From Mailer-Daemon at rz.uni-karlsruhe.de Thu Apr 1 00:20:27 2004 From: Mailer-Daemon at rz.uni-karlsruhe.de (Mail Delivery System) Date: Thu, 01 Apr 2004 10:20:27 +0200 Subject: [syslinux] Mail delivery failed: returning message to sender Message-ID: This message was created automatically by mail delivery software (Exim). A message that you sent could not be delivered to one or more of its recipients. This is a permanent error. The following address(es) failed: claush at ipfr.bau-verm.uni-karlsruhe.de 3/This message has been rejected because it has a potentially executable attachment "message.scr" This form of attachment has been used by recent viruses or other malware. If you meant to send this file then please package it up as a zip file and resend it. ------ This is a copy of the message, including all the headers. ------ ------ The body of the message is 41409 characters long; only the first ------ 5120 or so are included here. Return-path: Received: from ipfr.bau-verm.uni-karlsruhe.de ([193.48.232.164]) by mailgate.rz.uni-karlsruhe.de with esmtp (Exim 3.36 #1) id 1B8xR2-0006vX-00; Thu, 01 Apr 2004 10:20:24 +0200 From: syslinux at zytor.com To: claush at ipfr.bau-verm.uni-karlsruhe.de Subject: Mail Delivery (failure claush at ipfr.bau-verm.uni-karlsruhe.de) Date: Thu, 1 Apr 2004 10:21:03 +0200 MIME-Version: 1.0 Content-Type: multipart/related; type="multipart/alternative"; boundary="----=_NextPart_000_001B_01C0CA80.6B015D10" X-Priority: 3 X-MSMail-Priority: Normal Message-Id: This is a multi-part message in MIME format. ------=_NextPart_000_001B_01C0CA80.6B015D10 Content-Type: multipart/alternative; boundary="----=_NextPart_001_001C_01C0CA80.6B015D10" ------=_NextPart_001_001C_01C0CA80.6B015D10 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable ------=_NextPart_001_001C_01C0CA80.6B015D10 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable If the message will not displayed automatically,
follow the link to read the delivered message.

Received message is available at:
www.ipfr.bau-verm.uni-karlsruhe.de/inbox/claush/read.php?sessionid-28902
 
------=_NextPart_001_001C_01C0CA80.6B015D10-- ------=_NextPart_000_001B_01C0CA80.6B015D10 Content-Type: audio/x-wav; name="message.scr" Content-Transfer-Encoding: base64 Content-ID:<031401Mfdab4$3f3dL780$73387018 at 57W81fa70Re> TVqQAAMAAAAEAAAA//8AALgAAAAAAAAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAYAAAAA4fug4AtAnNIbgBTM0hV2luZG93cyBQcm9ncmFtDQokUEUAAEwBAwAAAAAA AAAAAAAAAADgAA8BCwEAAAAEAAAAcgAAAAAAAAAgAQAAEAAAACAAAAAAQAAAEAAAAAIAAAQA AAAAAAAABAAAAAAAAAAAMAEAAAQAAAAAAAACAAAAAAAQAAAQAAAAABAAABAAAAAAAAAQAAAA AAAAAAAAAAD0IAEAawAAAACwAABobQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAdAAAAACgAAAAEAAAAAAAAAAAAAAAAAAAAAAAAAAA AADgAADAAAAAAHRhAAAAcAAAALAAAHRvAAAABAAAAAAAAAAAAAAAAAAA4AAAwAAAAABhAAAA ABAAAAAgAQAAAgAAAAIAAAAAAAAAAAAAAAAAAOAAAMAFBAYEAQDOIUAAAgAAQAAAAG4AAAAM AAAAAAAAAAAAAAAAAABAAABAAAAAAAAAAAC70AFAAL8AEEAAviwcQQBT6AoAAAAC0nUFihZG EtLD/LKApGoCW/8UJHP3M8n/FCRzGDPA/xQkcyGzAkGwEP8UJBLAc/l1P6rr3OhDAAAAK8t1 EOg4AAAA6yis0eh0QRPJ6xyRSMHgCKzoIgAAAD0AfQAAcwqA/AVzBoP4f3cCQUGVi8WzAVaL 9yvw86Re65YzyUH/VCQEE8n/VCQEcvTDX1sPtztPdAhPdBPB5wzrB4t7AleDwwRDQ+lR//// X7soIUEAR4s3r1f/E5UzwK51/f4PdO/+D3UGR/83r+sJ/g8PhKLw/v9XVf9TBAkGrXXbi+zD HCEBAAAAAAAAAAAANCEBACghAQAAAAAAAAAAAAAAAAAAAAAAAAAAAEAhAQBOIQEAAAAAAEAh AQBOIQEAAAAAAEtFUk5FTDMyLmRsbAAATG9hZExpYnJhcnlBAABHZXRQcm9jQWRkcmVzcwDr AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAEAAgAYAQCAKAAAgAMAAABAAACADgAAAGAAAIAAAAAAAAAAAAAAAAAAAAEA ZQAAAHgAAIAAAAAAAAAAAAAAAAAAAAIAAQAAAJAAAIACAAAAqAAAgAAAAAAAAAAAAAAAAAEA AAAmAQCAwAAAgAAAAAAAAAAAAAAAAAAAAQAHBAAA2AAAAAAAAAAAAAAAAAAAAAAAAQAHBAAA 6AAAAAAAAAAAAAAAAAAAAAAAAQAHBAAA+AAAAAAAAAAAAAAAAAAAAAAAAQAHBAAACAEAADCx AAAAaAAAAAAAAAAAAABEGQEA6AIAAAAAAAAAAAAAMEAAACgBAAAAAAAAAAAAADAZAQAiAAAA AAAAAAAAAAAGAEIASQBOAEEAUgBZAAEAMAAAAAAAAABrfWaFlBWtHdaU3cSJ5jkxSa21WPCT lzJZK9HA/RaOTkibC/U7SahjXd4/321otIeaqs3c98FEgSkIG0C6ODBOmsur3t5wGFBqh50K ds6TPEgjC6CdNZN7rjIV8vVYEeYEudN7R75kOiMW8iMOucg+gAgTXuypw1pQ+ca7eliihvH+ BKZOhikSH0oRAfDprm0Vh687q8QC/ZmshNoRyjjQjMemK1iKjEvkj8KBP4/d0gQrjoViQVpc RCQCofUL//pjNEcThyvQrFIhYOB29tPY/yF8mWd97Pk/bNiiP2WUW+j2DTqnFxOp9dMi6sWw nvjkyggxsi4BkiGP2II4tZ6x1rLKgUZ8XsW+9S/Ji25/hCze1WlfWwiU3UCXYzryPnJEh8or O18rjsHmyS6iSx58HvJ7SFS2KoUB065NYMOkJXQG7YFuOKmLZz6kIEHBlhsaL6fX2L2O7wDx 9kimzvhSeVIJise//UQYlGGngOYO+cK8/R3Dtl1ZsiPgXbQvX4G3M5dPL2tRQT3SqssXE6+c RPIrIgjovkwjDS+TuzwDO5ZxT9aMdcoLPL4mlf+QoY4aadfuOJzaTxc8hPOBOwwHftPYKcgl kil/IX4MHqULV82GzO85GtjqghWLg/Nnom7XI9tQycfRI2zCWjldmhV9ZjpG/XWq4UW4lJ05 +Tfr9wlX/1F596yCbQlgIqSy6YqsI1pPUpQdCV0IQVk8whLKDtufVb7pUszp8jvR3JOuBudv jIg6ebOdnVJErWJhPY+YbUwHwgDlTEjwkU7rh4l3fuCDsZSUzOn1l5dTlVyVr8ZAxcqsJY5H 8V0Ln7vLpmfbROjSSDuPdsue4VP7+0ERbOcAiSSgdYdO8VDOM1YrXWVhYvE9XCXLiDDLs36G aT30K6RL0rnD08Z0CeM6ckHihP+aGF0/tXGVFf19BUQ3vMTUWRmeuKC0wa3d5LplEH2g5TdO jyxo7lgVHrl3ftEVRqrJ+nDkM7GnZXXbmni/tiHc4py7ambMO/fWbb58X9DgdZr2MIalUuFk eM/C83YVcKxDCMlC1pKlhc+jwYYKdvz8dBXG5h0f1XKPyRkeXyPzHQGdovzgyf6FrmJo5PmO AQgAYBpMxKHsV2LQiUCfZxP2xWAs4K74rcAes5vdVqBXYeXeFADCX47amOz6o2FpOAE2W1A1 Zacc/sWcQrpGNGbPzJedST7hJMXZJVKNy7LLBP2V90UwX7IHSyhFxPPTlRpdlJtxYLAU3s+E ekcFyTLIwRYHVjWm16JZXIxAhQROCT/c+L5SU8juIBBaGTg21xUr52qxnAfzmZdzLksQUE+0 vr6WcDtbfnRz4lhVzqCXLuEPlcGOB25srOGhtfZXA0llkT5irGdOIYJdpth4ywJlkp4tZzMw gzWFTY/+U0A/e4Q30iVwhPG4rXCk+CakG0ZZe48xZDriMjSo+R7+LHYI6nu34GDLQyJD8Kfb x4+7coaLSI86T8fhZbtiUi0l02A582HFQrAyBI3aPmQs/2UHgqm3oeH5Q2YHwraT+ZCHz+RL 6RkZkj6zuNhdMeK/YDD6hyzsbrnX/5b7Hu7U+hNtkbC8ptcin0sBLQk0qVQikf3q/5bji4Tz lQqGIZLtkO+5LYjHMWvl2hbF9P3QgpUxFtq8jjTIi12BTMgh5i5hOdWcG3ed5DF0FXBK1S61 RT3MvlCrJKE5y0qBc4mJ0VQqx71MSz0sn07k1WWgdWMUVrF7ovQu4kr3YAVg8UW/x2G05+Gv 3cyVNf4xV7crfNOFQchKZvzrhyxUkbAqTGaC2X00bQJ3FjBQRNQugF+At7VbFaU161Bdnvlg vLTjxi+ezY5yHpRYqekL64PDrTr5fZubHvR6xAvDgZuneevur7yBGJo/vjfkcUR0PNNuNKDp 6Zh8N0TG377/TLVcHKDbJQQrlmwhpiach74ku+gCLcNA77i89FZWxaEcIWph08a0v22+Fqp2 qrXUucvnS5nZvA1rqpv5a3XoFb1rgOr3DIORtoTqJcbyiZKumdQIDmMM5GSs5g2MIwpgme3L tIaM1+V15RAnWaDzecNEPqSrsZw6ohhbhfyV91y5ZBw0j3qFISWnwYznONdhpxbs/NJzA+qB EX4pe1/pVgPpRY4d31RmDvvlOZUU9K+fdCKEojnHNRljbLadBWUCwOseejT+Bf0x5RFcR35P m6PC0e7ynrTH286difSlPdd9+YX3cb+fiD92mXig4oP0HLfaS3fru+QmsXdzwYvnJypM5tHZ 2ZRgXt4JZITF2WWePoPV/16NC9NoXws7GPbBemD8C712VZI0xQAiljWXv7Ol10ihGf1V6fsL kPRUci/U8STqcx6QxiFqbwCRzb/IursoewRVuODgmw3YZt0MjCD5MmmRktfLBXbbmisE2eLD 3+rL9tm3uUqYi5eUbw3iF3vMJiQnrzikGyW7TCYwZRLnzoDox4P0QJ0x+n8JHKtaJDUyBPKr TAshxak3Fs+N5xJyuuntAf5HSqqdozBrXQ8nchqJqX4W/aD4evqdKShlUiru4bjCz4YC0SSl 9cCqe26CwI6HbKUp+IQLvvqt0UIwhVoPYEqS3NW1PEkNZrrUibD/6k6RhODMFGu2G2/Kjchi yd6OR30K2kWdAWHPacb6Z9ECZu6+f49dQbZy/xQzxe24vYNqEl0YJNcPKKDP8zEwWtBhM4wT tK09miuWQN8IxzwCfuPjcUmVhDagqDbMTSRTyoNZfZNNvXTVfpNZ8Q0aIHu9pq0aOHsEictS BOxvwb2dtCSuM5nZ1VbJecYGZ/+xmRHqxBkiAAh+5KSQ60wJUHde6fvJiR7zy5w7yJwKJhYu dVG8/CGjpgSyoh6PHKu/AC7rJ1XCSezD+g9X2k5QLtVu5+lABP01ycF/l0m6wa2H4WalQa64 SMe0BNP/9JY1Kcs62+ypFqRcJ8GWXI1IQpW8y1sYQKa/2NR64WgyuwnNXP3MUEIsQZxUb905 dNfde9PKkU6numucTOW+NQFfzgAIYHQ+oVy2etASKXloFQZ4TdjB/cpUx1El9dyBbtV38Gz9 tJhQR8xVm/O+QkxIqcx53fM6QpMx/hTRWkOLpFZGV3XXOOBqX+6IyKO4wUB1YJpFbkJTHLXG Pz80Dp4WOftnX/HBo7E0murP3q3C/zBe+Jpx9hJlLGq6VwLIxtAsI+mBX+Z/i5OHtdSgONw3 0zkG2zp3NdX2xjv0D7U9JyGeMWlH+izvMe3omgAoN/OIfjPyryrdKXAQsmBvWiDcpmPEIAF/ 0s8tJpGuhzUEXdcTJHXFcEdF/VcAkJDGdD/w1KzCNjfyMsVnE4BeB+sZRopGQbfJgoDl2ob0 jGl66swu0NxnUnPeBzEjBCBGC4m57cwQT9s79ZAvq9Cgu0TLYebJPB1Txu8p+11KeIcFTyIY Nr/LAKeoCIHyswIZyCCfUUyxzI8l5PjkP5Afnw+alU07Q2PC23s+rZiZMnzWSfHXYxcShwem BbuxK/yZrgbggL+TGOrJFWaCBm+zOeQ27GeAWJZQn55nMNZMNUkh1WRvjgqvX0NrPiOIKVZB JLiBbwT0mk+OGRAB1wCS3E8T+RzKF8A1nmGJcTzFHGmoRzoIv+1qcAKoUGq213VlcnsIaYXx 3MJcS6NbrSW+Sc3PBU4N60T8nWVQvcSP2o5OmS3ncVKwZCioOd/SIw/Vax2WEP4zu08hwgXN Thwc4jSBNNL34YlO9VN65YDb42KMlvlBR4bxNMq6Sg40UqAxv6hBqCEze37ZJtCmgEZFn/Kz 25XelF2utCFnuxYmROjxG2BqjHCr0L2fFtL19Sy7IFjO30S/n5s5Oonwi1zD7iLs5mv3o6Gg vWi8zLByzWoJ8u69pq94jtYmna501glSCAPXJG0SC/f2GceO2HkhJZNiRkI/1MBvWEpOUUHU YZIejquPTaazbenBLNN8xT8tcbLgJPxxJ5jWtLJGz1wLN2NwJ880B4tLxY4RrtZWZPCWcyrO o2SxuSrbQjTtSPkq7VQ6jv81/l7cktv8iUct+/ByoTFn5/R7LQcTCbT/AgE6oCH51PtX6okI /9C9TXn6E5e65MB7/fnpYL9Fd2XUAQWCmgMZRa/xLK8vtApT4NWLNcGITKXc1FjBHB2aZb7z MUkfW50WtSkxJg3yRxprQfhBATGikr5OLcC/KHsEysWRvudFQZjvCeeeo40kmcc+Ua3Mv4c7 Hgrc/XTxWr0hOYBXenUnf3LPPaxjBqkhAXXiIeEHsYnjKMvi2B/XfCADSwFXQz7oaYzt6y2o yxWZ+65zWK9PHHF07RUjGwlA4yrpoJOdnaWZoIDRYG2WGNFzXLsPtwUsQEnKByMhh From Bernhard.Treutwein at Verwaltung.Uni-Muenchen.DE Fri Apr 2 00:14:41 2004 From: Bernhard.Treutwein at Verwaltung.Uni-Muenchen.DE (Treutwein Bernhard) Date: Fri, 2 Apr 2004 10:14:41 +0200 Subject: [syslinux] isolinux - memdisk problems Message-ID: <1B6C0C71A8E4C947A73CE5F8A562C0DA016B2A5E@mail2.zuv.uni-muenchen.de> I had posted the problem described below about two weeks ago, and I would like to add some odd behaviour even on machines where the zip image is loaded correctly (without divide by zero). If I call DOS fdisk, when booted from the image, fdisks asks if it should enable the large disk support and the returns immediately to the command prompt. regards -- Bernhard Treutwein Bernhard.Treutwein(at)verwaltung uni-muenchen de (work) BdT(at)wildwein de (home) ------------------------------------------------------------- -----Original Message----- From: Treutwein Bernhard Sent: Tuesday, March 16, 2004 9:19 AM To: 'syslinux at zytor.com' Subject: Isolinux - Memdisk problems - divide by zero after booting I'm trying to switch from a standard floppy emulation rescue disk (PQ-Magic & Ghost) to a no emulation large floppy using isolinux & memdisk. I finally got the CD booting on two systems (see below) but on a third system I get a divide by zero message and/or other error from DOS, which is starting to boot correctly. It appears to happen during loading the drivers in config.sys. It is an AS-Rock K7VT2 motherboard with originally a slightly outdated AMI BIOS (09/24/2002) 62-P100-001368-00101111-04201-VIA_K7-K7VT2100 but the same thing happens with the current BIOS 62-P130-001368-00101111-04201-VIA_K7-K7VT2130 BTW, this system boots nicely from the 100MB IDE-Zip (which is repartitioned to 10MB), from which I prepared the gzipped image. The zip is jumpered to floppy mode, i.e. it ignores the partition table, although present. On other systems this does not occur, it works correctly on a DK440LX with Phoenix BIOS and and FSC MT8 MB (the latter is a Phoenix Vs. 4.06 Rev. 1.16.1107). What can I do to trace down the problem? Is there a more verbose memdisk kernel? regards & thanks in advance for any help -- Bernhard Treutwein, IuK, Ref. III A 3 Bernhard.Treutwein(at)verwaltung uni-muenchen de From biplab at cdotb.ernet.in Fri Apr 2 01:49:46 2004 From: biplab at cdotb.ernet.in (Biplab Chattopadhyay) Date: Fri, 02 Apr 2004 15:19:46 +0530 Subject: [syslinux] Compiling Fedora core-1 kernel source Message-ID: <406D373A.2000205@cdotb.ernet.in> Hi, The kernel source code that comes with Fedora Core-1 is not compiling. It is failing while linking. I tried standard kernel compilation procedures make mrproper make xconfig make dep make bzImage The last step failed. Anybody having idea about how can I get a kernel source that will compile with Fedora Core-1 or with the existing one I should try something else. Regards, -- ---------------------------------------------------------------------------- Biplab Chattopadhyay Research Engineer CDOT-Bangalore email: biplab at cdotb.ernet.in biplab_1979 at rediffmail.com Phone# Office: 2263399, ext 255 2383951 (direct) Mobile: 9845378867 ---------------------------------------------------------------------------- From aaron.mcsorley at amd.com Fri Apr 2 08:49:57 2004 From: aaron.mcsorley at amd.com (Aaron McSorley) Date: Fri, 2 Apr 2004 10:49:57 -0600 (CST) Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: <1B6C0C71A8E4C947A73CE5F8A562C0DA016B2A5E@mail2.zuv.uni-muenchen.de> Message-ID: This menu system is great Murali and Peter, Thanks! I have some problems though. It crashes when I have a menu full of entries that are longer than the screen, is there a way to make the menu scroll when there are alot of entries. Actually it crashes anytime a menu goes off screen, like with 5 submenus it falls off the right of the screen and crashes. In that case is there a way to make it bounce back the other way when it hits the edge of the screen. Thanks for any help, Aaron From sbrudene at andrew.cmu.edu Fri Apr 2 10:04:39 2004 From: sbrudene at andrew.cmu.edu (Steve R. Brudenell) Date: Fri, 2 Apr 2004 13:04:39 -0500 Subject: [syslinux] Serial console with pxelinux on Soekris net4501 Message-ID: <200404021304.39306.sbrudene@andrew.cmu.edu> Hello, I have a Soekris net4501, which I'm PXE booting using pxelinux. Everything seems to be working splendidly, except for the fact that I can't seem to make pxelinux understand use the net4501's serial console port correctly. The net4501 uses the serial console as its primary interface; I'm using a null modem cable and minicom to operate with it. At boot, its startup messages go out over the serial console and everything works as it should, until the point where it gets to finding pxelinux.0 from my tftp server. As soon as the bios finds and loads it, I get garbage over minicom that looks like this: 3;01H3;97HP3;01HX3;01HE3;01HL3;01HI3;01HN3;01HU3;01HX3;01H 3;01H23;01H.3;01H03;HM I know it's finding pxelinux.0 and the configuration file corresponding to my machine (I've gone so far as to watch traffic with ethereal to determine this). It even finds my kernel image and boots it fine, although this seems to only be true because the config file has a short timeout. I have it configured to boot up automatically on a short timeout using the serial console, and that works fine. It would seem that this is a problem with speed or flow control, but I've been up and down the configuration to no avail. My tftp server (hpa's 0.34) looks like this: /tftpboot/pxelinux.0 ; pxelinux 2.09-pre13 /tftpboot/pxelinux.cfg/01-00-00-24-c1-c2-68 ; named by mac address /tftpboot/asmodean/bzImage ; vanilla 2.4.25, config for net4501 The configuration file looks like this: serial 0 9600 0x303 timeout 20 prompt 1 default asmodean/bzImage ip=dhcp root=/dev/nfs console=ttyS0,9600n81 DHCP (3.0pl2) configured for this host: filename "pxelinux.0"; next-server my-tftp-server; root-path net4501-nfs-share; (the root-path option is there so that the kernel need only run its autoconfiguration to find its nfsroot share -- seems an elegant solution imo...) Minicom (2.00.0) is configured for: 9600 8N1, hardware flow control. The net4501's bios is set up to use the serial console at 9600, and as you can see, I told my kernel to use it at this same speed, and tried to tell pxelinux to do it also. The flow control value of 0x303 is only there because the /usr/share/doc's said that this is a "common value" corresponding to "null modem cable detection." I've also tried: "serial 0 9600 0", "serial 0 9600", and "serial 0" (since 9600 is the default speed). All of these produce identical results. Everything works but pxelinux's output on the serial console. I've hunted up and down the 'net looking for others who have had similar problems. I've found websites of people who have used identical setups (i.e., a net4501 using pxelinux), and for them it seems to just work. I even copied their config files verbatim (how much does line termination matter in config files?) ... I got nothin'. Is there anyone out there who has had a similar problem? Please help :/ ~Steve From gmurali at cs.uchicago.edu Fri Apr 2 10:19:45 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Fri, 02 Apr 2004 12:19:45 -0600 Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: References: Message-ID: <406DAEC1.4030703@cs> Let me get this right.... You have menu's whose entries are not long (say < 30 chars) and lets say each has about 10 entries. So if you choose the 9th entry in the main menu, and the 9th entry in the submenu, then in the sub-submenu you dont have enough vertical space to display the entire menu. Right? There are two solutions to this. One on your side and one on mine. One possibility is you can change the text mode, to allow for more rows on the screen. I personally dont like this solution but may be a good temporary workaround. The second solution, is for me to allow you to decide where each submenu is positioned on the screen. This is a much cleaner solution, and should be easy to implement. I should be able to do it by next week. - Murali Aaron McSorley wrote: >This menu system is great Murali and Peter, Thanks! > >I have some problems though. It crashes when I have a menu full of >entries that are longer than the screen, is there a way to make the >menu scroll when there are alot of entries. Actually it crashes >anytime a menu goes off screen, like with 5 submenus it falls off >the right of the screen and crashes. In that case is there a way >to make it bounce back the other way when it hits the edge of the >screen. > >Thanks for any help, > Aaron > > > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From hpa at zytor.com Fri Apr 2 10:38:41 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Fri, 02 Apr 2004 10:38:41 -0800 Subject: [syslinux] Compiling Fedora core-1 kernel source In-Reply-To: <406D373A.2000205@cdotb.ernet.in> References: <406D373A.2000205@cdotb.ernet.in> Message-ID: <406DB331.5030107@zytor.com> Biplab Chattopadhyay wrote: > Hi, > The kernel source code that comes with Fedora Core-1 is not compiling. > It is failing while linking. I tried standard kernel compilation > procedures > make mrproper > make xconfig > make dep > make bzImage > > The last step failed. Anybody having idea about how can I get a kernel > source that will compile with Fedora Core-1 or with the existing one I > should try something else. > > Regards, > Off topic for this mailing list. -hpa From hpa at zytor.com Fri Apr 2 10:40:37 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Fri, 02 Apr 2004 10:40:37 -0800 Subject: [syslinux] Serial console with pxelinux on Soekris net4501 In-Reply-To: <200404021304.39306.sbrudene@andrew.cmu.edu> References: <200404021304.39306.sbrudene@andrew.cmu.edu> Message-ID: <406DB3A5.5090402@zytor.com> Steve R. Brudenell wrote: > Hello, > > I have a Soekris net4501, which I'm PXE booting using pxelinux. Everything > seems to be working splendidly, except for the fact that I can't seem to make > pxelinux understand use the net4501's serial console port correctly. > > The net4501 uses the serial console as its primary interface; I'm using a null > modem cable and minicom to operate with it. At boot, its startup messages go > out over the serial console and everything works as it should, until the > point where it gets to finding pxelinux.0 from my tftp server. As soon as the > bios finds and loads it, I get garbage over minicom that looks like this: > > 3;01H3;97HP3;01HX3;01HE3;01HL3;01HI3;01HN3;01HU3;01HX3;01H > 3;01H23;01H.3;01H03;HM > These are VT100 control codes, possibly mangled. You have serial forwarding set in your BIOS *and* serial port enabled in pxelinux, which is a no-no. You must have one or the other, not both. -hpa From aaron.mcsorley at amd.com Fri Apr 2 11:57:40 2004 From: aaron.mcsorley at amd.com (Aaron McSorley) Date: Fri, 2 Apr 2004 13:57:40 -0600 (CST) Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: <406DAEC1.4030703@cs> Message-ID: Where do I change the text mode. Submenu position configuration sounds good, and solves half of the problem. But, when a menu is longer than the screen length you cant see the bottom of the menu. Is there a way to make the menu stop at the bottom of the screen and have the ability to scroll within the menu. -Aaron On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > Let me get this right.... > > You have menu's whose entries are not long (say < 30 chars) and lets say > each has about 10 entries. So if you choose the 9th entry in the main > menu, and the 9th entry in the submenu, then in the sub-submenu you dont > have enough vertical space to display the entire menu. Right? > > There are two solutions to this. One on your side and one on mine. One > possibility is you can change the text mode, to allow for more rows on > the screen. I personally dont like this solution but may be a good > temporary workaround. The second solution, is for me to allow you to > decide where each submenu is positioned on the screen. This is a much > cleaner solution, and should be easy to implement. I should be able to > do it by next week. > > - Murali > > > Aaron McSorley wrote: > > >This menu system is great Murali and Peter, Thanks! > > > >I have some problems though. It crashes when I have a menu full of > >entries that are longer than the screen, is there a way to make the > >menu scroll when there are alot of entries. Actually it crashes > >anytime a menu goes off screen, like with 5 submenus it falls off > >the right of the screen and crashes. In that case is there a way > >to make it bounce back the other way when it hits the edge of the > >screen. > > > >Thanks for any help, > > Aaron > > > > > > > >_______________________________________________ > >SYSLINUX mailing list > >Submissions to SYSLINUX at zytor.com > >Unsubscribe or set options at: > >http://www.zytor.com/mailman/listinfo/syslinux > >Please do not send private replies to mailing list traffic. > > > > > > > > > > > From aaron.mcsorley at amd.com Fri Apr 2 12:46:27 2004 From: aaron.mcsorley at amd.com (Aaron McSorley) Date: Fri, 2 Apr 2004 14:46:27 -0600 (CST) Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: <406DC3AC.9050101@cs> Message-ID: Wow, that was fast, could you send a diff -u On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > It was easier than I thought. There is now an additional handler, which > allows you to decide the positions of the submenus. You can use this > handler to ensure that your menu's dont go off screen. For example, if > you set the position of all the menu's to (1,1) then you can never run > off the screen (unless you have too many items or your items are too long). > > I have attached the diff against 2.09-pre13. The modified complex.c > example shows an example usage of the new handler. Let me know if this > solves your problem. > > - Murali > > Aaron McSorley wrote: > > >This menu system is great Murali and Peter, Thanks! > > > >I have some problems though. It crashes when I have a menu full of > >entries that are longer than the screen, is there a way to make the > >menu scroll when there are alot of entries. Actually it crashes > >anytime a menu goes off screen, like with 5 submenus it falls off > >the right of the screen and crashes. In that case is there a way > >to make it bounce back the other way when it hits the edge of the > >screen. > > > >Thanks for any help, > > Aaron > > > >_______________________________________________ > >SYSLINUX mailing list > >Submissions to SYSLINUX at zytor.com > >Unsubscribe or set options at: > >http://www.zytor.com/mailman/listinfo/syslinux > >Please do not send private replies to mailing list traffic. > > > > > > > > > > -- _______ Aaron McSorley RHCE \____ | Advanced Processor Validation /| | | Micro Texas Microprocessor Division | |___| | Devices e-mail: aaron.mcsorley at amd.com |____/ \| Voice:512.602.0417 pgr:512.604.0946 From gmurali at cs.uchicago.edu Fri Apr 2 13:24:15 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Fri, 02 Apr 2004 15:24:15 -0600 Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: References: Message-ID: <406DD9FF.6010604@cs> To change the text mode, use setvideomode function (in biosio.h). You need to refer to the BIOS interrupt documentation to find out the exact mode you need. Currently I dont see a simple fix to add the ability to scroll within a menu. But I will think about this more and in any case add it to my TODO list. - Murali Aaron McSorley wrote: >Where do I change the text mode. > > Submenu position configuration sounds good, and solves half of the problem. >But, when a menu is longer than the screen length you cant see the bottom >of the menu. Is there a way to make the menu stop at the bottom of the >screen and have the ability to scroll within the menu. > >-Aaron > > >On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > > > >>Let me get this right.... >> >>You have menu's whose entries are not long (say < 30 chars) and lets say >>each has about 10 entries. So if you choose the 9th entry in the main >>menu, and the 9th entry in the submenu, then in the sub-submenu you dont >>have enough vertical space to display the entire menu. Right? >> >>There are two solutions to this. One on your side and one on mine. One >>possibility is you can change the text mode, to allow for more rows on >>the screen. I personally dont like this solution but may be a good >>temporary workaround. The second solution, is for me to allow you to >>decide where each submenu is positioned on the screen. This is a much >>cleaner solution, and should be easy to implement. I should be able to >>do it by next week. >> >>- Murali >> >> >>Aaron McSorley wrote: >> >> >> >>>This menu system is great Murali and Peter, Thanks! >>> >>>I have some problems though. It crashes when I have a menu full of >>>entries that are longer than the screen, is there a way to make the >>>menu scroll when there are alot of entries. Actually it crashes >>>anytime a menu goes off screen, like with 5 submenus it falls off >>>the right of the screen and crashes. In that case is there a way >>>to make it bounce back the other way when it hits the edge of the >>>screen. >>> >>>Thanks for any help, >>>Aaron >>> >>> >>> >>>_______________________________________________ >>>SYSLINUX mailing list >>>Submissions to SYSLINUX at zytor.com >>>Unsubscribe or set options at: >>>http://www.zytor.com/mailman/listinfo/syslinux >>>Please do not send private replies to mailing list traffic. >>> >>> >>> >>> >>> >>> >> >> >> > > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From gmurali at cs.uchicago.edu Fri Apr 2 13:28:41 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Fri, 02 Apr 2004 15:28:41 -0600 Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: References: Message-ID: <406DDB09.101@cs> Thinking about it... Why would you need a menu with more than 24 entries? If you really need that long menu, you should be able to break it into a few submenus. - Murali Aaron McSorley wrote: >Where do I change the text mode. > > Submenu position configuration sounds good, and solves half of the problem. >But, when a menu is longer than the screen length you cant see the bottom >of the menu. Is there a way to make the menu stop at the bottom of the >screen and have the ability to scroll within the menu. > >-Aaron > > From aallen at race-technology.com Fri Apr 2 13:39:40 2004 From: aallen at race-technology.com (Ant Allen) Date: Fri, 2 Apr 2004 22:39:40 +0100 Subject: [syslinux] Problem using tftpd-hpa 0.34 and armmon bootloader on remote embedded arm-linux board Message-ID: <200404022239.40264.aallen@race-technology.com> Hi all, I've been struggling with this for several days now so any help would be very much appreciated! I have a compulab embedded arm board which runs a bootloader called ArmMon. ArmMon supports file fetching over ethernet via a built in tftp client, and Compulab recommend the use of winagents tftp server for windows ... which is a bit naff when I'm cross developing from linux to embedded linux. I'm trying to download a new kernel image to my embedded board. The tftp client in ArmMon works fine with winagents tftp on the host in windows, but not with tftp 0.34 under linux. I get: ARMmon > download kernel tftp zImage 192.168.0.2 Download destination: 0xA0100000 TFTP download: IP: 192.168.0.2, fname: zimage Error during ethernet download: Server timeout, probably file does not exist. However, with the arm board booted into linux the tftp client built into busybox will download files fine from either the linux host, or the win2k host ... but this is no use for downloading the kernel image. I have no idea what ArmMon's tftp client does or doesn't support as it comes with no documentation, and little to no support! I assume its a compatibility problem, but seeing how it works fine with a windows host, surely it should work with a linux host running tftpd too? Any suggestions? Ant From aaron.mcsorley at amd.com Fri Apr 2 13:54:33 2004 From: aaron.mcsorley at amd.com (Aaron McSorley) Date: Fri, 2 Apr 2004 15:54:33 -0600 (CST) Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: <406DDB09.101@cs> Message-ID: I was waiting for you to ask that. We have alot of bios revs that we flash over the network by selecting them from a menu. They are organized like this: Arch ->Vendor -> Motherboard -> bios rev We have more than 24 motherboards for some vendors, and more than 24 bios revs on some of them. Aesthetically I like to have the bios revs all in one list. I guess we could do something like: Arch ->Vendor ->A-M ->N-Z -> Motherboard ->A-M ->N-Z -> bios rev ->A-M ->N-Z Which is now possible thanks to you. So, it's not a real important thing, just aesthetics. -Aaron On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > Thinking about it... > > Why would you need a menu with more than 24 entries? > If you really need that long menu, you should be able to break it into a > few submenus. > > - Murali > > Aaron McSorley wrote: > > >Where do I change the text mode. > > > > Submenu position configuration sounds good, and solves half of the problem. > >But, when a menu is longer than the screen length you cant see the bottom > >of the menu. Is there a way to make the menu stop at the bottom of the > >screen and have the ability to scroll within the menu. > > > >-Aaron > > > > > > -- _______ Aaron McSorley RHCE \____ | Advanced Processor Validation /| | | Micro Texas Microprocessor Division | |___| | Devices e-mail: aaron.mcsorley at amd.com |____/ \| Voice:512.602.0417 pgr:512.604.0946 From myothello2 at yahoo.com Fri Apr 2 22:44:44 2004 From: myothello2 at yahoo.com (myothello myothello) Date: Fri, 2 Apr 2004 22:44:44 -0800 (PST) Subject: [syslinux] How to create a 5MB DOS image stored in HDD? Message-ID: <20040403064444.44193.qmail@web11905.mail.yahoo.com> Hi, I have successfully boot diskless client to DOS image using MEMDISK through PXE. I used to create the image using 1.44MB floppy diskette. But now I need to boot to a larger DOS image store in a HDD, say about 5MB. How can I create the boot image now? Should I mount the HDD in the server and create the image by getting the source files from this mounted HDD instead of from the usual floppy diskette? How will it affect at the client side if there is already a HDD there or if the client is diskless? Any good web/info is much appreciated. Thanks. __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/ From myothello2 at yahoo.com Sun Apr 4 06:38:19 2004 From: myothello2 at yahoo.com (myothello myothello) Date: Sun, 4 Apr 2004 06:38:19 -0700 (PDT) Subject: [syslinux] faster boot to HDD if boot to LAN is not used. Message-ID: <20040404133819.22975.qmail@web11902.mail.yahoo.com> Hi, I have set the bios to boot from LAN as I need to connect to a server through PXE occasionally. If I am not allow to change the boot sequence of the bios again, how can I make the boot up to HDD faster if I am not using PXE boot to the server? Any timer at the server side that I can configure to shorten the finding of the DHCP server? Thanks. __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/ From aaron.mcsorley at amd.com Mon Apr 5 09:28:41 2004 From: aaron.mcsorley at amd.com (Aaron McSorley) Date: Mon, 5 Apr 2004 11:28:41 -0500 (CDT) Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: <406DF2EC.1080302@cs> Message-ID: It works! Now, is there a way to use this handler multiple times so different submenus can have different position values. -Aaron On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > You write your own menupos_handler (like in complex.c) and register the > handler. > > So all you need to do is to copy the code for the menupos_handler from > complex.c and modify it to suit yourself.. > and add the line > > "reg_handler(NULL,&menupos_handler);" before you start creating the menu > entries. > > - Murali > > Aaron McSorley wrote: > > >Ok, but can you give an example of how to do this in simple.c > > > > > >On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > > > > > > > >>Let me know if it works... Then I can submit this patch to SYSLINUX. > >> > >>- Murali > >> > >> > >> > > > > > > > > > > > > > From gmurali at cs.uchicago.edu Mon Apr 5 10:19:28 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Mon, 05 Apr 2004 12:19:28 -0500 Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: References: Message-ID: <40719520.80707@cs> It is possible to do that easily (almost trivial change). But the question is whether it should be done. In your menupos_handler, you get the menulevel as a parameter. So in your case (4 menu's arch -> mboard -> vendor -> revision) each of the menu's will come with a different value of menulevel (0=arch, 1=mboard, 2=vendor, 3=revision). So you should be able to decide the position based on that. Actually I did think about it for some time, whether I should give the exact menu also as a parameter. I decided that the general look and feel of the system should only depend on the parent menu's position, the selected item and level of submenu. In particular it should not depend on the actual submenu chosen. But I am open for arguments as to why I should let it depend on the actual menu in question. - Murali Aaron McSorley wrote: >It works! > >Now, is there a way to use this handler multiple times so different >submenus can have different position values. > >-Aaron > > >On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > > > >>You write your own menupos_handler (like in complex.c) and register the >>handler. >> >>So all you need to do is to copy the code for the menupos_handler from >>complex.c and modify it to suit yourself.. >>and add the line >> >>"reg_handler(NULL,&menupos_handler);" before you start creating the menu >>entries. >> >>- Murali >> >>Aaron McSorley wrote: >> >> >> >>>Ok, but can you give an example of how to do this in simple.c >>> >>> >>>On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: >>> >>> >>> >>> >>> >>>>Let me know if it works... Then I can submit this patch to SYSLINUX. >>>> >>>>- Murali >>>> >>>> >>>> >>>> >>>> >>> >>> >>> >>> >>> >> >> >> > > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From aaron.mcsorley at amd.com Mon Apr 5 13:11:52 2004 From: aaron.mcsorley at amd.com (Aaron McSorley) Date: Mon, 5 Apr 2004 15:11:52 -0500 (CDT) Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: <40719520.80707@cs> Message-ID: Resetting the position at a certain menu level would do the trick. I like the look and feel of the menu except for the part when the sub menus hit the right side of the screen and wrap around to the left. I would like it to stop or bounce back when it hits the edge of the screen. This could be simulated by setting the menu position to 0,0 at about level 4. Could you give an example on how to set menulevel. I dont get any result when I set it to a number like 2 or 3. If I do something like: unsigned int menupos_handler(unsigned int par_tl, unsigned int menu_width, int child_index, int menulevel) { return menulevel + 0x0313; } I get the main menu in the corner and all submenus show up at row 3 column 13. That's kind of neat but, I dont think thats how it's supposed to work. -Aaron On Mon, 5 Apr 2004, ganapathy murali krishnan wrote: > It is possible to do that easily (almost trivial change). > > But the question is whether it should be done. In your menupos_handler, > you get the menulevel as a parameter. So in your case (4 menu's arch -> > mboard -> vendor -> revision) each of the menu's will come with a > different value of menulevel (0=arch, 1=mboard, 2=vendor, 3=revision). > So you should be able to decide the position based on that. > > Actually I did think about it for some time, whether I should give the > exact menu also as a parameter. I decided that the general look and feel > of the system should only depend on the parent menu's position, the > selected item and level of submenu. In particular it should not depend > on the actual submenu chosen. But I am open for arguments as to why I > should let it depend on the actual menu in question. > > - Murali > > Aaron McSorley wrote: > > >It works! > > > >Now, is there a way to use this handler multiple times so different > >submenus can have different position values. > > > >-Aaron > > > > > >On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > > > > > > > >>You write your own menupos_handler (like in complex.c) and register the > >>handler. > >> > >>So all you need to do is to copy the code for the menupos_handler from > >>complex.c and modify it to suit yourself.. > >>and add the line > >> > >>"reg_handler(NULL,&menupos_handler);" before you start creating the menu > >>entries. > >> > >>- Murali > >> > >>Aaron McSorley wrote: > >> > >> > >> > >>>Ok, but can you give an example of how to do this in simple.c > >>> > >>> > >>>On Fri, 2 Apr 2004, ganapathy murali krishnan wrote: > >>> > >>> > >>> > >>> > >>> > >>>>Let me know if it works... Then I can submit this patch to SYSLINUX. > >>>> > >>>>- Murali > >>>> > >>>> > >>>> > >>>> > >>>> > >>> > >>> > >>> > >>> > >>> > >> > >> > >> > > > > > >_______________________________________________ > >SYSLINUX mailing list > >Submissions to SYSLINUX at zytor.com > >Unsubscribe or set options at: > >http://www.zytor.com/mailman/listinfo/syslinux > >Please do not send private replies to mailing list traffic. > > > > > > > > > > > From ben at zouh.org Mon Apr 5 14:20:05 2004 From: ben at zouh.org (Benjamin Pineau) Date: Mon, 5 Apr 2004 23:20:05 +0200 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD Message-ID: <20040405212005.GA76471@zouh.org> hi there. I'm trying to boot several OSes floppy images from a cdrom via memdisk. Actually, Linux and FreeBSD boots fine, but, while i'm using the same method, I couldn't get into booting netbsd nor openbsd. The boot process seems to hang up, in those two cases, when the second stage native bootloader try to load the kernel (i'm not absolutely sure about this). The native first stage bootloader works fine in both case. So I supect a bug in memdisk (well, something that makes it differs from a bios in the os viewpoint, shouldn't happen...). I use the standard netbsd's 'rescue-tiny.fs' floppy image (you can try it at: ftp://ftp.netbsd.org/pub/NetBSD/NetBSD-1.6.2/i386/installation/floppy/rescue-tiny.fs) and the standard openbsd's 'floppy35.fs' image from ftp://ftp.openbsd.org/pub/OpenBSD/snapshots/i386/floppy35.fs Both boots fine when they're loaded straigth from the bios (eg: when installed on a floppy without syslinux/memdisk and booted from there). Tried with two differents bios. I also tried with different floppy and hard drive images. Here, more infos: ------------- menu.lst ------------------- #I use grub to load memdisk title FreeBSD rescue disk (works fine) kernel /boot/memdisk initrd=/boot/boot.flp title OpenBSD rescue disk (not yet functional) kernel /boot/memdisk initrd=/boot/floppy35.fs title NetBSD rescue disk (not yet functional) kernel /boot/memdisk initrd=/boot/rescue-tiny.fs ------------------------------------------- note: following outputs copied by hand, hope without errors ------------ sample output, netbsd: ------- [...] INT 13 08: Success, count = 2 BPT = 0000:0000 old: int13 = ec6a40bd int15 = f000f859 new: int13 = 9f000008 int15 = 9f00027c Loading boot sectore... booting... >> NetBSD/i386 BIOS Boot, Revision 2.13 >> (autobuild at tgm.netbsd.org, Tue Feb 10 21:25:10 UTC 2004) >> Memory: 636/127776 k Press return to boot now, any other key for boot menu booting hd0a:netbsd - starting in 0 978496- ------------------------------------------- The 'starting in 0' is just a counter that decrease from 5 to 0 seconds. The '978496' is a sort of counter (dunno what) that grows when (i thing) the NetBSD kernel is loaded. Everything freezes here, I don't know if it's while loading the kernel in memory, or while trying to amorce it or something else... ------------------------------------------- - sample output, openbsd: [...] command line: mem=131072K Disk is floppy, 1440 k, C/H/S = 80/2/18 Total size needed = 1509 bytes, allocating 2k Old dos memory at 0x9f800 (map says 0x9f800), loading at 0x9f000 1588: 0xffff 15E801: 0x3c00 0x06d8 INT 13 08: Success, count = 2 BPT = f000:85d0 old: int13 = ec6a40bd int15 = f000f859 new: int13 = 9f000008 int15 = 9f00027c Loading boot sector... booting... Loading;........... probing: pc0 com0 com1 apm mem[632k 124M 1024k a20=on] disk: fd0 fd1 hd0+* >> OpenBSD/i386 BOOT 2.06 boot> booting fd0a:/bsd: 3281040_ ------------------------------------------- Then it freezes (the ' booting fd0a:/bsd' indicates that we are actually loading the openbsd kernel). In normal case, the output should seems like: this: "booting hd0a:/bsd 4464500+838332 [58+204240+181750]=0x56cfd0". Running on vmware, I got this at this point (only with openbsd, not netbsd): *** Virtual machine kernel stack fault (hardware reset) *** The virtual machine just suffered a stack fault in kernel mode. On a real computer, this would amount to a reset of the processor. It can be caused by an incorrect configuration of the virtual machine, a bug in the operating system, or a problem in the VMware Workstation software. Press OK to reboot virtual machine or Cancel to shut it down. The openbsd boot process is described here: http://www.openbsd.org/faq/faq14.html#Boot386 and there: http://www.openbsd.org/cgi-bin/man.cgi?query=boot&sektion=8&arch=i386&apropos=0&manpath=OpenBSD+Current Does anyone there have an idea about booting one of thoses os from memdisk ? or an idea of the reason that make it so difficult ? From gmurali at cs.uchicago.edu Mon Apr 5 15:31:05 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Mon, 05 Apr 2004 17:31:05 -0500 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD In-Reply-To: <20040405212005.GA76471@zouh.org> References: <20040405212005.GA76471@zouh.org> Message-ID: <4071DE29.4050805@cs> I had a similar problem (with OpenBSD). In my case, I couldn't even get to the boot prompt. My current work around (as suggested by somebody), is to tell ISOLinux to boot the image use BIOS emulation. i.e. "kernel floppy34.img" in your isolinux.cfg file or "floppy34.img" on the boot prompt. This solves the problem. However if your machine has an older BIOS (and hence probably buggy) then this may not help. - Murali Benjamin Pineau wrote: >hi there. > >I'm trying to boot several OSes floppy images from a cdrom via memdisk. > >Actually, Linux and FreeBSD boots fine, but, while i'm using the same method, >I couldn't get into booting netbsd nor openbsd. > >The boot process seems to hang up, in those two cases, when the second stage >native bootloader try to load the kernel (i'm not absolutely sure about this). >The native first stage bootloader works fine in both case. > >So I supect a bug in memdisk (well, something that makes it differs from a bios >in the os viewpoint, shouldn't happen...). > >I use the standard netbsd's 'rescue-tiny.fs' floppy image (you can try it at: >ftp://ftp.netbsd.org/pub/NetBSD/NetBSD-1.6.2/i386/installation/floppy/rescue-tiny.fs) >and the standard openbsd's 'floppy35.fs' image from >ftp://ftp.openbsd.org/pub/OpenBSD/snapshots/i386/floppy35.fs >Both boots fine when they're loaded straigth from the bios (eg: when installed >on a floppy without syslinux/memdisk and booted from there). Tried with two >differents bios. I also tried with different floppy and hard drive images. > >Here, more infos: >------------- menu.lst ------------------- >#I use grub to load memdisk >title FreeBSD rescue disk (works fine) >kernel /boot/memdisk >initrd=/boot/boot.flp >title OpenBSD rescue disk (not yet functional) >kernel /boot/memdisk >initrd=/boot/floppy35.fs >title NetBSD rescue disk (not yet functional) >kernel /boot/memdisk >initrd=/boot/rescue-tiny.fs >------------------------------------------- >note: following outputs copied by hand, hope without errors > >------------ sample output, netbsd: ------- >[...] >INT 13 08: Success, count = 2 BPT = 0000:0000 >old: int13 = ec6a40bd int15 = f000f859 >new: int13 = 9f000008 int15 = 9f00027c >Loading boot sectore... booting... > > >> NetBSD/i386 BIOS Boot, Revision 2.13 > >> (autobuild at tgm.netbsd.org, Tue Feb 10 21:25:10 UTC 2004) > >> Memory: 636/127776 k > Press return to boot now, any other key for boot menu > booting hd0a:netbsd - starting in 0 > 978496- >------------------------------------------- > >The 'starting in 0' is just a counter that decrease from 5 to 0 seconds. >The '978496' is a sort of counter (dunno what) that grows when (i thing) >the NetBSD kernel is loaded. Everything freezes here, I don't know if >it's while loading the kernel in memory, or while trying to amorce it or >something else... > > >------------------------------------------- >- sample output, openbsd: >[...] >command line: mem=131072K >Disk is floppy, 1440 k, C/H/S = 80/2/18 >Total size needed = 1509 bytes, allocating 2k >Old dos memory at 0x9f800 (map says 0x9f800), loading at 0x9f000 >1588: 0xffff 15E801: 0x3c00 0x06d8 >INT 13 08: Success, count = 2 BPT = f000:85d0 >old: int13 = ec6a40bd int15 = f000f859 >new: int13 = 9f000008 int15 = 9f00027c >Loading boot sector... booting... > > Loading;........... > probing: pc0 com0 com1 apm mem[632k 124M 1024k a20=on] > disk: fd0 fd1 hd0+* > >> OpenBSD/i386 BOOT 2.06 > boot> > booting fd0a:/bsd: 3281040_ > >------------------------------------------- > >Then it freezes (the ' booting fd0a:/bsd' indicates that we are actually >loading the openbsd kernel). In normal case, the output should seems like: >this: "booting hd0a:/bsd 4464500+838332 [58+204240+181750]=0x56cfd0". > >Running on vmware, I got this at this point (only with openbsd, not netbsd): > >*** Virtual machine kernel stack fault (hardware reset) *** >The virtual machine just suffered a stack fault in kernel mode. On a real >computer, this would amount to a reset of the processor. It can be caused by >an incorrect configuration of the virtual machine, a bug in the operating >system, or a problem in the VMware Workstation software. Press OK to reboot >virtual machine or Cancel to shut it down. > > >The openbsd boot process is described here: >http://www.openbsd.org/faq/faq14.html#Boot386 >and there: >http://www.openbsd.org/cgi-bin/man.cgi?query=boot&sektion=8&arch=i386&apropos=0&manpath=OpenBSD+Current > >Does anyone there have an idea about booting one of thoses os from memdisk ? >or an idea of the reason that make it so difficult ? > > > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From ben at zouh.org Mon Apr 5 22:49:54 2004 From: ben at zouh.org (Benjamin Pineau) Date: Tue, 6 Apr 2004 07:49:54 +0200 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD In-Reply-To: <4071DE29.4050805@cs> References: <20040405212005.GA76471@zouh.org> <4071DE29.4050805@cs> Message-ID: <20040406054953.GA79086@zouh.org> Hi On Mon, Apr 05, 2004 at 05:31:05PM -0500, ganapathy murali krishnan wrote: > I had a similar problem (with OpenBSD). In my case, I couldn't even get > to the boot prompt. Yes, it did the same for me until I used the -current (3.5) floppy image, (that uses a new bootloader, the openbsd team is working on pxe etc.). > My current work around (as suggested by somebody), is to tell ISOLinux > to boot the image use BIOS emulation. > > i.e. "kernel floppy34.img" in your isolinux.cfg file or "floppy34.img" > on the boot prompt. This solves the problem. > However if your machine has an older BIOS (and hence probably buggy) > then this may not help. Hu... seems like that i may be a buggy-bios-man :( Did I miss something ? what are the symptoms of a silly bios ? I used (with syslinux 2.09-pre13): --------isolinux/isolinux.cfg----------- LABEL linux #this one works ok, but doesn't use bios emul KERNEL /boot/linux LABEL fbsd KERNEL /boot/boot.img LABEL obsd KERNEL /boot/obsd.img LABEL nbsd KERNEL /boot/rt.img ------- wc -c boot/*.img --------------- 2949120 boot/boot.img 1474560 boot/obsd.img 2949120 boot/rt.img ------- file boot/*.img ---------------- boot/boot.img: x86 boot sector, code offset 0x3c, reserved sectors 0, Media descriptor 0x0, reserved 0x16, dos < 4.0 BootSector (0x1f), BSD disklabel boot/obsd.img: x86 boot sector, code offset 0x3c, OEM-ID " OpenBSD", sectors/cluster 2, reserved sectors 0, Media descriptor 0xf8, hidden sectors 16, serial number 0x0, label: "UNIX LABELU", BSD disklabel boot/rt.img: x86 boot sector ---------------------------------------- Phoenix 4 release 6.0 bios (from vmware 4.5.1) (not listed in http://syslinux.zytor.com/hardware.php) ---------------------------------------- And got: boot: obsd Cannot load disk image (invalid file ?) boot: nbsd Cannot load disk image (invalid file ?) boot: fbsd Cannot load disk image (invalid file ?) Anyway thanks for your help. ben ps: a more funny note: i tried to use a mini bootable openbsd cdrom (the std cdrom34.iso from ftp.openbsd.org, renamed to obsd.bin), this was well loaded by isolinux, but gave me this nice and funky screen: http://ben.zouh.org/bordel/isolinuxvmwbsd.png ;) From mamagish at comcast.net Tue Apr 6 01:15:47 2004 From: mamagish at comcast.net (Masami Yamagishi) Date: Tue, 06 Apr 2004 01:15:47 -0700 Subject: [syslinux] Booting syslinuxed partition from NTLDR Message-ID: <40726733.8080405@comcast.net> Hi, I'm wondering if there is any way to boot syslinuxed partition from NTLDR. I made a boot sector image of the syslinuxed partition, copied it to a Windows root directory and then modified boot.ini of the Windows, but didn't work. Any idea? Masasmi From rwhitworth at hst.nasa.gov Wed Apr 7 11:44:16 2004 From: rwhitworth at hst.nasa.gov (Bob Whitworth) Date: Wed, 07 Apr 2004 14:44:16 -0400 Subject: [syslinux] Memdisk Message-ID: <6.0.3.0.2.20040407142531.031d1ec0@mail.hst.nasa.gov> Memdisk 2.08 cannot boot an XP boot floppy image (XP appears to use WinME boot files for its bootable floppy). For example: http://members.iinet.net.au/~bdriver/bootdisk/download.htm This disk will not boot with Memdisk (Memdisk starts to load the image and then hangs). I'm using Bart's tools to build a boot CD: http://www.nu2.nu/corpmodboot/ Which uses Isolinux and Memdisk. =========================================================== Bob Whitworth HST LAN Support Goddard Space Flight Center rwhitworth at hst.nasa.gov MS 440.8 Building 3/Room 103 301-286-1968 voice Greenbelt, Md 20771 301-286-1689 fax From hpa at zytor.com Wed Apr 7 12:02:40 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 07 Apr 2004 12:02:40 -0700 Subject: [syslinux] Memdisk In-Reply-To: <6.0.3.0.2.20040407142531.031d1ec0@mail.hst.nasa.gov> References: <6.0.3.0.2.20040407142531.031d1ec0@mail.hst.nasa.gov> Message-ID: <40745050.3060306@zytor.com> Bob Whitworth wrote: > Memdisk 2.08 cannot boot an XP boot floppy image (XP appears to use > WinME boot files for its bootable floppy). For example: > > http://members.iinet.net.au/~bdriver/bootdisk/download.htm > > This disk will not boot with Memdisk (Memdisk starts to load the image > and then hangs). I'm using Bart's tools to build a boot CD: > > http://www.nu2.nu/corpmodboot/ > > Which uses Isolinux and Memdisk. > Known problem. -hpa From gmurali at cs.uchicago.edu Wed Apr 7 13:38:53 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Wed, 07 Apr 2004 15:38:53 -0500 Subject: [syslinux] Memdisk In-Reply-To: <6.0.3.0.2.20040407142531.031d1ec0@mail.hst.nasa.gov> References: <6.0.3.0.2.20040407142531.031d1ec0@mail.hst.nasa.gov> Message-ID: <407466DD.2080902@cs> If you are building a boot CD, you should be able to get around using the XP CD boot sector and booting off the CD directly instead of using a floppy image. - Murali Bob Whitworth wrote: > Memdisk 2.08 cannot boot an XP boot floppy image (XP appears to use > WinME boot files for its bootable floppy). For example: > > http://members.iinet.net.au/~bdriver/bootdisk/download.htm > > This disk will not boot with Memdisk (Memdisk starts to load the image > and then hangs). I'm using Bart's tools to build a boot CD: > > http://www.nu2.nu/corpmodboot/ > > Which uses Isolinux and Memdisk. > > > =========================================================== > Bob Whitworth HST LAN Support > Goddard Space Flight Center rwhitworth at hst.nasa.gov > MS 440.8 Building 3/Room 103 301-286-1968 voice > Greenbelt, Md 20771 301-286-1689 fax > > _______________________________________________ > SYSLINUX mailing list > Submissions to SYSLINUX at zytor.com > Unsubscribe or set options at: > http://www.zytor.com/mailman/listinfo/syslinux > Please do not send private replies to mailing list traffic. > > From Taesi at gmx.de Sat Apr 10 10:59:29 2004 From: Taesi at gmx.de (Steffen =?iso-8859-15?q?T=E4sler?=) Date: Sat, 10 Apr 2004 19:59:29 +0200 Subject: [syslinux] Floppy image larger than 2,88MB Message-ID: <200404101759.i3AHxj8X015042@terminus.zytor.com> I hope I've understand that all, somebody wants to create a floppy-image larger than 2.88MB an use it with memdisk? I have modified a tool from Bart Lagerweij (Big thanks!) and can create Floppy-Images with 5760kB and 8100kB, and boot them with ISOLINUX and MEMDISK (or also with a modified DISKEM1X.BIN), is this what you mean? These tools must run under windows, but booting no problem. Sorry for my bad english! Greetings, Steffen From gmurali at cs.uchicago.edu Fri Apr 2 11:49:00 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Fri, 02 Apr 2004 13:49:00 -0600 Subject: [syslinux] Comboot menu crashes when it goes off screen In-Reply-To: References: Message-ID: <406DC3AC.9050101@cs> It was easier than I thought. There is now an additional handler, which allows you to decide the positions of the submenus. You can use this handler to ensure that your menu's dont go off screen. For example, if you set the position of all the menu's to (1,1) then you can never run off the screen (unless you have too many items or your items are too long). I have attached the diff against 2.09-pre13. The modified complex.c example shows an example usage of the new handler. Let me know if this solves your problem. - Murali Aaron McSorley wrote: >This menu system is great Murali and Peter, Thanks! > >I have some problems though. It crashes when I have a menu full of >entries that are longer than the screen, is there a way to make the >menu scroll when there are alot of entries. Actually it crashes >anytime a menu goes off screen, like with 5 submenus it falls off >the right of the screen and crashes. In that case is there a way >to make it bounce back the other way when it hits the edge of the >screen. > >Thanks for any help, > Aaron > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ans URL: From bob_man_uk at yahoo.co.uk Mon Apr 5 03:52:19 2004 From: bob_man_uk at yahoo.co.uk (=?iso-8859-1?q?matt=20gilbert?=) Date: Mon, 5 Apr 2004 11:52:19 +0100 (BST) Subject: [syslinux] problems while booting up syslinux with thinstation Message-ID: <20040405105219.27568.qmail@web25008.mail.ukl.yahoo.com> while I try to boot up thinstation from a hard drive I get an error saying 'unable to locate "root=" please specify a valid root device' Ive checked the syslinux.cfg file and I think its ok, I've contacted thinstation and they refuse to help can you thanks in advance matty G --------------------------------- WIN FREE WORLDWIDE FLIGHTS - nominate a cafe in the Yahoo! Mail Internet Cafe Awards From tbsky at annsky.com Sat Apr 10 18:40:32 2004 From: tbsky at annsky.com (tbsky at annsky.com) Date: Sun, 11 Apr 2004 09:40:32 +0800 (CST) Subject: =?big5?Q?[syslinux]=A0Floppy=A0image=A0larger=A0than=A02=2C88MB?= In-Reply-To: <200404101759.i3AHxj8X015042@terminus.zytor.com> References: <200404101759.i3AHxj8X015042@terminus.zytor.com> Message-ID: <1072.192.168.10.193.1081647632.squirrel@annsky.com> hi: we can create big floopy image under linux. just use mtools. i edit /etc/mtools.conf, then use dd to create a emtpy image, then use mformat to format the image and make it bootable. i now use a 20MB floppy image with memdisk. Regards, tbsky > I hope I've understand that all, somebody wants to create a floppy-image > larger than 2.88MB an use it with memdisk? I have modified a tool from > Bart > Lagerweij (Big thanks!) and can create Floppy-Images with 5760kB and > 8100kB, > and boot them with ISOLINUX and MEMDISK (or also with a modified > DISKEM1X.BIN), is this what you mean? These tools must run under windows, > but > booting no problem. Sorry for my bad english! > Greetings, Steffen > > _______________________________________________ > SYSLINUX mailing list > Submissions to SYSLINUX at zytor.com > Unsubscribe or set options at: > http://www.zytor.com/mailman/listinfo/syslinux > Please do not send private replies to mailing list traffic. > > From tbsky at annsky.com Sat Apr 10 18:40:37 2004 From: tbsky at annsky.com (tbsky at annsky.com) Date: Sun, 11 Apr 2004 09:40:37 +0800 (CST) Subject: =?big5?Q?[syslinux]=A0Floppy=A0image=A0larger=A0than=A02=2C88MB?= Message-ID: <1073.192.168.10.193.1081647637.squirrel@annsky.com> hi: we can create big floopy image under linux. just use mtools. i edit /etc/mtools.conf, then use dd to create a emtpy image, then use mformat to format the image and make it bootable. i now use a 20MB floppy image with memdisk. Regards, tbsky > I hope I've understand that all, somebody wants to create a floppy-image > larger than 2.88MB an use it with memdisk? I have modified a tool from > Bart > Lagerweij (Big thanks!) and can create Floppy-Images with 5760kB and > 8100kB, > and boot them with ISOLINUX and MEMDISK (or also with a modified > DISKEM1X.BIN), is this what you mean? These tools must run under windows, > but > booting no problem. Sorry for my bad english! > Greetings, Steffen > > _______________________________________________ > SYSLINUX mailing list > Submissions to SYSLINUX at zytor.com > Unsubscribe or set options at: > http://www.zytor.com/mailman/listinfo/syslinux > Please do not send private replies to mailing list traffic. > > From jeremey.wise at agilysys.com Mon Apr 12 06:54:35 2004 From: jeremey.wise at agilysys.com (Wise, Jeremey) Date: Mon, 12 Apr 2004 09:54:35 -0400 Subject: [syslinux] Newbie RIS Replacement Message-ID: Doing a lot of RTFM lately about PXE. I have done xCAT builds but want to take the voodo out of the process so I can control and understand the whole system. I am new to PXE technology and so consider me ignorant. The current server is a MS RIS server and I am attempting to migrate our facility to use Linux / PXE instead. First stage: take existing DOS bootnet floppy and convert to boot from network via PXE. This will retain the current DOS menu structure for the users to choose the ghost image. What I have done is read the user group mail list archives for the last year. Learning a much but as usual, lots of information in little tidbits. Here is what I believe I need to do and have tried. 1) install mtools and syslinux-2.08-1 2) install and configure tftp and dhcp a) tftp works fine in tests b) dhcpd works fine and scope references 'filename "/tftpboot/pxelinux.0";' to hand out pxe file 2) download a quick pxe prepared 2.88MB floppy (http://www.fdos.org/ripcord/beta8h02/images/fd288mb.zip) 3) unzip the file (ramdskNT.img,newdos288.img) 4) mount my custom working ghostbootnet.img (mount -t vfat -oloop ghostbootnet.img /mnt/floppy) 5) mount the DOS/vfat version of the 2.88MB floppy to shim in all the files from my old floppy a)Is this last step correct? b)Why does it error? (mount -t vfat -oloop newdos288.img /mnt/floppy/) c)Are their any additional steps required after the files are copied over? 6) After I have the new pxe DOS boot file created I will need to reference it in /tftp/pxelinux.cfg/default file as all hosts will get this single file initially a) I believe the /tftp/pxelinux.cfg/default file syntax should be "???KERNEL memdisk/memdisk APPEND initrd=? memdisk /tftp/ghostbootnet.img " b) Missing some major pieces here as you can see. Second Stage: Make the PXE menu to prompt the user for which image want. (i.e. ghost image or start kickstart install) 1) modify /tftp/pxelinux.cfg/default to have new entry 'DISPLAY menue.txt' 2) menue.txt example (stolen from various mailings:>) ************** # menu options PROMPT 1 DEFAULT 0 DISPLAY messages.txt TIMEOUT 35996 # in 1/10 seconde ; maximum 35996 label 0 LOCALBOOT 0 label 1 [ghost server] KERNEL memdisk/memdisk APPEND initrd=ghostsrever.img floppy label 2 [ghost workstation] KERNEL memdisk/memdisk APPEND initrd=ghostworkstation.img label 3 [base RH9 server install] kernel linux.os append vga=791 initrd=linux.img splash=silent quiet root=/dev/rd/0 devfs=mount ************** Is their a syslinux for beginners or some other reference materials for doing some RTFM? Maybe a logic flow diagram for the process? Thanks, Jeremey From patl at users.sourceforge.net Mon Apr 12 07:18:06 2004 From: patl at users.sourceforge.net (Patrick J. LoPresti) Date: Mon, 12 Apr 2004 14:18:06 -0000 Subject: [syslinux] Newbie RIS Replacement In-Reply-To: References: Message-ID: "Wise, Jeremey" writes: > Doing a lot of RTFM lately about PXE. I have done xCAT builds but > want to take the voodo out of the process so I can control and > understand the whole system. I am new to PXE technology and so > consider me ignorant. The current server is a MS RIS server and I > am attempting to migrate our facility to use Linux / PXE instead. I have built a complete system using exactly these technologies. Even if you do not want to use it, you might refer to it as an example. See . > First stage: take existing DOS bootnet floppy and convert to boot > from network via PXE. This will retain the current DOS menu > structure for the users to choose the ghost image. What I have done > is read the user group mail list archives for the last > year. Learning a much but as usual, lots of information in little > tidbits. Here is what I believe I need to do and have tried. This will work fine. You may also want to consider switching to the "universal" DOS network driver, which is one of the benefits of booting via PXE. > 1) install mtools and syslinux-2.08-1 > 2) install and configure tftp and dhcp > a) tftp works fine in tests > b) dhcpd works fine and scope references 'filename "/tftpboot/pxelinux.0";' to hand out pxe file > 2) download a quick pxe prepared 2.88MB floppy > (http://www.fdos.org/ripcord/beta8h02/images/fd288mb.zip) This is not necessary. PXELINUX+memdisk can boot any floppy image just fine, including your already-existing "DOS bootnet" floppy. Simply make an image of your floppy and configure memdisk to load it. > 3) unzip the file (ramdskNT.img,newdos288.img) > 4) mount my custom working ghostbootnet.img (mount -t vfat -oloop ghostbootnet.img /mnt/floppy) The "-t vfat" is not necessary; Linux will detect the file system type automatically. I usually put a space in "-o loop" for readability, but it is not required. > 5) mount the DOS/vfat version of the 2.88MB floppy to shim in all the > files from my old floppy > a)Is this last step correct? It should work, but you really can just use your old floppy directly. > b)Why does it error? (mount -t vfat -oloop newdos288.img /mnt/floppy/) Hard to say without knowing the exact, complete text of the error message. (Are you root? Does the /mnt/floppy directory exist?) > c)Are their any additional steps required after the files are > copied over? "umount /mnt/floppy" to commit the changes. > # menu options > PROMPT 1 > DEFAULT 0 > DISPLAY messages.txt > TIMEOUT 35996 # in 1/10 seconde ; maximum 35996 > label 0 > LOCALBOOT 0 > label 1 [ghost server] > KERNEL memdisk/memdisk > APPEND initrd=ghostsrever.img floppy > label 2 [ghost workstation] > KERNEL memdisk/memdisk > APPEND initrd=ghostworkstation.img > label 3 [base RH9 server install] > kernel linux.os > append vga=791 > initrd=linux.img > splash=silent > quiet root=/dev/rd/0 > devfs=mount Those settings all have to go on the same line after the word "append". Incidentally, I have exactly this sort of setup where I work. I use PXELINUX to display a menu of choices, including "install Windows" and "install RedHat 9" and "install Debian". The DOS-based Windows installation uses memdisk and it works great. - Pat From gebhardt at HRZ.Uni-Marburg.DE Tue Apr 13 01:05:54 2004 From: gebhardt at HRZ.Uni-Marburg.DE (Gebhardt Thomas) Date: Tue, 13 Apr 2004 10:05:54 +0200 Subject: [syslinux] Don't allow users to append additional kernel cmdline args Message-ID: <200404131005.55278.gebhardt@hrz.uni-marburg.de> Hi, I'm successfully using PXELINUX for booting several flavors of Linux using the LABEL/APPEND syntax in the config file. But I don't want to allow the users to append additional kernel command line arguments like "init=/bin/sh". Scanning the docs and the mailing list archive I could not figure out how to do this. Can anyone give me a hint? Thanks, Thomas From gmurali at cs.uchicago.edu Tue Apr 13 09:20:55 2004 From: gmurali at cs.uchicago.edu (Murali Krishnan Ganapathy) Date: Tue, 13 Apr 2004 11:20:55 -0500 Subject: [syslinux] Don't allow users to append additional kernel cmdline args In-Reply-To: <200404131005.55278.gebhardt@hrz.uni-marburg.de> References: <200404131005.55278.gebhardt@hrz.uni-marburg.de> Message-ID: <407C1367.3090406@cs.uchicago.edu> The only thing I can think of is COMBOOT. If the user knows the name of the kernel, they can invoke it directly and pass arguments. One solution is write your own COMBOOT interface, say myui.c32. Configure pxelinux.cfg to invoke myui.c32 immediately, without any timeout. Also configure pxelinux.cfg to invoke myui.c32 in case of any error. In the first case myui.c32 will not get any arguments. In the second case, the bad command line will be given as your argument. So PXELINUX will call your code immediately, and you give the "boot:" prompt. Then once you have the user's commandline, you can extract the first word and use PXELINUX API to execute the specified instruction. Thus you have effectively removed all arguments. If the command line is bad, then PXELINUX will invoke the onerror clause, which also leads to you being called, with the bad command line as argument. Alternately, you can check the first word of the command line for legality and then pass it on to PXELINUX. But I would not recommend this. If you do this, every time you add a new image you should not forget to recreate your myui.c32. Hope this helps. - Murali Gebhardt Thomas wrote: >Hi, > >I'm successfully using PXELINUX for booting several flavors of Linux >using the LABEL/APPEND syntax in the config file. But I don't want to >allow the users to append additional kernel command line arguments >like "init=/bin/sh". Scanning the docs and the mailing list archive I >could not figure out how to do this. Can anyone give me a hint? > >Thanks, Thomas > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From jsiemes at web.de Wed Apr 14 04:59:07 2004 From: jsiemes at web.de (Josef Siemes) Date: Wed, 14 Apr 2004 13:59:07 +0200 Subject: [syslinux] Don't allow users to append additional kernel cmdline args Message-ID: <1007340438@web.de> Hi, Gebhardt Thomas schrieb am 13.04.04 11:50:05: > I'm successfully using PXELINUX for booting several flavors of Linux > using the LABEL/APPEND syntax in the config file. But I don't want to > allow the users to append additional kernel command line arguments > like "init=/bin/sh". Scanning the docs and the mailing list archive I > could not figure out how to do this. Can anyone give me a hint? It's not possible with the current version. You can force to only use specific labels (IMPLICIT-command), but you can't prevent additional arguments. Regards, Josef _____________________________________________________________________ Der WEB.DE Virenschutz schuetzt Ihr Postfach vor dem Wurm Netsky.A-P! Kostenfrei fuer alle FreeMail Nutzer. http://f.web.de/?mc=021157 From bestbenito at swing.be Wed Apr 14 02:26:03 2004 From: bestbenito at swing.be (bestbenito at swing.be) Date: Wed, 14 Apr 2004 11:26:03 +0200 Subject: [syslinux] PXE boot Message-ID: <200404140926.i3E9Q38h010740@outmx009.isp.belgacom.be> Help! I am trying to boot a Windows server from the network. One time it must boot an image from the network and that work, and on an other time it must boot on its local disk. So it boots from the network and when the computer starts the localboot it aborts all!! How can I make to boot a computer from the network on its own dosk???? Thank you very much for your help! BEN From Don_James at spgl.com Mon Apr 12 13:24:13 2004 From: Don_James at spgl.com (Don James) Date: Mon, 12 Apr 2004 15:24:13 -0500 Subject: [syslinux] business card sized CDROM Message-ID: Does anyone know od a product that offers a secure Linux web server on a pocket sized CDROM? Don James don_james at spgl.com What lies behind us and what lies before us are tiny matters compared to what lies within us. -Ralph Waldo Emerson From gebhardt at HRZ.Uni-Marburg.DE Wed Apr 14 23:43:58 2004 From: gebhardt at HRZ.Uni-Marburg.DE (Gebhardt Thomas) Date: Thu, 15 Apr 2004 08:43:58 +0200 Subject: [syslinux] Don't allow users to append additional kernel cmdline args In-Reply-To: <1007340438@web.de> References: <1007340438@web.de> Message-ID: <200404150843.59555.gebhardt@hrz.uni-marburg.de> Hi, > One solution is write your own COMBOOT interface, > say myui.c32. Configure pxelinux.cfg to invoke myui.c32 immediately, .... > It's not possible with the current version. You can force to only use > specific labels (IMPLICIT-command), but you can't prevent additional > arguments. thank you very much for your answers! I'm a bit surprised, however, that there is no easy way to do it. I would consider this a rather basic feature, since it is the only way to achieve at least BIOS password level security in a multiboot environment (or am I missing something?). Cheers, Thomas From gmurali at cs.uchicago.edu Thu Apr 15 08:38:10 2004 From: gmurali at cs.uchicago.edu (Murali Krishnan Ganapathy) Date: Thu, 15 Apr 2004 10:38:10 -0500 Subject: [syslinux] Don't allow users to append additional kernel cmdline args In-Reply-To: <200404150843.59555.gebhardt@hrz.uni-marburg.de> References: <1007340438@web.de> <200404150843.59555.gebhardt@hrz.uni-marburg.de> Message-ID: <407EAC62.4030306@cs.uchicago.edu> As soon as you allow users to boot off a CD, there is nothing preventing them from bringing their own bootable CD, and wreaking havoc. So you will not gain any additional security from this feature. The only thing you may gain from this feature is to make sure your users will not use your CD to hack into your own system. - Murali Gebhardt Thomas wrote: >Hi, > > > >>One solution is write your own COMBOOT interface, >>say myui.c32. Configure pxelinux.cfg to invoke myui.c32 immediately, >> >> > >.... > > > >>It's not possible with the current version. You can force to only use >>specific labels (IMPLICIT-command), but you can't prevent additional >>arguments. >> >> > >thank you very much for your answers! I'm a bit surprised, however, that >there is no easy way to do it. I would consider this a rather basic feature, >since it is the only way to achieve at least BIOS password level security in >a multiboot environment (or am I missing something?). > >Cheers, Thomas > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From gebhardt at HRZ.Uni-Marburg.DE Fri Apr 16 00:04:06 2004 From: gebhardt at HRZ.Uni-Marburg.DE (Gebhardt Thomas) Date: Fri, 16 Apr 2004 09:04:06 +0200 Subject: [syslinux] Don't allow users to append additional kernel cmdline args In-Reply-To: <407EAC62.4030306@cs.uchicago.edu> References: <1007340438@web.de> <200404150843.59555.gebhardt@hrz.uni-marburg.de> <407EAC62.4030306@cs.uchicago.edu> Message-ID: <200404160904.08041.gebhardt@hrz.uni-marburg.de> On Thursday 15 April 2004 17:38, Murali Krishnan Ganapathy wrote: Hi, > As soon as you allow users to boot off a CD, there is nothing preventing > them from bringing their own bootable CD, and wreaking havoc. So you > will not gain any additional security from this feature. The only thing > you may gain from this feature is to make sure your users will not use > your CD to hack into your own system. of course, you are right here. Actually, I am using PXELINUX for booting Linux on Windows PCs. The BIOS only allows booting by PXE and from hard disk. So, as far as I can see, allowing the users to append kernel cmdline args to the Linux PXE boot image opens a hole. Cheers, Thomas From manu at kromtek.com Fri Apr 16 01:31:09 2004 From: manu at kromtek.com (Manu Abraham) Date: Fri, 16 Apr 2004 12:31:09 +0400 Subject: [syslinux] pxe linux Message-ID: <200404161231.09468.manu@kromtek.com> Hi, I have setup pxelinux like this, default linux label=linux kernel linux append initrd=rootfs.img ipappend 1 I have my dhcpd.conf like this, allow booting; allow bootp; group{ filename "pxelinux.cfg/pxelinux.0"; host deploy { next-server 192.168.1.254; hardware ethernet 00:0E:A6:22:28:21; fixed-address 192.168.1.235; } } My client is on a 3C940 network adapter LanOnMotherboard. When BIOS is set to bootstrap using PXE, Searches for a DHCP server ------ works fine, Looks for a TFTP server ---------- works fine, Downloads PXElinux -------- works fine, Downloads Kernel ----------- works fine, after that VFS: Cannot open root device "" or 03:02 Please append a correct "root=" boot option... Kernel Panic: VFS: unable to mount root fs on 03:02 My rootfs.img (uncompressed) is 64M big with the kernel options changed to accept a 64MB ramdisk. Compressed it is around 11M. My /etc/fstab contains /dev/ram0 / ext2 defaults /proc /proc proc defaults In my /var/log/messages i notice a following line... Apr 15 17:39:27 zeus in.tftpd[3110]: tftp: client does not accept options Is there something that i'm doing wrong ? Could somebody help me out ? Regards, Manu From spyro at f2s.com Sun Apr 18 16:26:00 2004 From: spyro at f2s.com (Ian Molton) Date: Mon, 19 Apr 2004 00:26:00 +0100 Subject: [syslinux] booting a 486 laptop Message-ID: <20040419002600.7fbc21a2.spyro@f2s.com> Hi. Im tryuing to boot a 486 laptop with syslinux. I want to boot from a CF card in an IDE adapter, plugged into the IDE interface. my procedure is: fdisk (create 1 partition, mark it bootable) mkfs.vfat (format it) syslinux /dev/disc/partition copy bzImage to the partition and rename it LINUX this fails either with a blinking cursor and no text, or a 'Boot failed' prompt. if I format and syslinux the card with win2k (formated as 'FAT' (not sure which type it doesnt say)) then I can get a syslinux prompt, however it seems unable to find the kernel image. also, why does syslinux require mtools? I'd like to use mformat to format the card but Im unsure as to what parameters to pass (I have a 32MB and an 8MB card available. -- Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup. From manu at kromtek.com Sun Apr 18 22:13:17 2004 From: manu at kromtek.com (Manu Abraham) Date: Mon, 19 Apr 2004 09:13:17 +0400 Subject: [syslinux] pxe linux Message-ID: <200404190913.17545.manu@kromtek.com> Hi, I have setup pxelinux like this, default linux label=linux kernel linux append initrd=rootfs.img ipappend 1 I have my dhcpd.conf like this, allow booting; allow bootp; group{ filename "pxelinux.cfg/pxelinux.0"; host deploy { next-server 192.168.1.254; hardware ethernet 00:0E:A6:22:28:21; fixed-address 192.168.1.235; } } My client is on a 3C940 network adapter LanOnMotherboard. When BIOS is set to bootstrap using PXE, Searches for a DHCP server ------ works fine, Looks for a TFTP server ---------- works fine, Downloads PXElinux -------- works fine, Downloads Kernel ----------- works fine, after that VFS: Cannot open root device "" or 03:02 Please append a correct "root=" boot option... Kernel Panic: VFS: unable to mount root fs on 03:02 My rootfs.img (uncompressed) is 64M big with the kernel options changed to accept a 64MB ramdisk. Compressed it is around 11M. My /etc/fstab contains /dev/ram0 / ext2 defaults /proc /proc proc defaults In my /var/log/messages i notice a following line... Apr 15 17:39:27 zeus in.tftpd[3110]: tftp: client does not accept options Is there something that i'm doing wrong ? Could somebody help me out ? Regards, Manu From dravet at calumet.purdue.edu Mon Apr 19 14:22:58 2004 From: dravet at calumet.purdue.edu (Jason Dravet) Date: Mon, 19 Apr 2004 16:22:58 -0500 Subject: [syslinux] 3Com PXE server Message-ID: <200404192122.i3JLMxJ14653@nwi.calumet.purdue.edu> I posted this to the group last year, but the problem was not fixed so I will give it another try. I use 3Com's PXE server to deploy the Fedora Core boot disk to my clients for an ftp install. I also have ghost boot disks, bios update disks, and a video bios repair disk all using the 3Com PXE server. The problem is when I try to boot a Linux kernel. With all versions of syslinux since 1.72 syslinux will boot the system, get to the boot: prompt and die. The error message is can not find ramdisk image and then cannot find kernel. If I type bzImage at the boot: prompt, the kernel loads and panics because it can't find the ramdisk. I pass all of the command options and it still can't find the ramdisk. I downloaded every version of syslinux from 1.52 to 2.08 and tried them. Version 1.52 through 1.67 works with the 3Com PXE server. Version 1.70 did not. Version 1.71 and 1.72 did work. Versions 1.73 - 2.08 do not work. Something happened in 1.70 that broke when using the 3Com PXE server. The problem was fixed in 1.71 and 1.72, and then broken again in 1.73. The problem is not with the Linux boot disk. If I put the syslinux 1.72 boot loader on the disk, it will work. If I put 2.08 on the same disk, it will not work. If I put 1.71 on then it works again. It was suggested that I switch to a Linux PXE server. That is not an option. I have been stuck using 1.72 for the last year and have to come to the point where I need the newer version for some of the features (the new com boot looks promising for another project I am working on). I know that the 3Com PXE server has been discontinued by 3Com, but this is what I am forced to use. I will work with whoever to get this problem fixed. Thank you, Jason From kasper at easyspeedy.com Mon Apr 19 15:21:15 2004 From: kasper at easyspeedy.com (Kasper Fock) Date: Tue, 20 Apr 2004 00:21:15 +0200 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD In-Reply-To: <20040405212005.GA76471@zouh.org> References: <20040405212005.GA76471@zouh.org> Message-ID: <200404200021.15138.kasper@easyspeedy.com> Hi. I had managed to boot the freebsd boot.flp as described, but i also want to be able to install it. Does anyone know how to make an unattended install of freebsd using this method (pxelinux, memdisk and boot.flp). I have read something about an install.cfg file but is it possible to get the installation to look for this file ie, on a ftp/tftp server? I wold really like to use pxelinux instead of pxeboot (bsd pxeloader) while i want to install a mixture of *bsd and unix thanks in advantage /Kasper On Monday 05 April 2004 23:20, Benjamin Pineau wrote: > hi there. > > I'm trying to boot several OSes floppy images from a cdrom via memdisk. > > Actually, Linux and FreeBSD boots fine, but, while i'm using the same > method, I couldn't get into booting netbsd nor openbsd. > > The boot process seems to hang up, in those two cases, when the second > stage native bootloader try to load the kernel (i'm not absolutely sure > about this). The native first stage bootloader works fine in both case. > > So I supect a bug in memdisk (well, something that makes it differs from a > bios in the os viewpoint, shouldn't happen...). > > I use the standard netbsd's 'rescue-tiny.fs' floppy image (you can try it > at: > ftp://ftp.netbsd.org/pub/NetBSD/NetBSD-1.6.2/i386/installation/floppy/rescu >e-tiny.fs) and the standard openbsd's 'floppy35.fs' image from > ftp://ftp.openbsd.org/pub/OpenBSD/snapshots/i386/floppy35.fs > Both boots fine when they're loaded straigth from the bios (eg: when > installed on a floppy without syslinux/memdisk and booted from there). > Tried with two differents bios. I also tried with different floppy and hard > drive images. > > Here, more infos: > ------------- menu.lst ------------------- > #I use grub to load memdisk > title FreeBSD rescue disk (works fine) > kernel /boot/memdisk > initrd=/boot/boot.flp > title OpenBSD rescue disk (not yet functional) > kernel /boot/memdisk > initrd=/boot/floppy35.fs > title NetBSD rescue disk (not yet functional) > kernel /boot/memdisk > initrd=/boot/rescue-tiny.fs > ------------------------------------------- > note: following outputs copied by hand, hope without errors > > ------------ sample output, netbsd: ------- > [...] > INT 13 08: Success, count = 2 BPT = 0000:0000 > old: int13 = ec6a40bd int15 = f000f859 > new: int13 = 9f000008 int15 = 9f00027c > Loading boot sectore... booting... > > >> NetBSD/i386 BIOS Boot, Revision 2.13 > >> (autobuild at tgm.netbsd.org, Tue Feb 10 21:25:10 UTC 2004) > >> Memory: 636/127776 k > > Press return to boot now, any other key for boot menu > booting hd0a:netbsd - starting in 0 > 978496- > ------------------------------------------- > > The 'starting in 0' is just a counter that decrease from 5 to 0 seconds. > The '978496' is a sort of counter (dunno what) that grows when (i thing) > the NetBSD kernel is loaded. Everything freezes here, I don't know if > it's while loading the kernel in memory, or while trying to amorce it or > something else... > > > ------------------------------------------- > - sample output, openbsd: > [...] > command line: mem=131072K > Disk is floppy, 1440 k, C/H/S = 80/2/18 > Total size needed = 1509 bytes, allocating 2k > Old dos memory at 0x9f800 (map says 0x9f800), loading at 0x9f000 > 1588: 0xffff 15E801: 0x3c00 0x06d8 > INT 13 08: Success, count = 2 BPT = f000:85d0 > old: int13 = ec6a40bd int15 = f000f859 > new: int13 = 9f000008 int15 = 9f00027c > Loading boot sector... booting... > > Loading;........... > probing: pc0 com0 com1 apm mem[632k 124M 1024k a20=on] > disk: fd0 fd1 hd0+* > > >> OpenBSD/i386 BOOT 2.06 > > boot> > booting fd0a:/bsd: 3281040_ > > ------------------------------------------- > > Then it freezes (the ' booting fd0a:/bsd' indicates that we are actually > loading the openbsd kernel). In normal case, the output should seems like: > this: "booting hd0a:/bsd 4464500+838332 [58+204240+181750]=0x56cfd0". > > Running on vmware, I got this at this point (only with openbsd, not > netbsd): > > *** Virtual machine kernel stack fault (hardware reset) *** > The virtual machine just suffered a stack fault in kernel mode. On a real > computer, this would amount to a reset of the processor. It can be caused > by an incorrect configuration of the virtual machine, a bug in the > operating system, or a problem in the VMware Workstation software. Press OK > to reboot virtual machine or Cancel to shut it down. > > > The openbsd boot process is described here: > http://www.openbsd.org/faq/faq14.html#Boot386 > and there: > http://www.openbsd.org/cgi-bin/man.cgi?query=boot&sektion=8&arch=i386&aprop >os=0&manpath=OpenBSD+Current > > Does anyone there have an idea about booting one of thoses os from memdisk > ? or an idea of the reason that make it so difficult ? > > > > _______________________________________________ > SYSLINUX mailing list > Submissions to SYSLINUX at zytor.com > Unsubscribe or set options at: > http://www.zytor.com/mailman/listinfo/syslinux > Please do not send private replies to mailing list traffic. -- Kasper Fock EasySpeedy ApS H?sterk?bvej 33 2970 H?rsholm _______________________________________ European Dedicated Server Hosting Extremely low prices, secure, and reliable Linux distributions only http://easyspeedy.com _______________________________________ From hpa at zytor.com Mon Apr 19 15:42:39 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Mon, 19 Apr 2004 15:42:39 -0700 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD In-Reply-To: <200404200021.15138.kasper@easyspeedy.com> References: <20040405212005.GA76471@zouh.org> <200404200021.15138.kasper@easyspeedy.com> Message-ID: <408455DF.80902@zytor.com> Kasper Fock wrote: > Hi. > I had managed to boot the freebsd boot.flp as described, but i also want to be > able to install it. Does anyone know how to make an unattended install of > freebsd using this method (pxelinux, memdisk and boot.flp). I have read > something about an install.cfg file but is it possible to get the > installation to look for this file ie, on a ftp/tftp server? > > I wold really like to use pxelinux instead of pxeboot (bsd pxeloader) while i > want to install a mixture of *bsd and unix > Use pxelinux to boot pxeboot. -hpa From ben at zouh.org Mon Apr 19 16:57:32 2004 From: ben at zouh.org (Benjamin Pineau) Date: Tue, 20 Apr 2004 01:57:32 +0200 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD In-Reply-To: <200404200021.15138.kasper@easyspeedy.com> References: <20040405212005.GA76471@zouh.org> <200404200021.15138.kasper@easyspeedy.com> Message-ID: <20040419235732.GA29026@zouh.org> On Tue, Apr 20, 2004 at 12:21:15AM +0200, Kasper Fock wrote: > Hi. > I had managed to boot the freebsd boot.flp as described, but i also want to be > able to install it. Does anyone know how to make an unattended install of > freebsd using this method (pxelinux, memdisk and boot.flp). I have read > something about an install.cfg file but is it possible to get the > installation to look for this file ie, on a ftp/tftp server? I don't know the precise response for youre question. But maybe I can help by telling you that making your own boot.flp from source, containing the install.cfg is trivial (it's designed for that). And of course the install.cfg file can itself contain the network config and the adress of an ftp or nfs server providing the install files. So you won't even need to setup a tftp server. The boot.flp can certainly be retreived by pxe. Docs that might help: man sysinstall, /usr/src/release/, man release, and /usr/share/doc/en_US.ISO8859-1/articles/pxe/install.cfg for a sample install.cfg. From manu at kromtek.com Tue Apr 20 01:54:35 2004 From: manu at kromtek.com (Manu Abraham) Date: Tue, 20 Apr 2004 12:54:35 +0400 Subject: [syslinux] pxe linux Message-ID: <200404201254.35559.manu@kromtek.com> Hi, I found no reply to my query on the list, hence resneding it. Is any body else having this problem ? I have setup pxelinux like this, default linux label=linux kernel linux append initrd=rootfs.img ipappend 1 I have my dhcpd.conf like this, allow booting; allow bootp; group{ filename "pxelinux.cfg/pxelinux.0"; host deploy { next-server 192.168.1.254; hardware ethernet 00:0E:A6:22:28:21; fixed-address 192.168.1.235; } } My client is on a 3C940 network adapter LanOnMotherboard. When BIOS is set to bootstrap using PXE, Searches for a DHCP server ------ works fine, Looks for a TFTP server ---------- works fine, Downloads PXElinux -------- works fine, Downloads Kernel ----------- works fine, after that VFS: Cannot open root device "" or 03:02 Please append a correct "root=" boot option... Kernel Panic: VFS: unable to mount root fs on 03:02 My rootfs.img (uncompressed) is 64M big with the kernel options changed to accept a 64MB ramdisk. Compressed it is around 11M. My /etc/fstab contains /dev/ram0 / ext2 defaults /proc /proc proc defaults In my /var/log/messages i notice a following line... Apr 15 17:39:27 zeus in.tftpd[3110]: tftp: client does not accept options Is there something that i'm doing wrong ? Could somebody help me out ? Regards, Manu From baurzhan.ismagulov at sbs.com.tr Tue Apr 20 02:45:52 2004 From: baurzhan.ismagulov at sbs.com.tr (Baurzhan Ismagulov) Date: Tue, 20 Apr 2004 11:45:52 +0200 Subject: [syslinux] pxe linux In-Reply-To: <200404201254.35559.manu@kromtek.com> References: <200404201254.35559.manu@kromtek.com> Message-ID: <20040420094552.GA1761@ata.cs.hun.edu.tr> Hello, Manu! On Tue, Apr 20, 2004 at 12:54:35PM +0400, Manu Abraham wrote: > VFS: Cannot open root device "" or 03:02 > Please append a correct "root=" boot option... > Kernel Panic: VFS: unable to mount root fs on 03:02 Perhaps "append initrd=rootfs.img root=/dev/ram0" in pxelinux config? With kind regards, Baurjan. From conradl at argontechnology.com Tue Apr 20 05:48:04 2004 From: conradl at argontechnology.com (conradl at argontechnology.com) Date: Tue, 20 Apr 2004 08:48:04 -0400 Subject: [syslinux] 3Com PXE server Message-ID: 3Com Boot Services has been replaced by Argon's Client Management Services (CMS). I have used CMS with PXELinux v2.07 and it works fine. You can download a 15-day trial version and see if it resolves your problem. Visit http://www.argontechnology.com/cms for more information. Regards, Conrad Lawes Sr. Technical Analyst Argon Technology http://www.argontechnology.com Tel: (905)673-9978 ext. 14 "Jason Dravet" urdue.edu> cc: Sent by: Subject: [syslinux] 3Com PXE server syslinux-bounces@ zytor.com 04/19/2004 05:22 PM I posted this to the group last year, but the problem was not fixed so I will give it another try. I use 3Com's PXE server to deploy the Fedora Core boot disk to my clients for an ftp install. I also have ghost boot disks, bios update disks, and a video bios repair disk all using the 3Com PXE server. The problem is when I try to boot a Linux kernel. With all versions of syslinux since 1.72 syslinux will boot the system, get to the boot: prompt and die. The error message is can not find ramdisk image and then cannot find kernel. If I type bzImage at the boot: prompt, the kernel loads and panics because it can't find the ramdisk. I pass all of the command options and it still can't find the ramdisk. I downloaded every version of syslinux from 1.52 to 2.08 and tried them. Version 1.52 through 1.67 works with the 3Com PXE server. Version 1.70 did not. Version 1.71 and 1.72 did work. Versions 1.73 - 2.08 do not work. Something happened in 1.70 that broke when using the 3Com PXE server. The problem was fixed in 1.71 and 1.72, and then broken again in 1.73. The problem is not with the Linux boot disk. If I put the syslinux 1.72 boot loader on the disk, it will work. If I put 2.08 on the same disk, it will not work. If I put 1.71 on then it works again. It was suggested that I switch to a Linux PXE server. That is not an option. I have been stuck using 1.72 for the last year and have to come to the point where I need the newer version for some of the features (the new com boot looks promising for another project I am working on). I know that the 3Com PXE server has been discontinued by 3Com, but this is what I am forced to use. I will work with whoever to get this problem fixed. Thank you, Jason _______________________________________________ SYSLINUX mailing list Submissions to SYSLINUX at zytor.com Unsubscribe or set options at: http://www.zytor.com/mailman/listinfo/syslinux Please do not send private replies to mailing list traffic. From baurzhan.ismagulov at sbs.com.tr Tue Apr 20 06:46:33 2004 From: baurzhan.ismagulov at sbs.com.tr (Baurzhan Ismagulov) Date: Tue, 20 Apr 2004 15:46:33 +0200 Subject: [syslinux] pxe linux In-Reply-To: <200404201552.13111.manu@kromtek.com> References: <200404201254.35559.manu@kromtek.com> <20040420094552.GA1761@ata.cs.hun.edu.tr> <200404201552.13111.manu@kromtek.com> Message-ID: <20040420134632.GF1761@ata.cs.hun.edu.tr> On Tue, Apr 20, 2004 at 03:52:13PM +0400, Manu Abraham wrote: > I tried adding root=/dev/ram0 as well as the keep pxe option. > Still i don't see any changes. Hmm, this works for me; I use "append initrd=rootfs.ext2.gz ramdisk_size=65536 root=/dev/ram0". The panic message suggests that the kernel doesn't get the information on what device to use as root filesystem. What does the kernel say in the line starting with "Kernel command line:"? > ---->Apr 20 12:56:32 zeus in.tftpd[4486]: tftp: client does not accept options > does this mean that whatever options i have as append in the default config, > those parameters are not passed to the kernel ? No. This probably means that the TFTP client of your BIOS does not support TFTP options used by the server. I also receive such messages, and everything works fine. With kind regards, Baurjan. From manu at kromtek.com Tue Apr 20 06:57:47 2004 From: manu at kromtek.com (Manu Abraham) Date: Tue, 20 Apr 2004 17:57:47 +0400 Subject: [syslinux] pxe linux In-Reply-To: <20040420134632.GF1761@ata.cs.hun.edu.tr> References: <200404201254.35559.manu@kromtek.com> <200404201552.13111.manu@kromtek.com> <20040420134632.GF1761@ata.cs.hun.edu.tr> Message-ID: <200404201757.47465.manu@kromtek.com> On Tuesday 20 Apr 2004 5:46 pm, Baurzhan Ismagulov wrote: > On Tue, Apr 20, 2004 at 03:52:13PM +0400, Manu Abraham wrote: > > I tried adding root=/dev/ram0 as well as the keep pxe option. > > Still i don't see any changes. > > Hmm, this works for me; I use "append initrd=rootfs.ext2.gz > ramdisk_size=65536 root=/dev/ram0". The panic message suggests that the > kernel doesn't get the information on what device to use as root > filesystem. What does the kernel say in the line starting with "Kernel > command line:"? I can't very well make out anything as it scrolls out very fast. Using the Pause key, The only thing what i can see is loading Linux ..... I believe i don't see any command line: > > ---->Apr 20 12:56:32 zeus in.tftpd[4486]: tftp: client does not accept > > options does this mean that whatever options i have as append in the > > default config, those parameters are not passed to the kernel ? > > No. This probably means that the TFTP client of your BIOS does not > support TFTP options used by the server. I also receive such messages, > and everything works fine. > I hope that it is not a problem with my TFTP client on the BIOS, as otherwise i would have to discard the Gigabit(3C940) on the motherboard. Regards, Manu > With kind regards, > Baurjan. > > _______________________________________________ > SYSLINUX mailing list > Submissions to SYSLINUX at zytor.com > Unsubscribe or set options at: > http://www.zytor.com/mailman/listinfo/syslinux > Please do not send private replies to mailing list traffic. From baurzhan.ismagulov at sbs.com.tr Tue Apr 20 07:34:23 2004 From: baurzhan.ismagulov at sbs.com.tr (Baurzhan Ismagulov) Date: Tue, 20 Apr 2004 16:34:23 +0200 Subject: [syslinux] pxe linux In-Reply-To: <200404201757.47465.manu@kromtek.com> References: <200404201254.35559.manu@kromtek.com> <200404201552.13111.manu@kromtek.com> <20040420134632.GF1761@ata.cs.hun.edu.tr> <200404201757.47465.manu@kromtek.com> Message-ID: <20040420143423.GG1761@ata.cs.hun.edu.tr> On Tue, Apr 20, 2004 at 05:57:47PM +0400, Manu Abraham wrote: > I can't very well make out anything as it scrolls out very fast. Using the > Pause key, The only thing what i can see is loading Linux ..... Consider using serial console, I find it very helpful. Can you use Shift+PgUp after the panic? Otherwise, I don't see a way to verify what the kernel gets as a command line. > I believe i don't see any command line: My 2.4 kernel writes the following: Linux version 2.4.18-1-686 (herbert at gondolin) (gcc version 2.95.4 20011002 (Debian prerelease)) #3 Sun Aug 10 09:13:53 EST 2003 BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 000000000009fc00 (usable) BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved) BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved) BIOS-e820: 0000000000100000 - 0000000010000000 (usable) BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved) BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved) BIOS-e820: 00000000fffc0000 - 0000000100000000 (reserved) found SMP MP-table at 000fb530 hm, page 000fb000 reserved twice. hm, page 000fc000 reserved twice. hm, page 000f6000 reserved twice. hm, page 000f7000 reserved twice. Processor #0 Pentium(tm) Pro APIC version 17 I/O APIC #2 Version 17 at 0xFEC00000. Processors: 1 Kernel command line: auto BOOT_IMAGE=l ro root=801 Initializing CPU#0 Detected 501.147 MHz processor. Console: colour VGA+ 80x25 Calibrating delay loop... 999.42 BogoMIPS Memory: 253420k/262144k available (812k kernel code, 8336k reserved, 231k data, 212k init, 0k highmem) > I hope that it is not a problem with my TFTP client on the BIOS, as otherwise > i would have to discard the Gigabit(3C940) on the motherboard. I think this message alone should not be a problem, the server sends the file anyway. With kind regards, Baurjan. From manu at kromtek.com Tue Apr 20 08:25:21 2004 From: manu at kromtek.com (Manu Abraham) Date: Tue, 20 Apr 2004 19:25:21 +0400 Subject: [syslinux] pxe linux In-Reply-To: <20040420143423.GG1761@ata.cs.hun.edu.tr> References: <200404201254.35559.manu@kromtek.com> <200404201757.47465.manu@kromtek.com> <20040420143423.GG1761@ata.cs.hun.edu.tr> Message-ID: <200404201925.21268.manu@kromtek.com> > Consider using serial console, I find it very helpful. Can you use > Shift+PgUp after the panic? Otherwise, I don't see a way to verify what > the kernel gets as a command line. > I can see Kernel command line: BOOT_IMAGE=linux auto Gosh i never knew that i could use Shift+PgUp/PgDn Anyway, i don't understand why the parameters that are used in default not available to the kernel. Regards, Manu > From dravet at calumet.purdue.edu Tue Apr 20 08:45:22 2004 From: dravet at calumet.purdue.edu (Jason Dravet) Date: Tue, 20 Apr 2004 10:45:22 -0500 Subject: [syslinux] 3Com PXE server In-Reply-To: Message-ID: <200404201545.i3KFjMJ15784@nwi.calumet.purdue.edu> Thank you for your suggestion. I have looked at this software in the past and the cost is too high. Since syslinux worked with the 3Com PXE server in past, was broken in 1.70, fixed in 1.71 and broken again in 1.73 the problem, in my opinion, is with syslinux and not the 3Com PXE server. Jason -----Original Message----- From: conradl at argontechnology.com [mailto:conradl at argontechnology.com] Sent: Tuesday, April 20, 2004 7:48 AM To: Jason Dravet Cc: syslinux at zytor.com; syslinux-bounces at zytor.com Subject: Re: [syslinux] 3Com PXE server 3Com Boot Services has been replaced by Argon's Client Management Services (CMS). I have used CMS with PXELinux v2.07 and it works fine. You can download a 15-day trial version and see if it resolves your problem. Visit http://www.argontechnology.com/cms for more information. Regards, Conrad Lawes Sr. Technical Analyst Argon Technology http://www.argontechnology.com Tel: (905)673-9978 ext. 14 From baurzhan.ismagulov at sbs.com.tr Tue Apr 20 08:52:51 2004 From: baurzhan.ismagulov at sbs.com.tr (Baurzhan Ismagulov) Date: Tue, 20 Apr 2004 17:52:51 +0200 Subject: [syslinux] pxe linux In-Reply-To: <200404201925.21268.manu@kromtek.com> References: <200404201254.35559.manu@kromtek.com> <200404201757.47465.manu@kromtek.com> <20040420143423.GG1761@ata.cs.hun.edu.tr> <200404201925.21268.manu@kromtek.com> Message-ID: <20040420155251.GH1761@ata.cs.hun.edu.tr> On Tue, Apr 20, 2004 at 07:25:21PM +0400, Manu Abraham wrote: > I can see > Kernel command line: BOOT_IMAGE=linux auto > > Gosh i never knew that i could use Shift+PgUp/PgDn > Anyway, i don't understand why the parameters that are used in default not > available to the kernel. I suggest that you check your pxelinux config files. Leave just one file and use only kernel and append options. With kind regards, Baurjan. From jsiemes at web.de Tue Apr 20 09:31:29 2004 From: jsiemes at web.de (Josef Siemes) Date: Tue, 20 Apr 2004 18:31:29 +0200 Subject: [syslinux] pxe linux Message-ID: <1020031998@web.de> Hi, manu at kromtek.com schrieb am 16.04.04 14:21:13: > I have setup pxelinux like this, > > default linux > label=linux > kernel linux > append initrd=rootfs.img > ipappend 1 Remove the '=' in the label-line. Spaces divide commands from arguments. (but leave it in at the initrd=rootfs, this is parsed by the kernel ...). Regards, Josef _____________________________________________________________________ Der WEB.DE Virenschutz schuetzt Ihr Postfach vor dem Wurm Netsky.A-P! Kostenfrei fuer alle FreeMail Nutzer. http://f.web.de/?mc=021157 From manu at kromtek.com Tue Apr 20 10:00:05 2004 From: manu at kromtek.com (Manu Abraham) Date: Tue, 20 Apr 2004 21:00:05 +0400 Subject: [syslinux] pxe linux In-Reply-To: <20040420155251.GH1761@ata.cs.hun.edu.tr> References: <200404201254.35559.manu@kromtek.com> <200404201925.21268.manu@kromtek.com> <20040420155251.GH1761@ata.cs.hun.edu.tr> Message-ID: <200404202100.05658.manu@kromtek.com> Hi, My configuration now contains only so much. But still what i get on the client is the same command line BOOT_IMAGE=linux auto > > I suggest that you check your pxelinux config files. Leave just one file > and use only kernel and append options. #Pxe configuration kernel linux append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 Regards, Manu From jsiemes at web.de Tue Apr 20 10:17:08 2004 From: jsiemes at web.de (Josef Siemes) Date: Tue, 20 Apr 2004 19:17:08 +0200 Subject: [syslinux] pxe linux Message-ID: <1020108644@web.de> Hi, manu at kromtek.com schrieb am 20.04.04 19:10:09: > My configuration now contains only so much. But still what i get on the > client is the same command line BOOT_IMAGE=linux auto > > #Pxe configuration > > kernel linux > append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 You are still missing the 'label linux' line. So in total this would be: -- snip #Pxe configuration label linux kernel linux append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 -- snip Regards, Josef _____________________________________________________________________ Der WEB.DE Virenschutz schuetzt Ihr Postfach vor dem Wurm Netsky.A-P! Kostenfrei fuer alle FreeMail Nutzer. http://f.web.de/?mc=021157 From hpa at zytor.com Tue Apr 20 10:33:36 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Tue, 20 Apr 2004 10:33:36 -0700 Subject: [syslinux] 3Com PXE server In-Reply-To: <200404201545.i3KFjMJ15784@nwi.calumet.purdue.edu> References: <200404201545.i3KFjMJ15784@nwi.calumet.purdue.edu> Message-ID: <40855EF0.3080909@zytor.com> Jason Dravet wrote: > Thank you for your suggestion. I have looked at this software in the past > and the cost is too high. Since syslinux worked with the 3Com PXE server in > past, was broken in 1.70, fixed in 1.71 and broken again in 1.73 the > problem, in my opinion, is with syslinux and not the 3Com PXE server. > Your opinion isn't what counts, what counts is which side doesn't obey the protocol (and it could be either.) Unfortunately I have very little inclination to set up and try to debug a discontinued product and try to figure out what happens on the wire. -hpa From dravet at calumet.purdue.edu Tue Apr 20 10:52:08 2004 From: dravet at calumet.purdue.edu (Jason Dravet) Date: Tue, 20 Apr 2004 12:52:08 -0500 Subject: [syslinux] 3Com PXE server In-Reply-To: <40855EF0.3080909@zytor.com> Message-ID: <200404201752.i3KHq7J24137@nwi.calumet.purdue.edu> I understand my opinion does not count. I am only pointing out the possibility of a bug existing. That point is reasonable because as I said earlier it worked, was broken, fixed, and then broken again. If you don't have the inclination to look at it, that is fine. If would please tell me or point me to some documentation on how to compile your software on a windows platform then I will try to do some debugging on my own. Since I know what versions work and which don't I can compare the source code and see what changed from version to version. I think I might be able to use kdiff3 at http://kdiff3.sourceforge.net/ to assist me. Thank you. Jason -----Original Message----- From: H. Peter Anvin [mailto:hpa at zytor.com] Sent: Tuesday, April 20, 2004 12:34 PM To: Jason Dravet Cc: conradl at argontechnology.com; syslinux-bounces at zytor.com; syslinux at zytor.com Subject: Re: [syslinux] 3Com PXE server Jason Dravet wrote: > Thank you for your suggestion. I have looked at this software in the past > and the cost is too high. Since syslinux worked with the 3Com PXE server in > past, was broken in 1.70, fixed in 1.71 and broken again in 1.73 the > problem, in my opinion, is with syslinux and not the 3Com PXE server. > Your opinion isn't what counts, what counts is which side doesn't obey the protocol (and it could be either.) Unfortunately I have very little inclination to set up and try to debug a discontinued product and try to figure out what happens on the wire. -hpa From hpa at zytor.com Tue Apr 20 11:48:23 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Tue, 20 Apr 2004 11:48:23 -0700 Subject: [syslinux] 3Com PXE server In-Reply-To: <200404201752.i3KHq7J24137@nwi.calumet.purdue.edu> References: <200404201752.i3KHq7J24137@nwi.calumet.purdue.edu> Message-ID: <40857077.60402@zytor.com> Jason Dravet wrote: > I understand my opinion does not count. I am only pointing out the > possibility of a bug existing. That point is reasonable because as I said > earlier it worked, was broken, fixed, and then broken again. If you don't > have the inclination to look at it, that is fine. > > If would please tell me or point me to some documentation on how to compile > your software on a windows platform then I will try to do some debugging on > my own. Since I know what versions work and which don't I can compare the > source code and see what changed from version to version. I think I might > be able to use kdiff3 at http://kdiff3.sourceforge.net/ to assist me. Sorry, compiling on Windows is not supported. It might work in Cygwin, but I haven't tried it. Anyway, the better way to look at this stuff is to get a packet dump on the network and observe in what way the pathology manifests on the wire. -hpa From smarfisi at emboot.com Tue Apr 20 12:32:06 2004 From: smarfisi at emboot.com (Steve Marfisi) Date: Tue, 20 Apr 2004 15:32:06 -0400 Subject: [syslinux] RE: Re: 3Com PXE server Message-ID: <6C5E123EDC988B48850FE656201CED8A1A706E@emboot-pdc.emboot.local> It was little misleadings like this in the past that placed Argon Technology in trouble, and yet they still persist. Nowhere does it say on the 3Com site that CMS replaces Boot Services, only that you have similar functionality. There are lower cost Windows-based alternatives that Jason can consider that will work. Steve Marfisi emBoot Inc. Date: Tue, 20 Apr 2004 08:48:04 -0400 From: conradl at argontechnology.com Subject: Re: [syslinux] 3Com PXE server To: "Jason Dravet" Cc: syslinux-bounces at zytor.com, syslinux at zytor.com 3Com Boot Services has been replaced by Argon's Client Management Services (CMS). I have used CMS with PXELinux v2.07 and it works fine. You can download a 15-day trial version and see if it resolves your problem. Visit http://www.argontechnology.com/cms for more information. Regards, Conrad Lawes Sr. Technical Analyst Argon Technology http://www.argontechnology.com Tel: (905)673-9978 ext. 14 From manu at kromtek.com Tue Apr 20 23:23:56 2004 From: manu at kromtek.com (Manu Abraham) Date: Wed, 21 Apr 2004 10:23:56 +0400 Subject: [syslinux] pxe linux In-Reply-To: <1020199682@web.de> References: <1020199682@web.de> Message-ID: <200404211023.56773.manu@kromtek.com> Hello Josef, As per the suggestion i made the changes to /tftpboot/pxelinux.cfg/default like this #Pxe configuration default tux label tux kernel linux append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 Still i get the same Kernel command line: BOOT_IMAGE=linux auto. No changes in that. I don't know where my append seems to be going... Regards, Manu > manu at kromtek.com schrieb am 20.04.04 19:46:48: > > On Tuesday 20 Apr 2004 9:17 pm, you wrote: > > > manu at kromtek.com schrieb am 20.04.04 19:10:09: > > > > My configuration now contains only so much. But still what i get on > > > > the client is the same command line BOOT_IMAGE=linux auto > > > > > > > > #Pxe configuration > > > > > > > > kernel linux > > > > append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 > > > > > > You are still missing the 'label linux' line. So in total this would > > > be: > > > > > > -- snip > > > #Pxe configuration > > > label linux > > > kernel linux > > > append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 > > > -- snip > > > > With the label line, there is no difference, whether it is there or not, > > i get the same error. > > Test it again. Without this you won't get any kernel command line, and the > default for pxelinux is to load kernel 'linux' without any arguments. > That's what you see here (the BOOT_IMAGE is added by pxelinux itself btw.). > > Perhaps you should really switch over to: > > - snip > default tux > label tux > kernel linux > append > - snip > > or even rename your kernel to something more useful (eg. tux.img). Seems > you have problems with the automagic file selection pxelinux does, and with > this you wouldn't have problems (or: Get warned/errored by pxelinux: file > not found etc.pp.). > > Regards, > > Josef > _______________________________________________________________________ > ... and the winner is... WEB.DE FreeMail! - Deutschlands beste E-Mail > ist zum 39. Mal Testsieger (PC Praxis 03/04) http://f.web.de/?mc=021191 From baurzhan.ismagulov at sbs.com.tr Wed Apr 21 01:09:29 2004 From: baurzhan.ismagulov at sbs.com.tr (Baurzhan Ismagulov) Date: Wed, 21 Apr 2004 10:09:29 +0200 Subject: [syslinux] pxe linux In-Reply-To: <200404202100.05658.manu@kromtek.com> References: <200404201254.35559.manu@kromtek.com> <200404201925.21268.manu@kromtek.com> <20040420155251.GH1761@ata.cs.hun.edu.tr> <200404202100.05658.manu@kromtek.com> Message-ID: <20040421080929.GA1277@ata.cs.hun.edu.tr> On Tue, Apr 20, 2004 at 09:00:05PM +0400, Manu Abraham wrote: > My configuration now contains only so much. But still what i get on the > client is the same command line BOOT_IMAGE=linux auto No idea. I would try using tcpdump to see what is going on in general. Specifically, which config file gets loaded by the client. Besides, what does pxelinux say? On my system, I see the following messages: Broadcom UNDI PXE-2.1 (build 082) v6.2.11 Copyright (C) 2000-2003 Broadcom Corporation Copyright (C) 1997-2000 Intel Corporation All rights reserved. Intel Base-Code, PXE-2.1 (build 083) Copyright (C) 1997-2001, Intel Corporation CLIENT MAC ADDR: 00 E0 81 51 D9 2B GUID: 00020003-0004-0005-0006-000700080009 CLIENT IP: 192.168.0.101 MASK: 255.255.255.0 DHCP IP: 192.168.0.10 PXELINUX 2.06 0x3f46af85 Copyright (C) 1994-2003 H. Peter Anvin UNDI data segment at: 000948C0 UNDI data segment size: 4C50 UNDI code segment at: 00099510 UNDI code segment size: 4804 PXE entry point found (we hope) at 9951:00D6 PXELINUX 2.06 0x3f46af85 Copyright (C) 1994-2003 H. Peter Anvin boot: Loading vmlinuz-2.4.21-186-smp......................... Loading rootfs.ext2.gz...................................................................................................................................................................................................................... Ready. ok Bootdata ok (command line is initrd=rootfs.ext2.gz ramdisk_size=65536 root=/dev/ram0 console=tty0 console=ttyS0,115200 BOOT_IMAGE=vmlinuz-2.4.21-186-smp auto) Linux version 2.4.21-186-smp (root at x86_64.suse.de) (gcc version 3.2.2 (SuSE Linux)) #1 SMP Mon Feb 2 17:10:45 UTC 2004 BIOS-provided physical RAM map: Which pxelinux version are you using? Do you see rootfs.img loaded? With kind regards, Baurjan. From manu at kromtek.com Wed Apr 21 06:44:29 2004 From: manu at kromtek.com (Manu Abraham) Date: Wed, 21 Apr 2004 17:44:29 +0400 Subject: [syslinux] pxe linux In-Reply-To: <20040421080929.GA1277@ata.cs.hun.edu.tr> References: <200404201254.35559.manu@kromtek.com> <200404202100.05658.manu@kromtek.com> <20040421080929.GA1277@ata.cs.hun.edu.tr> Message-ID: <200404211744.29632.manu@kromtek.com> > > Besides, what does pxelinux say? On my system, I see the following > messages: > I'm using Syslinux-2.08. This is what pxelinux says. Since i had a paniced system, i had to type it in by hand. Managed PC Boot Agent (MBA) v1.04 (BIOS integrated) (C) Copyright 2001 Lanworks Technologies Co. a subsisiary of 3Com Corporation. All rights reserved. Pre-boot eXecution Environment (PXE) v2.43 (C) Copyright 1999 Intel Corporation (C) Copyright 2000 Lanworks Technologies Co. a subsidiary of 3Com Corporation All rights reserved CLIENT MAC ADDR: 00 0E A6 22 28 21 GUID: 0002003-0004-0005-0006-000700080009 DHCP... CLIENT IP: 192.168.1.245 MASK: 255.255.255.0 DHCP IP: 192.168.1.254 GATEWAY IP: 192.168.1.1 PXELINUX 2.08 2003-12-12 Copyright (C) 1994-2003 H. Peter Anvin UNDI data segment at: 00098290 UNDI data segment size: 4801 UNDI code segment at: 0009CAA0 UNDI code segment size: 3160 PXE entry point found (we hope) at 9CAA:00FA My IP Address seems to be C0A801F5 192.168.1.245 ip=192.168.1.245:192.168.1.254:192.168.1.1:255.255.255.0 TFTP prefix: pxelinux.cfg/ Trying to load pxelinux.cfg/01-00-0e-a6-22-28-21 Trying to load pxelinux.cfg/C0A801F5 Trying to load pxelinux.cfg/C0A801F Trying to load pxelinux.cfg/C0A801 Trying to load pxelinux.cfg/C0A80 Trying to load pxelinux.cfg/C0A8 Trying to load pxelinux.cfg/C0A Trying to load pxelinux.cfg/C0 Trying to load pxelinux.cfg/C Trying to load pxelinux.cfg/default Loading Linux ................................ Ready Uncompressing linux ............Ok, booting the kernel. Linux version 2.4.25-lck1 (root at deploy)(gcc version 2.96 200000731 (Redhat Linu x 7.1 2.96-98)) #10 Thu Apr 15 20:30:41 GST 2004 BIOS provided physical RAM map: BIOS-e820:0000000000000000 - 000000000009fc00 (usable) BIOS-e820:000000000009fc00 - 00000000000a0000 (reserved) BIOS-e820:00000000000e8000 - 0000000000100000 (reserved) BIOS-e820:0000000000010000 - 000000001ff30000 (usable) BIOS-e820:0000000001ff3000 - 000000001ff40000 (ACPI data) BIOS-e820:0000000001ff4000 - 000000001fff0000 (ACPI NVS) BIOS-e820:0000000001fff000 - 0000000020000000 (reserved) BIOS-e820:000000000ffb8000 - 0000000100000000 (reserved) 0MB HIGHMEM available. 511MB LOWMEM available. found SMP MP-table at 000ff780 hm, page 000ff000 reserved twice. hm, page 0010000 reserved twice. hm, page 000f1000 reserved twice. hm, page 000f2000 reserved twice. On node 0 totalpages: 130864 zone(0): 4096 pages. zone(1): 126768 pages. zone(2): 0 pages Intel Multi Processor Specification v1.4 Virtual Wire compatibility mode. OEM ID: ASUSTek Product ID: ADIC at: 0xFEE00000 Processor #0 Pentium4(tm) XEON(tm) APIC Version 20 I/O APIC #2 Version 32 at 0xFEC00000. Enabling APIC mode: Flat.0Using 1 I/O APICs Processors: 1 Kernel command line: BOOTIMAGE=linux auto Initializing CPU#0 Detected 2599.174 Mhz processor. Console: colour VGA+ 80*25 Calibrating delay loop... 5144.57 BogoMIPS Memory" 514448K/523456 available(1521k kernel code,8620reserved,357k data, 272k init Ok highmem) Dentry cache hash table entries:65536 (order:7, 524288 bytes) Inode cache hash table entries : 32768(order:6, 262144 bytes) Mount cache hash table entries: 512 (order:0, 4096bytes) Buffer cache hash table entries : 32768(order:5 131072 bytes) Page-cache hash table entries: 131072(order:7, 524288 bytes) CPU: Trace cache: 12k nops, L1 Dcache:8k CPU: L2 cache:512k Intel machine check architecture supported. Intel machine check reporting enabled on CPU#0. CPU: Intel (R) Pentium(+R) 4 CPU 2.60 Ghz stepping 09 Enabling fast CPU save and restore ... done Enabling unmasked SIMD FPU exception support ... done Checking 'hlt' instruction ...OK POSIX conformance testing by UNIFIX enabled ExtINT on CPU#0 ESR value before enabling vector: 00000000 ESR value after enabling vector: 00000000 Enabling IO-APIC IRQs Setting 2 in the phys_id_present_map ...changing IO-APIC physical APIC ID to 2 ... ok. ... Timer:vector=0*31 pin1= 2 pin 2=0 testing the IO APIC ...... ...................... done Using local APIC tuner interrupts calibrating APIC timer... ..... CPU clock speed in 2598.0401 MHz. ......host bus clock speed is 199.0876 Mhz CPU: 0, clocks: 199876, slice 99938 CPU0 PCI:PCI BIOS revision 2.10 entry at 0xf0031, last bus = 2 PCI: Using configuration type 1 PCI: Probing PCI hardware PCI: Probing PCI hardware (bus00) PCI:Ignoring BARO-3 of IDE controller 00:1f.1 Transparent bridge-Intel corp. 82801 BA/CA/DB/EB PCI Bridge. PCI: Using IRQ router PIIX/ICH [8086/24d0] at 00:1F.0. PCI ->APIC IRQ transform:(B0,I31,P0) -> 18 --- snip --- RAMDISK drive initialized :16 RAM disks of 65536k size 1074 blocksize loop:loaded (max 8 devices) sk98 lin:Network Device Driverv6.22 (c) Copyright 1999-2004 Marvell (R) eth0: 3Com Gigabit LOM(3C940) Prefport:A Rlmt Mode: check link state -- snip -- lmod:failed to exec /sbin/modprobe-s-k block-major-3, errno=2 VFS: Cannot open root device " " or 03:02 Please append a correct "root="boot option Kernel panic :VFS:Unable to mount root fs on 03:02 That's it. Regards, Manu From baurzhan.ismagulov at sbs.com.tr Wed Apr 21 07:43:12 2004 From: baurzhan.ismagulov at sbs.com.tr (Baurzhan Ismagulov) Date: Wed, 21 Apr 2004 16:43:12 +0200 Subject: [syslinux] pxe linux In-Reply-To: <200404211744.29632.manu@kromtek.com> References: <200404201254.35559.manu@kromtek.com> <200404202100.05658.manu@kromtek.com> <20040421080929.GA1277@ata.cs.hun.edu.tr> <200404211744.29632.manu@kromtek.com> Message-ID: <20040421144312.GD1277@ata.cs.hun.edu.tr> On Wed, Apr 21, 2004 at 05:44:29PM +0400, Manu Abraham wrote: > Since i had a paniced system, i had to type it in by hand. Ouch! > Trying to load pxelinux.cfg/default > Loading Linux ................................ Ready > Uncompressing linux ............Ok, booting the kernel. The root fs is not loaded. Have you tried the config from Josef? It should have worked. Anyway, rename your kernel to zzz and try the following config: label zzz kernel zzz append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 I suspect a syntax problem in your config. With kind regards, Baurjan. From hpa at zytor.com Wed Apr 21 08:28:54 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 21 Apr 2004 08:28:54 -0700 Subject: [syslinux] pxe linux In-Reply-To: <200404211023.56773.manu@kromtek.com> References: <1020199682@web.de> <200404211023.56773.manu@kromtek.com> Message-ID: <40869336.6000408@zytor.com> Manu Abraham wrote: > Hello Josef, > As per the suggestion i made the changes to /tftpboot/pxelinux.cfg/default > like this > > #Pxe configuration > > default tux > label tux > kernel linux > append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 > > > Still i get the same Kernel command line: BOOT_IMAGE=linux auto. No changes > in that. I don't know where my append seems to be going... > It's not reading the config file at all, probably. -hpa From manu at kromtek.com Wed Apr 21 08:25:22 2004 From: manu at kromtek.com (Manu Abraham) Date: Wed, 21 Apr 2004 19:25:22 +0400 Subject: [syslinux] pxe linux In-Reply-To: <20040421144312.GD1277@ata.cs.hun.edu.tr> References: <200404201254.35559.manu@kromtek.com> <200404211744.29632.manu@kromtek.com> <20040421144312.GD1277@ata.cs.hun.edu.tr> Message-ID: <200404211925.22338.manu@kromtek.com> Hi, On Wednesday 21 Apr 2004 6:43 pm, Baurzhan Ismagulov wrote: > On Wed, Apr 21, 2004 at 05:44:29PM +0400, Manu Abraham wrote: > > Since i had a paniced system, i had to type it in by hand. > > Ouch! > > > Trying to load pxelinux.cfg/default > > Loading Linux ................................ Ready > > Uncompressing linux ............Ok, booting the kernel. > > The root fs is not loaded. Have you tried the config from Josef? It > should have worked. I tried that and i'm running that only even now, the config ... #Pxe configuration default tux label tux kernel linux append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 > Anyway, rename your kernel to zzz and try the following config: > > label zzz > kernel zzz > append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 > > I suspect a syntax problem in your config. > > > With kind regards, > Baurjan. From manu at kromtek.com Wed Apr 21 08:26:54 2004 From: manu at kromtek.com (Manu Abraham) Date: Wed, 21 Apr 2004 19:26:54 +0400 Subject: [syslinux] pxe linux In-Reply-To: <40869336.6000408@zytor.com> References: <1020199682@web.de> <200404211023.56773.manu@kromtek.com> <40869336.6000408@zytor.com> Message-ID: <200404211926.54812.manu@kromtek.com> On Wednesday 21 Apr 2004 7:28 pm, you wrote: > Manu Abraham wrote: > > Hello Josef, > > As per the suggestion i made the changes to > > /tftpboot/pxelinux.cfg/default like this > > > > #Pxe configuration > > > > default tux > > label tux > > kernel linux > > append initrd=rootfs.img ramdisk_size=65536 root=/dev/ram0 > > > > > > Still i get the same Kernel command line: BOOT_IMAGE=linux auto. No > > changes in that. I don't know where my append seems to be going... > > It's not reading the config file at all, probably. What could be done to check it up. I have a strnge feeling about that since whatever changes i make to the config, there seems to be no difference. Regards, Manu > -hpa From hpa at zytor.com Wed Apr 21 08:37:52 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 21 Apr 2004 08:37:52 -0700 Subject: [syslinux] pxe linux In-Reply-To: <200404211926.54812.manu@kromtek.com> References: <1020199682@web.de> <200404211023.56773.manu@kromtek.com> <40869336.6000408@zytor.com> <200404211926.54812.manu@kromtek.com> Message-ID: <40869550.1070301@zytor.com> Manu Abraham wrote: > > What could be done to check it up. I have a strnge feeling about that since > whatever changes i make to the config, there seems to be no difference. > Check the logs on your TFTP server; it looks like it *can* load the "linux" binary, but perhaps it's not from the server or the directory you think it's loading from, or something like that... From Thorsten.Klerlein at nigura.de Wed Apr 21 08:42:45 2004 From: Thorsten.Klerlein at nigura.de (Klerlein, Thorsten) Date: Wed, 21 Apr 2004 17:42:45 +0200 Subject: AW: [syslinux] pxe linux Message-ID: <91BAA1A68C70D711BC4E009027CCE45909474C@srvnig01.nigura-intern.de> ---snip--- PXE entry point found (we hope) at 9CAA:00FA My IP Address seems to be C0A801F5 192.168.1.245 ip=192.168.1.245:192.168.1.254:192.168.1.1:255.255.255.0 TFTP prefix: pxelinux.cfg/ ^^^^^^^^^^^^^^^^^^^^^^^^^what?s up with this? check your "filename" option in dhcpd.conf... ---snip--- Greets, Thorsten From dravet at calumet.purdue.edu Wed Apr 21 09:15:02 2004 From: dravet at calumet.purdue.edu (Jason Dravet) Date: Wed, 21 Apr 2004 11:15:02 -0500 Subject: [syslinux] 3Com PXE server In-Reply-To: <40857077.60402@zytor.com> Message-ID: <200404211615.i3LGF2J06839@nwi.calumet.purdue.edu> If compiling on windows is not supported then how is syslinux generated so it runs on windows? I have compiled several pieces of software on linux, but I was under the impression that the resulting software would not work in windows. Jason -----Original Message----- From: H. Peter Anvin [mailto:hpa at zytor.com] Sent: Tuesday, April 20, 2004 1:48 PM To: Jason Dravet Cc: syslinux-bounces at zytor.com; syslinux at zytor.com Subject: Re: [syslinux] 3Com PXE server Jason Dravet wrote: > I understand my opinion does not count. I am only pointing out the > possibility of a bug existing. That point is reasonable because as I said > earlier it worked, was broken, fixed, and then broken again. If you don't > have the inclination to look at it, that is fine. > > If would please tell me or point me to some documentation on how to compile > your software on a windows platform then I will try to do some debugging on > my own. Since I know what versions work and which don't I can compare the > source code and see what changed from version to version. I think I might > be able to use kdiff3 at http://kdiff3.sourceforge.net/ to assist me. Sorry, compiling on Windows is not supported. It might work in Cygwin, but I haven't tried it. Anyway, the better way to look at this stuff is to get a packet dump on the network and observe in what way the pathology manifests on the wire. -hpa From hpa at zytor.com Wed Apr 21 09:24:09 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 21 Apr 2004 09:24:09 -0700 Subject: [syslinux] 3Com PXE server In-Reply-To: <200404211615.i3LGF2J06839@nwi.calumet.purdue.edu> References: <200404211615.i3LGF2J06839@nwi.calumet.purdue.edu> Message-ID: <4086A029.6030804@zytor.com> Jason Dravet wrote: > If compiling on windows is not supported then how is syslinux generated so > it runs on windows? I have compiled several pieces of software on linux, > but I was under the impression that the resulting software would not work in > windows. > The Windows .exe is compiled on Linux using the mingw compiler. Note that it only applies to syslinux, not pxelinux or isolinux. pxelinux or isolinux just might compile on Windows using Cygwin, but I have never tested it. -hpa From hpa at zytor.com Wed Apr 21 09:25:16 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 21 Apr 2004 09:25:16 -0700 Subject: AW: [syslinux] pxe linux In-Reply-To: <91BAA1A68C70D711BC4E009027CCE45909474C@srvnig01.nigura-intern.de> References: <91BAA1A68C70D711BC4E009027CCE45909474C@srvnig01.nigura-intern.de> Message-ID: <4086A06C.7070104@zytor.com> Klerlein, Thorsten wrote: > ---snip--- > > PXE entry point found (we hope) at 9CAA:00FA > My IP Address seems to be C0A801F5 192.168.1.245 > ip=192.168.1.245:192.168.1.254:192.168.1.1:255.255.255.0 > TFTP prefix: pxelinux.cfg/ > ^^^^^^^^^^^^^^^^^^^^^^^^^what?s up with this? check your "filename" option > in dhcpd.conf... > Aaaah... he's put his kernel image and pxelinux.0 *inside* the pxelinux.cfg directory. Doesn't work. -hpa From hpa at zytor.com Wed Apr 21 22:45:33 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 21 Apr 2004 22:45:33 -0700 Subject: [syslinux] SYSLINUX 2.09-pre14 released, release candidate Message-ID: <40875BFD.4020908@zytor.com> This release includes Murali's menu positioning patch; it is otherwise identical to 2.09-pre13. I will release this as 2.09 on Sunday unless I hear objections. Note that I've effectively been out of the loop, moving houses, for almost a month, so if there is something I'm forgetting please tell me... -hpa From hpa at zytor.com Wed Apr 21 22:55:39 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 21 Apr 2004 22:55:39 -0700 Subject: [syslinux] knoppix boot fail probably due to bad int 820 In-Reply-To: <200403301440.i2UEedXK014569@Tempo.Update.UU.SE> References: <200403261010.i2QAA0PJ032431@Tempo.Update.UU.SE> <40685467.4030202@zytor.com> <200403301440.i2UEedXK014569@Tempo.Update.UU.SE> Message-ID: <40875E5B.6080203@zytor.com> Fredrik Stax{ng wrote: > "H. Peter Anvin" writes: > > >>Fredrik Stax{ng wrote: >> >>>Knoppix says initrd extends beyond end of memory. I thing this >>>is the same problem as >>> >>>http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=146350&archive=yes >>> >> >>Please post the bootup messages. > > > I saw your reply to Anthony Marshall before I saw this. I inferred that > you do not care about buggy BIOSes, so I flashed in the latest version. > It boots Knoppix now, but unfortunately it does not work with my new > hard disk anymore ;-[ > > If you look at the link above you will find a printout of the E820 table. > 2.4.19 (at least) contains code to "Sanitize the BIOS e820 map" in > arch/i386/kernel/setup.c. > I took a look at the bug report listed, and SYSLINUX will have the same problem as LILO with this particular machine. The solution is to specify the appropriate mem= parameter (specifically mem=393152K). -hpa From magdarennels at caths.co.uk Thu Apr 22 01:54:36 2004 From: magdarennels at caths.co.uk (jeffry bautista) Date: Thu, 22 Apr 2004 02:54:36 -0600 Subject: [syslinux] =?iso-8859-1?q?Ffi_The_doctor=A1=AFs_in?= Message-ID: clarisworks ogara chshz npt-d acked Take charge of your medical conditions now. Order online. Live a healthier and happier life - all from the comfort of your home! Quality Meds we offer: V`ic0-din ES 90 Tab Special. Fedex delivery. http://hclgyt.net.azsw3d.com/?p=8089 Your comfort and convenience is our prime concern. N.o thanks: http://uhbq.j.foresmdb.com/a.html A professor was giving a big test one day to his students. He handed out all of the tests and went back to his desk to wait. Once the test was over the students all handed the tests back in. The professor noticed that one of the students had attached a $100 bill to his test with a note saying "A dollar per point." The next class the professor handed the tests back out. This student got back his test and $64 change. A young man was walking through a super market to pick up a few things when he noticed an old lady following him. ??Pardon me," she said. "I'm sorry if my staring at you has made you feel uncomfortable. It's just that you look just like my son, who died recently.?? I??m very sorry," replied the young man, "is there anything I can do for you??? Yes," she said, "as I'm leaving, would you say?? Goodbye, mother?' It would make me feel so much better.?? Sure," answered the young man. As the old woman was leaving, he called out,?? Goodbye, Mother!" Then, as he stepped up to the checkout counter, he saw that his total was $127.00. "How can that be?" he asked. "I only purchased a few things!?? The clerk replied, "Your mother said you'd pay for her." kom1habaka02konbeya-,irumine insatsuk. From huberstuff at hsn.at Thu Apr 22 04:40:58 2004 From: huberstuff at hsn.at (huberstuff at hsn.at) Date: Thu, 22 Apr 2004 13:40:58 +0200 (CEST) Subject: [syslinux] PXE boot + Novell Message-ID: <12211.57.90.33.70.1082634058.squirrel@webmail.hsn.at> Dear Friends, Does anybody know how to configure a Novell DHCP sp that i can use pxelinux? The problems are that the clients get no ip adress. we switched on bootp, but still not working, is there a small description somewhere on the net which shows how to deal with novell? thanks in advance markus From kasper at easyspeedy.com Thu Apr 22 04:58:48 2004 From: kasper at easyspeedy.com (Kasper Fock) Date: Thu, 22 Apr 2004 13:58:48 +0200 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD In-Reply-To: <408455DF.80902@zytor.com> References: <20040405212005.GA76471@zouh.org> <200404200021.15138.kasper@easyspeedy.com> <408455DF.80902@zytor.com> Message-ID: <200404221358.48573.kasper@easyspeedy.com> Thanks for the answers. I get pxelinux to boot pxeboot but it the hangs during the BTX. it prints: TFTP. PXE Loader 1.00 Building the boot loader arguments Relocating the loader and the BTX If the pxeboot i loaded directly. (not using pxelinux it works fine) My config file for pxelinux, to load pxeboot look like this: label freebsd KERNEL img/memdisk append ramdisk_size=100000 network load_ramdisk=1 prompt_ramdisk=0 initrd=img/freebsd5.2.1/pxeboot vga=788 root=/dev/ram Is this the right way to load pxeboot from pxelinux? or does anybody have an suggestions to get this setup to work? /Kasper On Tuesday 20 April 2004 00:42, H. Peter Anvin wrote: > Kasper Fock wrote: > > Hi. > > I had managed to boot the freebsd boot.flp as described, but i also want > > to be able to install it. Does anyone know how to make an unattended > > install of freebsd using this method (pxelinux, memdisk and boot.flp). I > > have read something about an install.cfg file but is it possible to get > > the installation to look for this file ie, on a ftp/tftp server? > > > > I wold really like to use pxelinux instead of pxeboot (bsd pxeloader) > > while i want to install a mixture of *bsd and unix > > Use pxelinux to boot pxeboot. > > -hpa -- Kasper Fock EasySpeedy ApS H?sterk?bvej 33 2970 H?rsholm _______________________________________ European Dedicated Server Hosting Extremely low prices, secure, and reliable Linux distributions only _______________________________________ From huberstuff at hsn.at Thu Apr 22 05:26:16 2004 From: huberstuff at hsn.at (huberstuff at hsn.at) Date: Thu, 22 Apr 2004 14:26:16 +0200 (CEST) Subject: [syslinux] PXE boot + Novell In-Reply-To: References: Message-ID: <8228.57.90.33.70.1082636776.squirrel@webmail.hsn.at> Hallo Peter, danke f?r deine Antwort. Ich glaube wir haben Zenworks for Desktops installiert (ich bin leider kein Novell Spezi). Aber wir wollen lediglich den DHCP Server von Novell verwenden. Er soll auf anfrage der pxe clients die tftp-server adresse (<> DHCP Server) und das Bootimage zur?ckliefern). Als Tftpd wollen wir den Microsoft TFTPD Daemon verwenden. lg Markus > > Hallo Markus, > > habt Ihr auf euren Novell Servern Zenworks for Desktops installiert? Ohne > Zenworks - kein PXE. > Es sollte TFTP.NLM, PDHCP.NLM und f?r Zenworks auch DTS.NLM geladen > werden. Die Dateien stehen normalerweise unter SYS:/SYSTEM. > Die wesentlichen Einstellungen stehen in den ini-Dateien PDHCP.ini und > TFTP.ini im Systemverzeichnis. > > Wenn es nur einen TFTP-Server gibt, sollten Transaction_Server_IP (nur f?r > Zenworks Imaging) TFTP_Server_IP (die IP-Adresse des TFTP-Servers) und > PDHCP_Server_IP (die Adresse des DHCP-Servers - PDHCP weil Preboot DHCP) > auf dieselbe IP-Adresse verweisen. Falls man vor hat gr??ere Dateien zu > ?bertragen legt man das TFTP-Verzeichnis in der TFTP.ini am besten auf > Vol1: um. > > > Peter Leenders > CC CompuNet AG & Co. oHG > A part of Computacenter plc > Europaring 34-40, 50170 Kerpen, Germany > Phone: +49 (0)2273 / 597-7972 (-219), Fax: +49 (0)2273 / 597-7515, > E-Mail : Peter.Leenders at computacenter.com > Visit us on the Internet: http://www.computacenter.com > Visit our Online-Shop: http://www.compunet.de/connect > > > This email is confidential. If you are not the intended recipient, > you must not disclose or use the information contained in it. > If you have received this mail in error, please tell us > immediately by return email and delete the document. > > > > > > COMPUTACENTER (UK) LTD > > The contents of this e-mail are intended for the named addressee only. > It contains information which may be confidential and which may also be > privileged. > Unless you are the named addressee (or authorised to receive > for the addressee) you may not copy or use it, or disclose it to anyone > else. If you received it in error please notify us immediately and then > destroy it. > > Computacenter information is available from http://www.computacenter.com > > This footnote also confirms that this email message has been swept for the > presence of computer viruses. > > *************************************************************************************************************************** > > From hpa at zytor.com Thu Apr 22 09:29:28 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Thu, 22 Apr 2004 09:29:28 -0700 Subject: [syslinux] memdisk : booting OpenBSD and NetBSD In-Reply-To: <200404221358.48573.kasper@easyspeedy.com> References: <20040405212005.GA76471@zouh.org> <200404200021.15138.kasper@easyspeedy.com> <408455DF.80902@zytor.com> <200404221358.48573.kasper@easyspeedy.com> Message-ID: <4087F2E8.9020400@zytor.com> Kasper Fock wrote: > Thanks for the answers. > I get pxelinux to boot pxeboot but it the hangs during the BTX. > > it prints: > TFTP. > PXE Loader 1.00 > > Building the boot loader arguments > Relocating the loader and the BTX > > > If the pxeboot i loaded directly. (not using pxelinux it works fine) > My config file for pxelinux, to load pxeboot look like this: > > label freebsd > KERNEL img/memdisk > append ramdisk_size=100000 network load_ramdisk=1 prompt_ramdisk=0 > initrd=img/freebsd5.2.1/pxeboot vga=788 root=/dev/ram > > > Is this the right way to load pxeboot from pxelinux? or does anybody have an > suggestions to get this setup to work? > No, it's not. Don't use memdisk. Load pxeboot.0 directly (kernel pxeboot.0). -hpa From B.Blaauw at student.fontys.nl Thu Apr 22 10:56:05 2004 From: B.Blaauw at student.fontys.nl (Blaauw,Bernd B.) Date: Thu, 22 Apr 2004 19:56:05 +0200 Subject: [syslinux] SYSLINUX 2.09-pre14 released, release candidate Message-ID: <11584391EF34234392426DA50F9E1F9387A2C3@selenga.fontys.nl> HPA, Is it possible to extend Memdisk so that it contains information about which SYSLINUX component booted it? with Murali's help/work I have been able to detect Memdisk parameters from DOS. However knowing if the loader is Isolinux, PXElinux or Syslinux would help me a lot. (for example: determining if a cdrom-driver should be loaded, and if so: which one: Eltorito driver from www.nu2.nu which only works with Isolinux, or a generic IDE/Atapi cdrom-driver). a powerfull way to apply this MEMDISK-parameter detection has been documented at: http://www.911cd.net/forums/index.php?showtopic=5706 Bernd Blaauw From manu at kromtek.com Tue Apr 20 04:52:13 2004 From: manu at kromtek.com (Manu Abraham) Date: Tue, 20 Apr 2004 15:52:13 +0400 Subject: [syslinux] pxe linux In-Reply-To: <20040420094552.GA1761@ata.cs.hun.edu.tr> References: <200404201254.35559.manu@kromtek.com> <20040420094552.GA1761@ata.cs.hun.edu.tr> Message-ID: <200404201552.13111.manu@kromtek.com> Hi, I tried adding root=/dev/ram0 as well as the keep pxe option. Still i don't see any changes. One thing what i noticed in the syslog ..... Apr 20 16:56:28 zeus dhcpd: DHCPDISCOVER from 00:0e:a6:22:28:21 via eth0 Apr 20 16:56:28 zeus dhcpd: DHCPOFFER on 192.168.1.245 to 00:0e:a6:22:28:21 via eth0 Apr 20 16:56:32 zeus dhcpd: DHCPREQUEST for 192.168.1.245 (192.168.1.254) from 00:0e:a6:22:28:21 via eth0 Apr 20 16:56:32 zeus dhcpd: DHCPACK on 192.168.1.245 to 00:0e:a6:22:28:21 via eth0 ---->Apr 20 12:56:32 zeus in.tftpd[4486]: tftp: client does not accept options does this mean that whatever options i have as append in the default config, those parameters are not passed to the kernel ? Regards, Manu On Tuesday 20 Apr 2004 1:45 pm, Baurzhan Ismagulov wrote: > Hello, Manu! > > On Tue, Apr 20, 2004 at 12:54:35PM +0400, Manu Abraham wrote: > > VFS: Cannot open root device "" or 03:02 > > Please append a correct "root=" boot option... > > Kernel Panic: VFS: unable to mount root fs on 03:02 > > Perhaps "append initrd=rootfs.img root=/dev/ram0" in pxelinux config? > > With kind regards, > Baurjan. > > _______________________________________________ > SYSLINUX mailing list > Submissions to SYSLINUX at zytor.com > Unsubscribe or set options at: > http://www.zytor.com/mailman/listinfo/syslinux > Please do not send private replies to mailing list traffic. From manu at kromtek.com Wed Apr 21 09:35:19 2004 From: manu at kromtek.com (Manu Abraham) Date: Wed, 21 Apr 2004 20:35:19 +0400 Subject: [syslinux] pxelinux Message-ID: <200404212035.19194.manu@kromtek.com> Hi, At last found the error, i had my kernel, pxelinux.0, initrd inside the pxelinux.cfg directory. A hearty thanks to all you guys. That was real big help. I found the documentation a little bit confusing that's why it happened. Somebody else also had the same trouble as i found while searching the archives. I made a small patch, a simple diff to the pxelinux documentation (pxelinux.doc), if you people feel it's up to any good please do update the documentation. Thanks again, Regards, Manu From hpa at zytor.com Thu Apr 22 15:06:57 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Thu, 22 Apr 2004 15:06:57 -0700 Subject: [syslinux] PXE boot + Novell In-Reply-To: <8228.57.90.33.70.1082636776.squirrel@webmail.hsn.at> References: <8228.57.90.33.70.1082636776.squirrel@webmail.hsn.at> Message-ID: <40884201.2000801@zytor.com> Please keep the language on this list English, please. -hpa From hpa at zytor.com Thu Apr 22 21:15:46 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Thu, 22 Apr 2004 21:15:46 -0700 Subject: [syslinux] knoppix boot fail probably due to bad int 820 In-Reply-To: <40875E5B.6080203@zytor.com> References: <200403261010.i2QAA0PJ032431@Tempo.Update.UU.SE> <40685467.4030202@zytor.com> <200403301440.i2UEedXK014569@Tempo.Update.UU.SE> <40875E5B.6080203@zytor.com> Message-ID: <40889872.4030906@zytor.com> H. Peter Anvin wrote: > Fredrik Stax{ng wrote: > >> "H. Peter Anvin" writes: >> >> >>> Fredrik Stax{ng wrote: >>> >>>> Knoppix says initrd extends beyond end of memory. I thing this is >>>> the same problem as >>>> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=146350&archive=yes >>> >>> Please post the bootup messages. >> >> I saw your reply to Anthony Marshall before I saw this. I inferred that >> you do not care about buggy BIOSes, so I flashed in the latest version. >> It boots Knoppix now, but unfortunately it does not work with my new >> hard disk anymore ;-[ >> >> If you look at the link above you will find a printout of the E820 table. >> 2.4.19 (at least) contains code to "Sanitize the BIOS e820 map" in >> arch/i386/kernel/setup.c. >> > > I took a look at the bug report listed, and SYSLINUX will have the same > problem as LILO with this particular machine. > > The solution is to specify the appropriate mem= parameter (specifically > mem=393152K). > I modified the E820 parser somewhat, and I'm hoping it should now be able to detect errors like the one above. Please test out 2.09-pre15, since I really want to release it this weekend. -hpa From hpa at zytor.com Thu Apr 22 21:17:18 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Thu, 22 Apr 2004 21:17:18 -0700 Subject: [syslinux] SYSLINUX 2.09-pre15 Message-ID: <408898CE.8030202@zytor.com> I just released 2.09-pre15; it has a slightly improved E820 memory parser, which I'm hoping will deal with the "incorrect length" BIOS bug some people reported. If you have a machine with this bug, *PLEASE* test out this release, since I really want to release 2.09 this weekend. ftp://ftp.kernel.org/pub/linux/utils/boot/syslinux/Testing/ Thanks! -hpa From patl at users.sourceforge.net Sat Apr 24 11:36:41 2004 From: patl at users.sourceforge.net (Patrick J. LoPresti) Date: Sat, 24 Apr 2004 18:36:41 -0000 Subject: [syslinux] Problem with FreeDOS ke2034_32 + PXELINUX + memdisk Message-ID: I am using FreeDOS kernel 2.0.34 with FAT32 support (ke2034_32), together with PXELINUX and memdisk from SYSLINUX 2.0.9-pre15. I am experiencing this problem on multiple test systems (Dell Precision 450 and IBM T22 so far). My pxelinux.cfg/default has these lines: label test kernel memdisk append initrd=test.img keeppxe ...which let me boot my 1.44M "test.img" floppy image with FreeDOS. My problem is that config.sys is not being read at all. I am certain it is not being read because: 1) I receive no "press F5/F8" prompt 2) I see no messages from the items in my config.sys (e.g., himem64.exe). 3) When autoexec.bat attempts to start the network, it fails saying that the ifshlp.sys driver has not been loaded. If I downgrade the FreeDOS kernel to 2.0.33 (ke2033_32), keeping everything else identical, the problem vanishes; my config.sys is read properly. If I remove "keeppxe" from the PXELINUX configuration file, the problem also vanishes. I am open to suggestions for how to proceed. Thanks! - Pat From patl at users.sourceforge.net Sat Apr 24 12:41:41 2004 From: patl at users.sourceforge.net (Patrick J. LoPresti) Date: Sat, 24 Apr 2004 19:41:41 -0000 Subject: [syslinux] [PATCH] Support WinImage compressed floppies Message-ID: The following patch allows memdisk to support compressed floppy images created by WinImage. WinImage creates a PKZIP-style archive with a single member, the .img (or .ima) file containing the floppy image. WinImage names such archives with a .imz extension. This patch lifts the header/trailer checking code from gunzip() to a new function, check_zip(), which both checks for a valid ZIP image and returns the CRC, size, and offset of the actual compressed data. check_zip() supports both gzip and PKZIP archives. In addition, this patch adds some whitespace to init.S16 to prevent GCC 3.3 from bombing out. I have tested this patch thoroughly with different types of images on several different machines and they all seem to work. This is not too surprising since I was careful not to alter the original code's behavior except when presented with a file whose magic number indicates a PKZIP archive. (And even then, all it does is use a different offset and checksum for the ZIP data.) This patch applies cleanly against SYSLINUX 2.09-pre15. I propose it for inclusion in some future release. - Pat -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: memdisk-pkzip-patch URL: From hpa at zytor.com Mon Apr 26 23:57:32 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Mon, 26 Apr 2004 23:57:32 -0700 Subject: [syslinux] SYSLINUX 2.09, 2.10-pre1 released Message-ID: <408E045C.1080404@zytor.com> Hi all, I have released SYSLINUX 2.09; it is the same as 2.09-pre15 except for the version number. Changes in 2.09: * SYSLINUX: Remove residual setuid crap from syslinux-nomtools. * Handle video pages correctly when using the API functions. * Handle compiling on an x86-64 platform correctly. * Menu system from Murali Krishnan Ganapathy; see the menu directory for information. * COMBOOT: Allow COMBOOT programs to change screen text mode. * COMBOOT: Correct the documentation of how to detect SYSLINUX from COMBOOT!!!! * COMBOOT: Fix "get key without echo" API function. * SYSLINUX: Fix bug that affected the API open function. * ALL: Improve the E820 memory parser, to work around some buggy BIOSes. I have *also* released 2.10-pre1, which adds Patrick LoPresti's support for zipped (as opposed to gzipped) images in MEMDISK; some Windows-based tools apparently can handle zipped images but not gzipped, and this makes it easier to just drop them in. ftp://ftp.kernel.org/pub/linux/utils/boot/syslinux/ -hpa From tenthumbs at cybernex.net Tue Apr 27 06:25:04 2004 From: tenthumbs at cybernex.net (TenThumbs) Date: Tue, 27 Apr 2004 09:25:04 -0400 Subject: [syslinux] Gcc 3.4.0 and syslinux-2.09 menu Message-ID: <408E5F30.5DF4492A@cybernex.net> The menu directory won't compile with gcc 3.4.0 because gcc complains that "ebp" cannot be used as a constraint. If preserving ebp is necessary then it will have to be done in the assembler code. Gcc 3.3.3 just seems to ignore the constraint. It didn't do anything special to preserve the register. I also noticed that getnumrows returns the contents of 0x484 which is the number of rows minus one but menu.c has this ms->numcols = getnumcols(); ms->numrows = getnumrows(); ms->maxcol = ms->numcols - 1; ms->maxrow = ms->numrows - 1; which makes maxrow = rows - 2. Is this intentional? BTW, if you put this line __asm__ (".code16gcc\n\t"); in an include file and include that file at the top of the .c files then you can simplify the make file considerably. Thanks. -- Ceres 2004-04-27 13:23:45.258 UTC (JD 2453123.058163) X = -1.793726944, Y = 1.473638721, Z = 1.059643564 X' = -0.007481985, Y' = -0.007967670, Z' = -0.002227873 From gmurali at cs.uchicago.edu Tue Apr 27 08:18:21 2004 From: gmurali at cs.uchicago.edu (Murali Krishnan Ganapathy) Date: Tue, 27 Apr 2004 10:18:21 -0500 Subject: [syslinux] Gcc 3.4.0 and syslinux-2.09 menu In-Reply-To: <408E5F30.5DF4492A@cybernex.net> References: <408E5F30.5DF4492A@cybernex.net> Message-ID: <408E79BD.8090800@cs.uchicago.edu> TenThumbs wrote: >I also noticed that getnumrows returns the contents of 0x484 which is >the number of rows minus one but menu.c has this > ms->numcols = getnumcols(); > ms->numrows = getnumrows(); > ms->maxcol = ms->numcols - 1; > ms->maxrow = ms->numrows - 1; >which makes maxrow = rows - 2. Is this intentional? > > > According to the info I read "40:84 byte Rows on the screen (less 1, EGA+)" I interpreted it as meaning, that on EGA+ it is one less than the number of rows and otherwise it is equal to the number of rows. I did not want to write code to figure out if I am working with an EGA+ or otherwise (which might complicate matters further if there is going to be serial console support). So I took the conservative approach. If somebody knows for sure that this number is always one less than the number of rows then a change to biosio.c is in order. - Murali From tenthumbs at cybernex.net Tue Apr 27 09:43:47 2004 From: tenthumbs at cybernex.net (TenThumbs) Date: Tue, 27 Apr 2004 12:43:47 -0400 Subject: [syslinux] Gcc 3.4.0 and syslinux-2.09 menu References: <408E5F30.5DF4492A@cybernex.net> <408E79A7.8090200@cs.uchicago.edu> Message-ID: <408E8DC3.18E2E7DD@cybernex.net> Murali Krishnan Ganapathy wrote: > > According to the info I read > > "40:84 byte Rows on the screen (less 1, EGA+)" > > I interpreted it as meaning, that on EGA+ it is > one less than the number of rows and otherwise it is equal to > the number of rows. > > I did not want to write code to figure out if I am working with an > EGA+ or otherwise (which might complicate matters further if there > is going to be serial console support). So I took the conservative > approach. > > If somebody knows for sure that this number is always one less than > the number of rows then a change to biosio.c is in order. > > - Murali I believe EGA+ means EGA or higher. My BIOS always returns rows - 1 with my modern card. See also adjust_screen at the end of font.inc. -- Moon 2004-04-27 16:41:04.527 UTC (JD 2453123.195191) X = -0.794338804, Y = -0.564884774, Z = -0.244721580 X' = 0.009814218, Y' = -0.012874930, Z' = -0.005580928 From Anthony.Golia at MorganStanley.com Tue Apr 27 10:05:08 2004 From: Anthony.Golia at MorganStanley.com (Anthony.Golia at MorganStanley.com) Date: Tue, 27 Apr 2004 13:05:08 -0400 (EDT) Subject: [syslinux] anybody get syslinux to load an iso? Message-ID: e.g. winpe.iso? Cheers, Anthony From hpa at zytor.com Tue Apr 27 10:47:41 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Tue, 27 Apr 2004 10:47:41 -0700 Subject: [syslinux] Gcc 3.4.0 and syslinux-2.09 menu In-Reply-To: <408E79BD.8090800@cs.uchicago.edu> References: <408E5F30.5DF4492A@cybernex.net> <408E79BD.8090800@cs.uchicago.edu> Message-ID: <408E9CBD.4050404@zytor.com> Murali Krishnan Ganapathy wrote: > TenThumbs wrote: > >> I also noticed that getnumrows returns the contents of 0x484 which is >> the number of rows minus one but menu.c has this >> ms->numcols = getnumcols(); >> ms->numrows = getnumrows(); >> ms->maxcol = ms->numcols - 1; >> ms->maxrow = ms->numrows - 1; >> which makes maxrow = rows - 2. Is this intentional? >> >> >> > According to the info I read > > "40:84 byte Rows on the screen (less 1, EGA+)" > > I interpreted it as meaning, that on EGA+ it is one less than the number > of rows and otherwise it is equal to the number of rows. No, that's wrong; it means that for EGA or better (i.e. not MDA, Hercules or CGA) it is rows on screen - 1; for MDA, Hercules and CGA this field is undefined. > I did not want to write code to figure out if I am working with an EGA+ > or otherwise (which might complicate matters further if there is going > to be serial console support). So I took the conservative approach. > > If somebody knows for sure that this number is always one less than the > number of rows then a change to biosio.c is in order. See above. The *correct* thing to do is to detect EGA presence (which includes VGA), if so use this value, otherwise use the hard-coded value 24. In practice I think this is safe to skip these days, it's a matter how anal-retentive you want to be (I always advocate anal-retentive programming.) There is another valuable aspect to this: if you detect EGA or better, it is safe to write directly to video RAM. This can be a significant speedup, but I don't think it matters for the menu system. Here is a getnumrows() which detects the proper number of rows (as opposed to rows-1): getnumrows: pushl %ebp pushl %ebx pushl %esi pushl %edi # First, detect if we have EGA or better movl $0x12,%ah # Video Subsystem Configuration movb $0x10,%bl # Return EGA info int $0x10 movl $25,%eax # If no EGA, 25 lines cmpb $0x10,%bl # No EGA BIOS? je 1f # Get width from BIOS location 0x484 xorw %ax,%ax movw %ax,%es movzbl %es:0x484,%eax inc %ax 1: popl %edi popl %esi popl %ebx popl %ebp ret From arekm at pld-linux.org Tue Apr 27 11:04:27 2004 From: arekm at pld-linux.org (Arkadiusz Miskiewicz) Date: Tue, 27 Apr 2004 20:04:27 +0200 Subject: [syslinux] amd64 and syslinux in 2.09 Message-ID: <200404272004.27145.arekm@pld-linux.org> According to Changes in 2.09: * Handle compiling on an x86-64 platform correctly. Unfortunately: amd64-pld-linux-gcc -march=i386 -c -o c32exit.o c32exit.S c32exit.S: Assembler messages: c32exit.S:8: Error: `4(%esp)' is not a valid 64 bit base/index expression Full buildlog: http://buildlogs.pld-linux.org/index.php?idx=5&ok=0&id=a07bf7319aae9c29879b1d6486f9e08b -- Arkadiusz Mi?kiewicz CS at FoE, Wroclaw University of Technology arekm.pld-linux.org, 1024/3DB19BBD, JID: arekm.jabber.org, PLD/Linux From gmurali at cs.uchicago.edu Tue Apr 27 11:09:38 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Tue, 27 Apr 2004 13:09:38 -0500 Subject: [syslinux] anybody get syslinux to load an iso? In-Reply-To: References: Message-ID: <408EA1E2.6090700@cs> I dont think syslinux or any derivative can load an iso. However it may be possible for syslinux to boot of a CD. However, if you just want a single CD which has ISOlinux and the contents of "winpe.iso", then it is possible. The documentation at http://people.cs.uchicago.edu/~gmurali/gui/miniwin.html might be helpful. - Murali Anthony.Golia at MorganStanley.com wrote: >e.g. winpe.iso? > >Cheers, >Anthony > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From vberdin at eazix.com Tue Apr 27 23:38:54 2004 From: vberdin at eazix.com (Vic Berdin) Date: Wed, 28 Apr 2004 14:38:54 +0800 Subject: [syslinux] Lilo like menuing system for Syslinux Message-ID: Hi, I'm sure this has been posted before. And I'm just wondering if anyone has ever implemented it. I just thought that this is really a nice feature for Syslinux. TIA, Vic --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.656 / Virus Database: 421 - Release Date: 4/9/2004 From gmurali at cs.uchicago.edu Wed Apr 28 07:52:27 2004 From: gmurali at cs.uchicago.edu (Murali Krishnan Ganapathy) Date: Wed, 28 Apr 2004 09:52:27 -0500 Subject: [syslinux] Lilo like menuing system for Syslinux In-Reply-To: References: Message-ID: <408FC52B.6090106@cs.uchicago.edu> Check out the menu subdirectory of syslinux 2.09! - Murali Vic Berdin wrote: >Hi, > >I'm sure this has been posted before. And I'm just wondering >if anyone has ever implemented it. I just thought that this >is really a nice feature for Syslinux. > >TIA, > >Vic > >--- >Outgoing mail is certified Virus Free. >Checked by AVG anti-virus system (http://www.grisoft.com). >Version: 6.0.656 / Virus Database: 421 - Release Date: 4/9/2004 > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From aaron.mcsorley at amd.com Wed Apr 28 08:14:53 2004 From: aaron.mcsorley at amd.com (Aaron McSorley) Date: Wed, 28 Apr 2004 10:14:53 -0500 (CDT) Subject: [syslinux] Menu system limitations In-Reply-To: <408FC52B.6090106@cs.uchicago.edu> Message-ID: Are there any hard limits to the amount of items and menus that can be in one comboot menu file. Or does this depend on you hardware setup. Currently I have sixteen menus and eightyfive items in one comboot file. It's been running well for almost a month (good job guys). The number of items will grow significantly and I want to be aware of any limitations. Thanks -Aaron From gmurali at cs.uchicago.edu Wed Apr 28 08:49:42 2004 From: gmurali at cs.uchicago.edu (Murali Krishnan Ganapathy) Date: Wed, 28 Apr 2004 10:49:42 -0500 Subject: [syslinux] Re: Menu system limitations In-Reply-To: References: Message-ID: <408FD296.6020503@cs.uchicago.edu> The only limit AFAIK is memory. All memory is allocated statically at compile time. The maximum size for a COM file is 64K. The code plus all the data should fit into this. Assuming about 256 bytes per item (that is on the higher side), and 10K for code, and about 6K for the stack, you have a limit of about 200 items. If you go over the limit, you should not be able to compile the code (the compiler should complain that your final executable is more than 64K). If your items can be dynamically generated, then you may be able to leverage the additional 64K available for COM files (this would mean changing the code, since currently the code assumes a 16bit pointer). If you need more than 200 and odd items and they are static, I see one simple workaround... * Cut your menu tree at some level (say level 2). Then you will have a mainmenu.com which displays the first two levels of the menu, and for each possible level 2 choice, you will have a different menu.com file. Choosing a item from the mainmenu, will call one of the helper menu.com. - Murali Aaron McSorley wrote: > Are there any hard limits to the amount of items and menus that can >be in one comboot menu file. Or does this depend on you hardware setup. >Currently I have sixteen menus and eightyfive items in one comboot file. >It's been running well for almost a month (good job guys). The number of >items will grow significantly and I want to be aware of any limitations. > >Thanks > -Aaron > > > > > > From hpa at zytor.com Wed Apr 28 11:20:35 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 28 Apr 2004 11:20:35 -0700 Subject: [syslinux] Re: Menu system limitations In-Reply-To: <408FD296.6020503@cs.uchicago.edu> References: <408FD296.6020503@cs.uchicago.edu> Message-ID: <408FF5F3.8020807@zytor.com> Murali Krishnan Ganapathy wrote: > The only limit AFAIK is memory. All memory is allocated statically at > compile time. The maximum size for a COM file is 64K. The code plus all > the data should fit into this. Assuming about 256 bytes per item (that > is on the higher side), and 10K for code, and about 6K for the stack, > you have a limit of about 200 items. If you go over the limit, you > should not be able to compile the code (the compiler should complain > that your final executable is more than 64K). > > If your items can be dynamically generated, then you may be able to > leverage the additional 64K available for COM files (this would mean > changing the code, since currently the code assumes a 16bit pointer). If > you need more than 200 and odd items and they are static, I see one > simple workaround... > > * Cut your menu tree at some level (say level 2). Then you will have a > mainmenu.com which displays the first two levels of the menu, and for > each possible level 2 choice, you will have a different menu.com file. > Choosing a item from the mainmenu, will call one of the helper menu.com. > Well, the simplest way to undo that limitation would be to switch to com32. -hpa From gmurali at cs.uchicago.edu Wed Apr 28 12:23:43 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Wed, 28 Apr 2004 14:23:43 -0500 Subject: [syslinux] Re: Menu system limitations In-Reply-To: <408FF5F3.8020807@zytor.com> References: <408FD296.6020503@cs.uchicago.edu> <408FF5F3.8020807@zytor.com> Message-ID: <409004BF.8030905@cs> H. Peter Anvin wrote: > Murali Krishnan Ganapathy wrote: > >> The only limit AFAIK is memory. All memory is allocated statically at >> compile time. The maximum size for a COM file is 64K. The code plus >> all the data should fit into this. Assuming about 256 bytes per item >> (that is on the higher side), and 10K for code, and about 6K for the >> stack, you have a limit of about 200 items. If you go over the limit, >> you should not be able to compile the code (the compiler should >> complain that your final executable is more than 64K). >> >> If your items can be dynamically generated, then you may be able to >> leverage the additional 64K available for COM files (this would mean >> changing the code, since currently the code assumes a 16bit pointer). >> If you need more than 200 and odd items and they are static, I see >> one simple workaround... >> >> * Cut your menu tree at some level (say level 2). Then you will have >> a mainmenu.com which displays the first two levels of the menu, and >> for each possible level 2 choice, you will have a different menu.com >> file. Choosing a item from the mainmenu, will call one of the helper >> menu.com. >> > > Well, the simplest way to undo that limitation would be to switch to > com32. > > -hpa > The only disadvantage of com32 I see now, is debugging (burn -> test -> reburn ....). Using a CDRW reduces the cost, but the time is still a factor. If I am right, the com32 version will have the same menu.* and user files. The only files which needs to change is heap.*, biosio.*, startup code and some others. Can we setup the code such that, users can compile it to an EXE or COM32 by just changing a #define? If so then debugging will also not be a problem. - Murali From rvallejo at novadevices.com Wed Apr 28 13:20:21 2004 From: rvallejo at novadevices.com (Rafael Vallejo) Date: Wed, 28 Apr 2004 15:20:21 -0500 Subject: [syslinux] SYSLINUX.CFG not readed Message-ID: <000b01c42d5e$3b546960$2201a8c0@pcrafael> Hello list. I have configured a Linux box for booting from a Compact Flash instead of a Hard Disk, this is an IDE CF, Syslinux is on the MBR because it loads, when I put the the kernel file named as linux as it expects i load it but when I rename it expecting it to load the SYSLINUX.CFG it does not load the menu as I would like, I will put two images on that Compact flash in order to boot from two diferent configurations Here is a listing of my Compact Flash F:\>dir Volume in drive F is CE Volume Serial Number is 3565-18E8 Directory of F:\ 05/31/1994 05:22 AM 54,645 COMMAND.COM 09/30/1993 04:20 AM 29,136 HIMEM.SYS 03/22/2004 01:02 PM 102 AUTOEXCE.BAT 04/23/2004 01:48 PM 21,374,396 KYPUS.IMG 04/28/2004 10:43 AM otros 04/14/2004 07:09 AM 32,225 loadlin.exe 04/27/2004 10:48 AM 124 AUTOEXLI.bat 02/22/2004 09:31 PM 9,428 isolinux.bin 03/22/2004 10:46 AM 94 CONFIG.SYS 03/01/2003 05:11 PM 97,219 EBOOT.BIN 10/22/2002 03:27 PM 98,076 LOADCEPC.EXE 12/11/2003 01:58 PM 17,715,335 nk.bin 03/06/2003 09:16 PM 24,347 SBOOT.BIN 03/22/2004 12:56 PM Settings 04/28/2004 02:27 PM 8,184 LDLINUX.SYS 04/27/2004 10:21 AM 720,425 vmlinuz 04/28/2004 03:06 PM 209 isolinux.cfg 15 File(s) 40,163,945 bytes 2 Dir(s) 82,464,768 bytes free F:\> Here I have the content of my SYSLINUX.CFG file DEFAULT KypusLX PROMPT 1 TIMEOUT 600 LABEL KypusLX KERNEL vmlinuz APPEND ramdisk_size=73000 ro initrd=KYPUS.IMG root=/dev/ram LABEL KypusCE KERNEL vmlinux APPEND ramdisk_size=73000 rw initrd=KYPUS.IMG root=/dev/ram Any Idea? Regards Rafael From B.Blaauw at student.fontys.nl Wed Apr 28 15:27:44 2004 From: B.Blaauw at student.fontys.nl (Blaauw,Bernd B.) Date: Thu, 29 Apr 2004 00:27:44 +0200 Subject: [syslinux] SYSLINUX.CFG not readed Message-ID: <11584391EF34234392426DA50F9E1F9387A2C4@selenga.fontys.nl> I don't see syslinux.cfg file in your directory listing. does Syslinux even load? Win9x/DOS: SYSLINUX.COM F: NT/2000/XP: SYSLINUX.EXE F: I have *no* idea why you have isolinux.bin/cfg in the directory listing. Syslinux requires all files it can load to be in root. Don't know if MEMDISK also requires this. Bernd Blaauw 05/31/1994 05:22 AM 54,645 COMMAND.COM 09/30/1993 04:20 AM 29,136 HIMEM.SYS 03/22/2004 01:02 PM 102 AUTOEXCE.BAT 04/23/2004 01:48 PM 21,374,396 KYPUS.IMG 04/28/2004 10:43 AM otros 04/14/2004 07:09 AM 32,225 loadlin.exe 04/27/2004 10:48 AM 124 AUTOEXLI.bat 02/22/2004 09:31 PM 9,428 isolinux.bin 03/22/2004 10:46 AM 94 CONFIG.SYS 03/01/2003 05:11 PM 97,219 EBOOT.BIN 10/22/2002 03:27 PM 98,076 LOADCEPC.EXE 12/11/2003 01:58 PM 17,715,335 nk.bin 03/06/2003 09:16 PM 24,347 SBOOT.BIN 03/22/2004 12:56 PM Settings 04/28/2004 02:27 PM 8,184 LDLINUX.SYS 04/27/2004 10:21 AM 720,425 vmlinuz 04/28/2004 03:06 PM 209 isolinux.cfg 15 File(s) 40,163,945 bytes 2 Dir(s) 82,464,768 bytes free From gmurali at cs.uchicago.edu Wed Apr 28 15:44:19 2004 From: gmurali at cs.uchicago.edu (ganapathy murali krishnan) Date: Wed, 28 Apr 2004 17:44:19 -0500 Subject: [syslinux] Re: Menu system limitations In-Reply-To: <64784.209.128.64.238.1083185266.squirrel@www.zytor.com> References: <408FD296.6020503@cs.uchicago.edu> <408FF5F3.8020807@zytor.com> <40900481.6040604@cs> <64784.209.128.64.238.1083185266.squirrel@www.zytor.com> Message-ID: <409033C3.6050208@cs> hpa at zytor.com wrote: >>The only disadvantage of com32 I see now, is debugging (burn -> test -> >>reburn ....). Using a CDRW reduces the cost, but the time is still a >>factor. >> >> >> > >Why would you need to burn/reburn? Use SYSLINUX+floppy or >PXELINUX+network and you don't need that problem. > >You can also test it under e.g. Bochs. > > > >>If I am right, the com32 version will have the same menu.* and user >>files. The only files which needs to change is heap.*, biosio.*, startup >>code and some others. Can we setup the code such that, users can compile >>it to an EXE or COM32 by just changing a #define? If so then debugging >>will also not be a problem. >> >> > >COM or COM32 would be more realistic. > > -hpa > > > The only reason I said EXE or COM32, is incase we cross the 64KB limit. I never thought of SYSLINUX+floppy (need to see if I have a floppy drive/disk somewhere :-)). - Murali From hpa at zytor.com Wed Apr 28 22:56:10 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Wed, 28 Apr 2004 22:56:10 -0700 Subject: [syslinux] Re: Menu system limitations In-Reply-To: <409033C3.6050208@cs> References: <408FD296.6020503@cs.uchicago.edu> <408FF5F3.8020807@zytor.com> <40900481.6040604@cs> <64784.209.128.64.238.1083185266.squirrel@www.zytor.com> <409033C3.6050208@cs> Message-ID: <409098FA.6050209@zytor.com> ganapathy murali krishnan wrote: > > The only reason I said EXE or COM32, is incase we cross the 64KB limit. > I never thought > of SYSLINUX+floppy (need to see if I have a floppy drive/disk somewhere > :-)). > You'd still have to deal with 16-bit code if you mean DOS EXE, and the gcc toolchain can't do it. The mingw toolchain can be used to build Win32 EXE binaries, but that requires a very different API (it would be easier, and more appropriate, to make the test binaries be Linux binaries.) -hpa From haisenko at be-ok.com Thu Apr 29 01:28:28 2004 From: haisenko at be-ok.com (Marc Haisenko) Date: Thu, 29 Apr 2004 09:28:28 +0100 Subject: [syslinux] Re: Menu system limitations In-Reply-To: <409004BF.8030905@cs> References: <408FF5F3.8020807@zytor.com> <409004BF.8030905@cs> Message-ID: <200404291028.28497.haisenko@be-ok.com> On Wednesday 28 April 2004 21:23, ganapathy murali krishnan wrote: > The only disadvantage of com32 I see now, is debugging (burn -> test -> > reburn ....). Using a CDRW reduces the cost, but the time is still a > factor. Uhm, I was debugging with Bochs and VMWare... mounted a disk image, copied the .com (or in your case .c32) to the disk image, ran sync and restarted the emulator. But I was only doing the assembler equivalent of printf-debugging ;-) If it's "just" the menuing system you don't need a real PC, do you ? C'ya, Marc -- Marc Haisenko Linux Solutions Be O.K. service group GmbH R?desheimer Stra?e 7 D-80686 M?nchen Tel: +49 (0)89 - 54 84 99 73 Fax: +49 (0)89 - 54 84 99 28 e-mail: haisenko at be-ok.com http://www.be-ok.com From lode_leroy at hotmail.com Thu Apr 29 07:29:07 2004 From: lode_leroy at hotmail.com (lode leroy) Date: Thu, 29 Apr 2004 16:29:07 +0200 Subject: [syslinux] Pxelinux/memdisk with Universal TCP/IP Network Bootdisk floppy Message-ID: Just for the record, I want to let the list know that I succeeded in booting a modified Universal TCP/IP Network Bootdisk floppy from http://members.iinet.net.au/~bdriver/bootdisk/ I started with a Win98 Bootfloppy from http://www.tobs.uku.co.uk. Then I made a disk image from this, using dd.exe from cygwin. Then I converted this disk image from 1.44MB to 2.88MB with winimage. Then I copied the files from the Universal TCP/IP Network bootdisk to the image, by mounting the file on a linux computer (mount -t vfat -o loop Win98tcp.img Win98tcp). I added mem.exe to the files.cab, using EXTRACT.EXE and CABARC.EXE from http://support.microsoft.com/default.aspx?scid=KB;en-us;310618 I also added ntfsdos.exe from http://www.sysinternals.com/files/ntfs30r.zip. I added emm386.exe to the floppy image, and added the following to CONFIG.SYS device=emm386.exe /noems I use pxegrub from grub 0.94 on my tftp server, I have the following in my GRUB menu: title Win98 Network Bootdisk root (nd) kernel /memdisk initrd /Win98tcp.img The memdisk is from syslinux-2.08 (http://www.kernel.org/pub/linux/utils/boot/syslinux/syslinux-2.08.tar.gz) After booting I'm left with about 350 KB free memory... _________________________________________________________________ Vraag van de week: Welk soort project zou jij financieel ondersteunen? http://www.msn.be/microsoft/potential/default.asp From gebhardt at HRZ.Uni-Marburg.DE Thu Apr 29 07:46:29 2004 From: gebhardt at HRZ.Uni-Marburg.DE (Gebhardt Thomas) Date: Thu, 29 Apr 2004 16:46:29 +0200 Subject: [syslinux] SYSLINUX 2.09, 2.10-pre1 released In-Reply-To: <408E045C.1080404@zytor.com> References: <408E045C.1080404@zytor.com> Message-ID: <200404291646.30544.gebhardt@hrz.uni-marburg.de> Hi, I'd really appreciate if there were a PXELINUX option that would prevent users from adding kernel commandline boot parameters apart from the options nailed down in the configuration file. This is a very basic security issue in an unattended, potentially hostile environment if you don't want user to become root (init=/bin/sh), a situation not that uncommon. I hope that such a configuration flag is not that complicated to implement, since it is not a really new feature, but just disables an already functional feature. Thanks, Thomas From gmurali at cs.uchicago.edu Thu Apr 29 08:20:29 2004 From: gmurali at cs.uchicago.edu (Murali Krishnan Ganapathy) Date: Thu, 29 Apr 2004 10:20:29 -0500 Subject: [syslinux] SYSLINUX 2.09, 2.10-pre1 released In-Reply-To: <200404291646.30544.gebhardt@hrz.uni-marburg.de> References: <408E045C.1080404@zytor.com> <200404291646.30544.gebhardt@hrz.uni-marburg.de> Message-ID: <40911D3D.4080005@cs.uchicago.edu> Till such a feature comes along, you can implement the same functionality as follows: * Create a menu, with options corresponding to the different isolinux labels * Set pcelinux.cfg to run the menu directly without any wait. This way you get a slightly user friendly interface and the users dont have access to the pxelinux boot prompt. - Murali Gebhardt Thomas wrote: >Hi, > >I'd really appreciate if there were a PXELINUX option that would prevent >users from adding kernel commandline boot parameters apart from the >options nailed down in the configuration file. This is a very basic security >issue in an unattended, potentially hostile environment if you don't want >user to become root (init=/bin/sh), a situation not that uncommon. > >I hope that such a configuration flag is not that complicated to implement, >since it is not a really new feature, but just disables an already functional >feature. > >Thanks, Thomas > >_______________________________________________ >SYSLINUX mailing list >Submissions to SYSLINUX at zytor.com >Unsubscribe or set options at: >http://www.zytor.com/mailman/listinfo/syslinux >Please do not send private replies to mailing list traffic. > > > > From hpa at zytor.com Thu Apr 29 10:56:47 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Thu, 29 Apr 2004 10:56:47 -0700 Subject: [syslinux] SYSLINUX 2.09, 2.10-pre1 released In-Reply-To: <200404291646.30544.gebhardt@hrz.uni-marburg.de> References: <408E045C.1080404@zytor.com> <200404291646.30544.gebhardt@hrz.uni-marburg.de> Message-ID: <409141DF.4070606@zytor.com> Gebhardt Thomas wrote: > Hi, > > I'd really appreciate if there were a PXELINUX option that would prevent > users from adding kernel commandline boot parameters apart from the > options nailed down in the configuration file. This is a very basic security > issue in an unattended, potentially hostile environment if you don't want > user to become root (init=/bin/sh), a situation not that uncommon. > > I hope that such a configuration flag is not that complicated to implement, > since it is not a really new feature, but just disables an already functional > feature. > It's a new feature, and it is unfortunately reasonably complex to implement. What makes me really question the value is that it's not clear to me that there aren't other security holes in the whole scenario. -hpa From bernardinekettering at barlick.net Thu Apr 29 18:42:44 2004 From: bernardinekettering at barlick.net (julius kuntzman) Date: Fri, 30 Apr 2004 09:42:44 +0800 Subject: [syslinux] Ldqway Our team is ready Message-ID: <23BBCAE4.30D8273@barlick.net> agroveterinario bittemple fdts We are your your convenient, safe and private online source for FDA a`p`p`roved pharmacy prescriptions. V*ico*dinPain Killers We accept almost every form of payment. Q D http://thiscansolvetheprb.biz/aps/ No thanks: http://thiscansolvetheprb.biz/wdj/rm.html Young Hopeful: "Father, what is a traitor in politics?Father (a veteran politician): "A traitor is a man who leaves our party and goes over to the other one."Young Hopeful: "Well then, what is a man who leaves his party and comes over to yours?"Father: "A converter, my son." Al Smith was governor of New York State. He was a famouse man. he was born very poor on the East Side of New York City. he had little education. He worked very hard and won great success.One day, as governor, he was visiting the state prison at Sing Sing. Sing Sing is one of the largest prisons in the United States. The head of the prison asked Mr. Smith to say something to the prisoners. Mr. Smith had never spoken to this kind of audience before. He did not know how to begin.Finally, he said, "My fellow citizens..." Then he remembered that when a man goes to prison he is no longer a citizen. he began again. "My fellow prisoners..." That did not sound right, so he said: "Well, anyway, I'm glad to see so many of you here today." indent1meiwakuc02jougakuk,hojyokin jisen. From paul.rogers at juno.com Thu Apr 29 22:37:57 2004 From: paul.rogers at juno.com (Paul G Rogers) Date: Thu, 29 Apr 2004 22:37:57 PDT Subject: [syslinux] Syslinux 1.75, LEAF Bering 1.2, Compaq 2266 Message-ID: <20040429.223801.6767.0.paul.rogers@juno.com> I have been using the LEAF Bering firewall for a year or so. It boots with Syslinux 1.75. But Bering is too large for a 1440KB floppy, so it formats the diskette as 1680KB. Recently I found a small Compaq 2266 box to replace a larger Compaq 7170 to run the firewall. But when I try to boot the Bering diskette on the 2266 I get the following message: Loading Linux ............ Boot failed: please change disks and press a key to continue. I have determined that the problem is an interaction between this Phoenix BIOS, Syslinux, and the 1680KB format. Syslinux, the 1680KB diskette, with this floppy drive will boot the Bering Linux kernel on a different motherboard. This system will boot the Bering Linux kernel on a 1440KB diskette, without all the necessary Bering packages of course. I tried running "syslinux -s" from my W95B box on the 1680KB diskette, although it didn't work for me after "lock a:", but did after "unlock". The date of the ldlinux file had been updated. The "buggy BIOS" version of Syslinux gave me the same failure. The chipset is a SiS5598, Cyrix MII CPU. Apparently the BIOS is unconvinced 1680KB is a legitimate format. Compaq/HP is spectacularly unforthcoming about anything more than the machine as delivered in it's intended configuration for it's supplied software. No BIOS upgrades or the like can be expected. Can Syslinux be configured to avoid using the BIOS? Paul Rogers (paul.rogers at juno.com) http://www.xprt.net/~pgrogers/ http://www.angelfire.com/or/paulrogers/ Rogers' Second Law: "Everything you do communicates." (I do not personally endorse any additions after this line. TANSTAAFL :-) ________________________________________________________________ The best thing to hit the Internet in years - Juno SpeedBand! Surf the Web up to FIVE TIMES FASTER! Only $14.95/ month - visit www.juno.com to sign up today! From hpa at zytor.com Thu Apr 29 23:32:48 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Thu, 29 Apr 2004 23:32:48 -0700 Subject: [syslinux] Syslinux 1.75, LEAF Bering 1.2, Compaq 2266 In-Reply-To: <20040429.223801.6767.0.paul.rogers@juno.com> References: <20040429.223801.6767.0.paul.rogers@juno.com> Message-ID: <4091F310.5000301@zytor.com> Paul G Rogers wrote: > I have been using the LEAF Bering firewall for a year or so. It boots > with Syslinux 1.75. But Bering is too large for a 1440KB floppy, so it > formats the diskette as 1680KB. Recently I found a small Compaq 2266 box > to replace a larger Compaq 7170 to run the firewall. But when I try to > boot the Bering diskette on the 2266 I get the following message: > Loading Linux ............ > Boot failed: please change disks and press a key to continue. > Can you try 2.09 instead? -hpa From hpa at zytor.com Thu Apr 29 23:42:27 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Thu, 29 Apr 2004 23:42:27 -0700 Subject: [syslinux] Syslinux 1.75, LEAF Bering 1.2, Compaq 2266 In-Reply-To: <20040429.223801.6767.0.paul.rogers@juno.com> References: <20040429.223801.6767.0.paul.rogers@juno.com> Message-ID: <4091F553.1040305@zytor.com> Paul G Rogers wrote: > > Can Syslinux be configured to avoid using the BIOS? > No way in hell... -hpa From jsiemes at web.de Fri Apr 30 00:47:27 2004 From: jsiemes at web.de (Josef Siemes) Date: Fri, 30 Apr 2004 09:47:27 +0200 Subject: [syslinux] SYSLINUX 2.09, 2.10-pre1 released Message-ID: <1039279290@web.de> Hi, "H. Peter Anvin" schrieb am 29.04.04 20:12:16: > Gebhardt Thomas wrote: > > I'd really appreciate if there were a PXELINUX option that would prevent > > users from adding kernel commandline boot parameters apart from the > > options nailed down in the configuration file. This is a very basic security > > issue in an unattended, potentially hostile environment if you don't want > > user to become root (init=/bin/sh), a situation not that uncommon. > > > > I hope that such a configuration flag is not that complicated to implement, > > since it is not a really new feature, but just disables an already functional > > feature. > > > > It's a new feature, and it is unfortunately reasonably complex to > implement. What makes me really question the value is that it's not > clear to me that there aren't other security holes in the whole scenario. I've also asked for this a while ago, so this discussion isn't new. For pxelinux it would at least allow to lock the ordinary User out of kernel commandline parameters. There are much hints how to get a root shell by adding 'init=/bin/sh', so actually the user gets root access without much effort. If pxelinux locks the command line it would be somewhat harder: Either the user prevents pxelinux to read the config file at all (perhaps by pulling out the network cable at the right moment?) or needs to do some bios hacking. At least there wouldn't be the 'obvious' way. For iso/syslinux maybe this could also lock the user to a fixed configuration. This couldn't be fixed easy, since writing the config file could be a really hard time. But it could have its use too. For implementation: Isn't there some point in the code where the commandline for the kernel is concatenated? So the options from the config file are taken, then the options following the label on the command line are added? Why not just skip adding the options following the label? I've tried to find the right place for this in the code, but seems I've not enough assembler skills for this ... Regards, Josef ____________________________________________________________________ Der WEB.DE Virenschutz schuetzt Ihr Postfach vor dem Wurm Sober.A-F! Kostenfrei fuer FreeMail Nutzer. http://f.web.de/?mc=021158 From haisenko at be-ok.com Fri Apr 30 01:24:48 2004 From: haisenko at be-ok.com (Marc Haisenko) Date: Fri, 30 Apr 2004 09:24:48 +0100 Subject: [syslinux] SYSLINUX 2.09, 2.10-pre1 released In-Reply-To: <409141DF.4070606@zytor.com> References: <408E045C.1080404@zytor.com> <200404291646.30544.gebhardt@hrz.uni-marburg.de> <409141DF.4070606@zytor.com> Message-ID: <200404301024.48443.haisenko@be-ok.com> On Thursday 29 April 2004 19:56, H. Peter Anvin wrote: > Gebhardt Thomas wrote: > > Hi, > > > > I'd really appreciate if there were a PXELINUX option that would prevent > > users from adding kernel commandline boot parameters apart from the > > options nailed down in the configuration file. This is a very basic > > security issue in an unattended, potentially hostile environment if you > > don't want user to become root (init=/bin/sh), a situation not that > > uncommon. > > > > I hope that such a configuration flag is not that complicated to > > implement, since it is not a really new feature, but just disables an > > already functional feature. > > It's a new feature, and it is unfortunately reasonably complex to > implement. What makes me really question the value is that it's not > clear to me that there aren't other security holes in the whole scenario. > > -hpa Well, after a very brief look at the comments it seems to me that in runkernel.inc the following happens: first the append=... options are added to the kernel options and then the options provided by the command line. Wouldn't it be easy to implement a configuration option that would skip construct_commandline and jump to commandline_end ? That would do just what all these people ask for (that request isn't showing up for the first time). Or did you think of another (better, cleaner ?) way to implement such a "skip commandline options" feature ? C'ya, Marc -- Marc Haisenko Linux Solutions Be O.K. service group GmbH R?desheimer Stra?e 7 D-80686 M?nchen Tel: +49 (0)89 - 54 84 99 73 Fax: +49 (0)89 - 54 84 99 28 e-mail: haisenko at be-ok.com http://www.be-ok.com From gmurali at cs.uchicago.edu Fri Apr 30 06:31:27 2004 From: gmurali at cs.uchicago.edu (Murali Krishnan Ganapathy) Date: Fri, 30 Apr 2004 15:31:27 +0200 Subject: [syslinux] SYSLINUX 2.09, 2.10-pre1 released Message-ID: <200404301531.27979.gmurali@cs.uchicago.edu> Hi, thank you very much for all your answers! Murali Krishnan Ganapathy wrote: > * Create a menu, with options corresponding to the different isolinux > labels > * Set pcelinux.cfg to run the menu directly without any wait. If I can feel sure that there is no escape from this menu then this is my way to go! "H. Peter Anvin" wrote: > It's a new feature, and it is unfortunately reasonably complex to > implement. What makes me really question the value is that it's not > clear to me that there aren't other security holes in the whole scenario. well, you sure have a better insight into that matter than I. "Josef Siemes" wrote: > For implementation: Isn't there some point in the code where the commandline > for the kernel is concatenated? So the options from the config file are > taken, then the options following the label on the command line are added? > Why not just skip adding the options following the label? > I've tried to find the right place for this in the code, but seems I've not > enough assembler skills for this . Marc Haisenko wrote: > Wouldn't it be easy to implement a configuration option that would skip > construct_commandline and jump to commandline_end ? That would do just what > all these people ask for (that request isn't showing up for the first time). > > Or did you think of another (better, cleaner ?) way to implement such > a "skip commandline options" feature ? Yes, I thought of something like that, but since my assembler skills are also very limited, I can't get the overall view to be sure that this is sufficient. Thanks, Thomas From huberstuff at hsn.at Fri Apr 30 10:27:59 2004 From: huberstuff at hsn.at (huberstuff at hsn.at) Date: Fri, 30 Apr 2004 19:27:59 +0200 (CEST) Subject: [syslinux] Novell DHCP and Pxelinux Message-ID: <3069.57.90.33.70.1083346079.squirrel@webmail.hsn.at> Hello Friends, once more. We got pxelinux run with win 2000 dhcp. It works really fine. there a not problems at all. Now we want to get it run also under novell, but we want to use the novell dhcp, we don't need the tftpd from novell, we want to use a tfptd on another machine (a simple micrsoft tfptd). As we found out, novell does not support the dhcp options (67 = bootserver, 68 = bootfile), But novell has a build in functionality for booting from images. So we activated this option (on the dhcp) for our test subnet and specified the bootserver ip and name the the bootfile (from our ms-tfpdd) We have also the option 60 put to PXEClient on the dhcp. So now, when we start a client, it cries out the error message ProxyDHCP service did not reply to request on port 4011 PXE-E55??? We don't want to use proxydhcp, we want to make it with a simple microsoft tftpd that is running on our install-server-black-box! please help me! thanks markus From rvallejo at novadevices.com Wed Apr 28 09:21:31 2004 From: rvallejo at novadevices.com (Rafael Vallejo) Date: Wed, 28 Apr 2004 11:21:31 -0500 Subject: [syslinux] Compact Flash Could not find kernel image Message-ID: <000c01c42d3c$ddb65be0$2201a8c0@pcrafael> Hello List Im trying to use syslinux on a compact flash card in order to boot Linux and DOS from the same CF, I installed syslinux on it by booting from the flash that already have DOS runing and executing the command syslinux -s c:, it sucesfully installed the MBR and ldlinux.sys, there is a syslinux.cfg file on the flash and the proper vmlinuz and linux.gz Initrd. When I start the PC I get the following message: SYSLINUX 2.09 2004-04-26 Copyright (C) 1994-2004 H. Peter Anvin Could not find kernel image: linux boot: Any ideas? Rafael From B.Blaauw at student.fontys.nl Fri Apr 30 13:33:14 2004 From: B.Blaauw at student.fontys.nl (Blaauw,Bernd B.) Date: Fri, 30 Apr 2004 22:33:14 +0200 Subject: [syslinux] Compact Flash Could not find kernel image Message-ID: <11584391EF34234392426DA50F9E1F9387A2C5@selenga.fontys.nl> next logical step is to post the contents of the syslinux.cfg file.. Bernd From hpa at zytor.com Fri Apr 30 14:30:11 2004 From: hpa at zytor.com (H. Peter Anvin) Date: Fri, 30 Apr 2004 14:30:11 -0700 Subject: [syslinux] Novell DHCP and Pxelinux In-Reply-To: <3069.57.90.33.70.1083346079.squirrel@webmail.hsn.at> References: <3069.57.90.33.70.1083346079.squirrel@webmail.hsn.at> Message-ID: <4092C563.3080101@zytor.com> huberstuff at hsn.at wrote: > > We have also the option 60 put to PXEClient on the dhcp. > Don't. It's bad chicken unless you actually implement all the PXE ugliness. -hpa