[00:28] ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds [06:40] alexanderwz (~alexander@karuna.med.harvard.edu) left irc: Read error: Operation timed out [06:40] alexanderwz (~alexander@karuna.med.harvard.edu) joined #fai. [07:07] fai-guy (~fai-guy@p57B9BA53.dip0.t-ipconnect.de) joined #fai. [07:29] ErKa (keryell@keryell.pck.nerim.net) joined #fai. [08:03] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [08:28] Mrfai (~lange@kueppers.informatik.uni-koeln.de) joined #fai. [08:29] mat1 (~math@p54B48FB3.dip0.t-ipconnect.de) joined #fai. [08:29] how to solve: ERROR: live-initramfs was not installed inside the nfsroot.? [08:29] add live-initramfs in NFSROOT [08:30] and check your sources.list. [08:30] thanks [08:30] live-initramfs is not needed in NFSROOT if you are using recent fai versions [08:31] because fai-nfsroot depends on live-initramfs [08:31] if upgraded yesterday to 3.2.4 [08:31] aptitude install fai-quickstarts doesn't depend on fai-nfsroot, or am I wrong? [08:33] that's ok, because fai-nfsroot must not be installed on a normal system. It's only needed inside the nfsroot when you call fai-setup or make-fai-nfsroot [08:34] but if you install fai-quickstart, you will probably run fai-setup. [08:34] yes [08:35] check /etc/fai/apt/sources.list. Add the koeln repository which is mentioned on the FAI web page [08:35] ahh.. *klick* I didn't want to run "aptitude install fai-nfsroot" [08:36] on my server [08:36] ok [08:38] yep [08:44] when running fai-setup again it warns me about untrusted/unverified packages: *** WARNING *** Ignoring these trust violations because... Is this correct? ATM I don't know how to supply a key for the koeln repo serving 3.2.4 fai [08:44] ... [08:56] Nick change: MT -> Guest696 [08:56] Guest696 (~MT@dove.informatik.tu-muenchen.de) left irc: Read error: Connection reset by peer [08:56] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [08:58] you can ignore this [09:25] with the new fai version 3.2.4 there is the new /etc/fai/NFSROOT, and the fai-kernel package is deprecated, but my fai-client now complains about missing modules... the exact error message is not available, cause DHCP info messages spamming the screen, and shell access is not available at this stage... any ideas? [09:33] maybe ther kernel you boot does not match to the lib/modules inside the nfsroot [09:50] when choosing a new kernel in pxe-linux.cfg/XYZ have I to rerun the fai-setup script? [09:55] yes it was probably a initrd-kernel mismatch [09:56] but now BusyBox complains: /bin/sh can't access tty; job control turned off [10:04] live.log gives: mount call failed 13 [10:06] it sis my NFS export [10:06] mat1: any NFS related msgs in daemon.log on the nfs-server? [10:15] yes, bu I don't understand: refused mount request from frontend.service for /srv/fai/nfsroot,v3,tcp,rsize=32768,wsize=32768 (/): not exported [10:15] but exportfs -av says: [10:15] exporting 134.104.121.0/24:/srv/fai/nfsroot [10:16] is he trying to mount "/srv/fai/nfsroot,v3,tcp,rsize=32768,wsize=32768" instead of just "/srv/fai/nfsroot"?? [10:17] in other words, does he get the options right? [10:18] can you da a 'cat /proc/cmdline' on the box? [10:20] mount error 13 means permission denied [10:21] mhmm cat proc/cmdline is the same as the appendline in the pxe conf [10:23] what's in it? [10:33] vmlinuz-2.6.18-5-amd64 initrd=initrd.img-2.6.18-5-amd64 ip=dhcp FAI_ACTION=install root=/dev/nfs nfsroot=X.Y.Z.X:/srv/fai/nfsroot,v3,tcp,rsize=32768,wsize=32768 boot=live FAI_FLAGS=verbose,sshd,createvt [10:35] when in initramfs => mount command: "mount -t nfs X.Y.Z.X:/srv/fai/nfsroot /mnt" failed and says invalid argument *wondering*, but the nfsserver is ping-able [10:35] sorry for the delay in my answers [10:35] mat1: you NFS server support NFS v3? Or only v2? check the error messages on your NFS server [10:36] I assume he supports v3 and v4 [10:37] nfsstat, produces such output where I assume this from [10:38] nfs-kernel-server from etch [10:39] my exports looks like this: /srv/fai/nfsroot X.Y.Z.0/255.255.255.0(async,ro,no_subtree_check,no_root_squash) [10:39] Insomnia (~insomnia@exit.credativ.com) joined #fai. [10:40] check your daemon.log on your NFS server. [10:40] mountd[2766]: refused mount request from frontend for /srv/fai/nfsroot,v3,tcp,rsize=32768,wsize=32768 (/): not exported [10:41] can't interpret this.. :-/ [10:43] mat1: it says that your client tried to mount /srv/fai/nfsroot,v3,tcp,rsize=32768,wsize=32768 on / [10:43] but it should ask for /srv/fai/nfsroot [10:43] seems hat it interprets ,v3,tcp.... as part of the path [10:43] ahhh [10:43] you mount options are wrong [10:44] how do I do them right? [10:44] mat1: use fai-chboot for creating your pxelinux.cfg files [10:47] I've here another tool I must use for PXE access, so I run allways fai-chboot and take those arguments in the other tool [10:47] but the fai-chboot just says: nfsroot=/srv/fai/nfsroot [10:47] how does it know from which server? [10:48] and with which options? [10:49] the server IP address can be defined in dhcpd.conf. There's an option root-server [10:50] a of course [10:50] But it may also work with adding the IP address. But you have to remove those v3,tcp,.. stuff [10:51] I think there could be something wrong in the pxe config file [11:01] thanks without the v3,... options suffix it works [11:04] oz_: hmm, maybe you are right.. but with many other installers the config work.. thank you for the hint [12:02] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Remote host closed the connection [12:03] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [12:07] cVsup (~cVsup@janeway2.mav.com.br) joined #fai. [12:09] i use fai script setup_harddisk [12:09] but it not find disk1 [12:09] somebody can help me? [12:09] Probing disks: [12:09] Disks found: hda [12:09] Using config file: disk_config [12:09] Mapping disk name disk1 to [12:09] ERROR: could not read device /dev/ [12:31] ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds [12:31] somebody alive? [12:54] Insomnia (insomnia@exit.credativ.com) left #fai. [12:59] ErKa (keryell@m2.wifi.enstb.org) joined #fai. [12:59] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Read error: Connection reset by peer [12:59] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [13:03] cVsup: yep. [13:04] Mrfai, you can help me with setup_harddisks script? [13:04] Do you have an entry for hda in /dev? [13:04] yes [13:04] is udev running? [13:04] hm... whait [13:05] Mrfai, but setup_harddisks not use udev [13:06] cVsup: did you login to the install client console and check if the disk is really there? [13:06] e.g. by trying to look at it with cfdisk? [13:07] lazyb0y_, no [13:07] i run setup_harddisks in other system [13:07] cVsup: what do mean you run it in other system? [13:08] you are doing a fai install, aren't you? [13:08] i am use gentoo linux [13:08] and tryning use for it [13:08] wow, interesting, but can you then elaborate a bit how you use fai with gentoo? [13:09] or how you use setup harddisks exactly [13:09] I think this is very important for the debugging - neither fai nor setup harddisk is very much tested with gentoo, if at all [13:10] is your nfsroot gentoo, also? [13:10] and, by the way, if you are doing experimental stuf anyway, you might want to consider using the new setup storage, instead [13:11] lazyb0y_, the big problem is create and formating partitions [13:11] setup_harddisk found devices [13:11] cVsup: sure, that's what setup harddisk is used for [13:12] but it not associated disk1 with first device [13:12] still, what exactly is the environment you run setzup harddisk in [13:12] it's only tested to run in a debian based nfsroot [13:13] hm... [13:13] there can be many reasons - one being that kernels used in gentoo have disk information on a different place than kernels on debian etch, with which setup harddisk is tested [13:13] i use gentoo stage3 [13:13] yes [13:13] but look the error [13:14] setup_harddisks version 0.43fai [13:14] Probing disks: [13:14] Disks found: hda [13:14] Using config file: disk_config [13:14] Mapping disk name disk1 to [13:14] ERROR: could not read device /dev/ [13:14] let me give you some examples what kind of indo I am askingyou fro: are you doing a network install? if so, with a gentoo based nfsroot? [13:14] it found disks [13:14] can you run is with debugging options enabled? [13:14] no [13:15] do you run it by hand in a running system? [13:15] what parameters [13:15] yes [13:15] i run in chroot [13:16] osp [13:16] ops [13:16] and, really, in any way: doin these experimental stuff is really encouraged to do with the new setup storage - as setup_garddisk will not get major fixes, for example for such use cases as you have now [13:16] i run in complete gentoo system [13:16] does the chroot have things in /dev at all? [13:17] yes [13:17] doe scfdsik work in that chroot and find stuff? [13:17] sorry for so many typos... [13:17] i mean - can you do stuff with cfdisk in this chroot [13:19] lazyb0y_, but [13:19] setup_harddisk work with scfdisk? [13:21] setup harddisk works with sfdisk [13:22] i just say, for debugging purposes it's sometoime helpful to look at the disk with cfdisk of fdisk, to see if the disk is accessibe easily in any way [13:22] setup_storage will work with something else [13:22] and is much more sophiticated. [13:23] I really suggest you try with this: http://faiwiki.informatik.uni-koeln.de/index.php/Setup-storage [13:23] and then work on getting this running nicely in your setup [13:28] cVsup: what's the output of ls -l /dev/hda* [13:30] GRUB fails on reboot in stage 1.5, Error 15 (file not found), but if I ssh into live-install-system /target looks ok. And every path exists on /target.. The node installing grub is also in the GRUB class... [13:30] brw-r----- 1 root disk 3, 0 Dez 18 10:28 /dev/hda [13:30] brw-r----- 1 root disk 3, 1 Dez 18 12:28 /dev/hda1 [13:30] brw-r----- 1 root disk 3, 2 Dez 18 12:28 /dev/hda2 [13:30] brw-r----- 1 root disk 3, 3 Dez 18 10:28 /dev/hda3 [13:32] This is bad: Mapping disk name disk1 to [13:32] it should say mapping... disk1 to hda [13:33] What the output of cat /proc/partitions? [13:36] do you have the variable disklist defined in /tmp/fai/variables.log. If not setup_harddisks can't work. [13:36] But you can just define disklist=hda as an ugly workaround [13:37] Mrfai, i not have file variables.log [13:37] what how make it? [13:42] if you do not have varaibles.log then you do not run fai the usual way I guess. Use my workaround instead. [13:43] Mrfai, but i need auto detect hds [13:43] hard disks [13:44] then you need to start FAI in the normal way. I did not understand how you are booting your install clients. Please read the FAI guide. [13:46] Mrfai: as you can see above, he runs setup harddisk completely without fai, on gentoo, manually - a very unusual way [13:46] I think it's an intersting way to use parts of FAI, but as said, I'd rather recommend that he teste setup_storage, as setup_harddisks will not be supported to be run this way [13:47] for setup_storage, I can imagine we are interested to support it's use in such a way - to run it manually, as well as integrate it into other tools, xen-tools, for example [13:49] also the new tool needs this variable: defined( $ENV{disklist} ) or die "Environment variable disklist is not set"; [13:51] sure, also the new tool will not work by itself in this environment and may need some things properly setup, still, I think it's more or less wasted time to try with the old tool and fix things so the old tool works [13:51] cVsup: have a look at set_disk_info() in lib/subroutines-linux. There the variable disklist is set [13:52] why is in the default /etc/fai/NFSROOT the grub package under AMD64, when it should be only installed when selecting the GRUB class?? [13:52] as in package_config/DEFAULT [13:55] mat1: hmm, I think, you cannot install a kernel (at least modules atre required in nfsroot) without grub (not sure, a guess) [13:56] /etc/fai/NFSROOT is only used for building the nfsroot. Those packages will not be installed on your system [13:56] and, the stuff in NFSROOT does not land in the installed client - AFAIK the base,tgz is created before these things are installed into the nfsroot [14:09] blblack (~brandon@wasabi.dtmf.com) got netsplit. [14:09] rcy (~rcy@dialin-184-59.calgary.primus.ca) got netsplit. [14:09] cVsup (~cVsup@janeway2.mav.com.br) got netsplit. [14:09] pll (~user@c-76-24-131-239.hsd1.ma.comcast.net) got netsplit. [14:09] juri_ (NkDrE20Hwt@volumehost.com) got netsplit. [14:09] alexanderwz (~alexander@karuna.med.harvard.edu) got netsplit. [14:09] lazyb0y_ (~henning@v683.vanager.de) got netsplit. [14:09] Lin (~igor@200.179.57.57) got netsplit. [14:09] MT (~MT@dove.informatik.tu-muenchen.de) got netsplit. [14:09] ErKa (keryell@m2.wifi.enstb.org) got netsplit. [14:09] fai-guy (~fai-guy@p57B9BA53.dip0.t-ipconnect.de) got netsplit. [14:09] glance (glance@130.239.18.148) got netsplit. [14:09] mat1 (~math@p54B48FB3.dip0.t-ipconnect.de) got netsplit. [14:09] Bluemoon_ (mj@altf4.org) got netsplit. [14:09] tokkee (tokkee@ssh.faui2k3.org) got netsplit. [14:09] torkel (torkel@ip64.degernas.se) got netsplit. [14:09] oz_ (~oz@anubis.coresystems.de) got netsplit. [14:09] ed (~ed@zehome.com) got netsplit. [14:09] Mrfai (~lange@kueppers.informatik.uni-koeln.de) got netsplit. [14:09] allee (~ach@allee.mpe.mpg.de) got netsplit. [14:09] \sh_away (~shermann@server3.servereyes.de) got netsplit. [14:09] kugg (~kugg@user184.77-105-210.netatonce.net) got netsplit. [14:09] sep (~sep@40.211.jostedal.no) got netsplit. [14:09] h01ger (~holger@socket.layer-acht.org) got netsplit. [14:09] thpo (~thpo@dotrc.de) got netsplit. [14:09] eartoast (~glaweh@ssh.physik.fu-berlin.de) got netsplit. [14:09] maswan (maswan@kennedy.acc.umu.se) got netsplit. [14:11] sep (~sep@40.211.jostedal.no) returned to #fai. [14:11] kugg (~kugg@user184.77-105-210.netatonce.net) returned to #fai. [14:11] \sh_away (~shermann@server3.servereyes.de) returned to #fai. [14:11] allee (~ach@allee.mpe.mpg.de) returned to #fai. [14:11] Mrfai (~lange@kueppers.informatik.uni-koeln.de) returned to #fai. [14:11] alexanderwz (~alexander@karuna.med.harvard.edu) returned to #fai. [14:11] Lin (~igor@200.179.57.57) returned to #fai. [14:11] pll (~user@c-76-24-131-239.hsd1.ma.comcast.net) returned to #fai. [14:11] lazyb0y_ (~henning@v683.vanager.de) returned to #fai. [14:11] h01ger (~holger@socket.layer-acht.org) returned to #fai. [14:11] rcy (~rcy@dialin-184-59.calgary.primus.ca) returned to #fai. [14:11] thpo (~thpo@dotrc.de) returned to #fai. [14:11] eartoast (~glaweh@ssh.physik.fu-berlin.de) returned to #fai. [14:11] maswan (maswan@kennedy.acc.umu.se) returned to #fai. [14:11] juri_ (NkDrE20Hwt@volumehost.com) returned to #fai. [14:11] blblack (~brandon@wasabi.dtmf.com) returned to #fai. [14:12] MT (~MT@dove.informatik.tu-muenchen.de) returned to #fai. [14:12] ErKa (keryell@m2.wifi.enstb.org) returned to #fai. [14:12] mat1 (~math@p54B48FB3.dip0.t-ipconnect.de) returned to #fai. [14:12] fai-guy (~fai-guy@p57B9BA53.dip0.t-ipconnect.de) returned to #fai. [14:12] glance (glance@130.239.18.148) returned to #fai. [14:12] Bluemoon_ (mj@altf4.org) returned to #fai. [14:12] tokkee (tokkee@ssh.faui2k3.org) returned to #fai. [14:12] ed (~ed@zehome.com) returned to #fai. [14:12] torkel (torkel@ip64.degernas.se) returned to #fai. [14:12] oz_ (~oz@anubis.coresystems.de) returned to #fai. [14:20] cVsup (~cVsup@janeway2.mav.com.br) got lost in the net-split. [14:29] hmm, but with standard NFSROOT file in /etc/fai and simple config from fai-doc examples, I still get the Error: GRUB in stage 1.5, Error 15, /target/boot/grub/* looks ok, all files like stage1 are present, and menu.lst contains valid paths [14:36] mat1: you mean, at install time? [14:37] then, for some reason, the class GRUB ist still defined for your install client [14:39] or some package you define depends on grub... [14:54] lazyb0y_: but i want to install grub so I guess it is neccessary that my host is in the GRUB class [14:54] yes I mean install-time [14:55] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Remote host closed the connection [14:55] mat1: paste your fai.log and shell.log to paste.debian.net. [14:55] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [14:56] I'll will do just 2 minutes [14:57] cause I've to reinstall, to get access to the log files [15:03] mat1: oh, I misunderstood you until now, that you are wondering why grub is there even if you don't want it... [15:07] so I've posted fai.log and shell.log [15:10] and disk_var.sh: [15:10] BOOT_DEVICE=/dev/sda [15:10] ROOT_PARTITION=/dev/sda1 [15:10] BOOT_PARTITION=/dev/sda1 [15:10] SWAPLIST="/dev/sda5" [15:13] to clearify myself: I want to install grub, but it doesn't work. [15:14] i meant clarify, sorry [15:16] cVsup (~cVsup@janeway2.mav.com.br) joined #fai. [15:22] mat1: but it looks good. HAve a look at the log messages from ==== shell: GRUB/10-setup ===== [15:23] Mrfai: where do I find them? [15:23] or did you meant the shell.log [15:23] mat1: at the end of http://paste.debian.net/45187 [15:23] yep in shell.log [15:23] ok I've allready seen them [15:24] so I wonder why the GRUB fails to boot [15:24] IMO this is not a problem of FAI. [15:24] MAybe brand new hardware, so the initrd does not detects the hard disk? [15:24] ErKa (keryell@m2.wifi.enstb.org) left irc: Ping timeout: 480 seconds [15:24] Yes I asked on #grub as well, they say no questions on grub if you don't tested with grub2 [15:25] which kind of controller is it? SATA? which vendor? [15:25] the problem was, that with the same hardware and fai 3.1.8 everything worked. [15:25] no this is my problem [15:25] so a plain 2.6.18 debian kernel recognizes your disks? [15:26] yes [15:27] with 3.1.8 everything worked, except raid + lvm, hence I changed to 3.2.4 as you suggested yesterday... a lot has changed [15:27] I removed /srv/fai/config, and subsitude it with the simple example from fai-doc [15:28] check the menu.lst which was created. Check if the file which are used there really exist in /target during isntallation. [15:30] that I've carefully checked, but the GRUB doesn't come to menu.lst at all before I can select anything it complains (in stage 1.5) that a file is missing [15:30] I know this isn't #grub, sorry for OT [15:32] I've even checked the device.map... but the explanation from GNU Grub on Error 15 in stage 1.5 is: "File not found" [15:32] "This error is returned if the specified file name cannot be found, but everything else (like the disk/partition info) is OK." [15:42] sepski (~sep@217.17.211.51) joined #fai. [15:54] AHCI was disabled in BIOS, sorry now it works [15:54] fai-guy (~fai-guy@p57B9BA53.dip0.t-ipconnect.de) left irc: Quit: schluss für heute [15:54] BTW: thanks for this software, I kinda like it [16:03] mat1: don't forget to fill out the FAI questionnaire: http://www.informatik.uni-koeln.de/fai/questionnaire [16:06] I'll do :-) [16:26] ErKa (keryell@keryell.pck.nerim.net) joined #fai. [16:29] allee (~ach@allee.mpe.mpg.de) left irc: Quit: Konversation terminated! [16:51] 'svn co http://svn.debian.org/wsvn/fai/people/michael/features/setup_harddisks_2' geht nicht, wie komme ich denn da ran? [16:51] oops sorry [16:51] svn co svn:// ... [16:51] (und dann kein wsvn sondern svn) [16:52] yes, thanks [17:01] sepski (~sep@217.17.211.51) left irc: Quit: Leaving [17:05] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Remote host closed the connection [17:05] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [17:09] Mrfai (~lange@kueppers.informatik.uni-koeln.de) left irc: Quit: leaving [17:21] Corey (~Corey@199.17.59.129) joined #fai. [17:31] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds [17:54] may this config: http://paste.debian.net/45195 work? [17:59] changed to http://paste.debian.net/45196 [18:04] sepski (~sep@217.17.211.51) joined #fai. [18:14] fai-guy (~fai-guy@p54BD7A08.dip.t-dialin.net) joined #fai. [18:18] storage magic can't locate init.pm, but @INC contains the path /srv/fai/config/sm/lib, where init.pm resides, any ideas? [18:39] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [18:52] cVsup (cVsup@janeway2.mav.com.br) left #fai (Leaving). [19:03] fai-guy (~fai-guy@p54BD7A08.dip.t-dialin.net) left irc: Ping timeout: 480 seconds [19:17] bye [19:17] mat1 (math@p54B48FB3.dip0.t-ipconnect.de) left #fai. [19:24] I need some help. [19:26] I'm trying to get fai setup in my university's lab and I finished setting up the fai server and the install clients boot off the network from it but then stops with a bin\sh can't accessto tty job control turned off [19:27] ask your uestion, it will be read [19:28] Corey, which FAI version, which distribution [19:28] how do I check [19:29] I'm running debian and just did an apt-get install fai-quickstart [19:29] etch [19:30] dpkg -l | grep fai [19:30] but I guess it should be 3.1.8 [19:30] it's 3.2.4 [19:31] are you really running etch? [19:31] that should be lenny or sid then ... [19:31] unless, of course, you added the uni-koeln repository [19:31] fai-guy (~fai-guy@p54BD7A08.dip.t-dialin.net) joined #fai. [19:32] yes I added the uni-koeln repository to my apt sources.list [19:33] ok [19:33] according to the installation guide the message I'm getting it says there is a permissions problem on my server [19:34] on the server when I type mount 127.0 [19:34] the question is what is printed before you end up in the shell? [19:34] the, say, 10 lines before that one [19:34] mount 127.0.0.1:/srv/fai/nfsroot it says failed permission denied [19:36] Corey: 127.0.0.1 is quitew a strange nfsserver [19:36] the install client says begin mounting root filesystem [19:36] I guess Corey tried that on the server [19:36] "no place but 127.0.0.1", but it this your value for $SERVER is wrong. [19:36] Corey, check your /etc/exports [19:37] Corey: how does your /etc/exports? [19:37] is nfsd running? [19:38] When I type showmount -e HP16 on the server it says its it's exporting the /srv/fai/nfsroot 10.0.6.69/255.0.0.0 and same for the config [19:39] I have it the defaults Fai-setup created [19:39] do you also have the koeln repo in /etc/fai/apt/sources.list? [19:39] yes [19:40] good :) [19:41] Corey: do the values 10.0.6.69/255.0.0.0 match your network? [19:41] yes the 10.0.6.69 is the server IP and the 255.0.0.0 is the network subnet mask [19:42] what about the server logs, do they tell anything about mount requests? [19:43] Corey: how does your /etc/export look like? [19:43] where do I find that [19:43] by typing `cat /etc/exports` ? [19:44] i suspect you just export to 10.0.6.69, not the entire network... [19:44] oz_ /srv/fai/nfsroot 10.0.6.69/255.0.0.0(async,rw,no_subtree_check,no_root_squash) and the config directory is the same [19:45] try to change it into 10.0.6.*/255.0.0.0(async,rw,no_subtree_check,no_root_squash) [19:45] and restart nfsd [19:45] I have tried to delete the IP and just put * instead but that didn't work either [19:47] on the install client when i get the (initramfs) I tried typing nfsmount 10.0.6.69:/srv/fai/nfsroot and it says no such file or directory [19:47] try to change it into "/srv/fai/nfsroot 10.0.6.*(async,rw,no_subtree_check,no_root_squash)" [19:47] subnet mask should be passed to nfsd by NIC configuration [19:48] /srv/fai/nfsroot does exist? [19:48] yes [19:49] the logs ... look at the logs ... that's what they are good for ... [19:49] what does /var/log/daemon.log tell you? [19:50] MT: interpreting the logs is _everything_, isn't it? ;) [19:50] the first thing is having the logs :-) [19:50] that's why M$ is so hard to debug :-) [19:54] daemon.log has nfsd[22218]: Unknown keyword "no_subtree_check" in export file and it has mountd[22220]: Unknown keyword "no_subtree_check" in export file [19:54] both of those are listed twice [19:54] just remove them from /etc/exports [19:54] try to change it into "/srv/fai/nfsroot 10.0.6.*(async,rw,subtree_check,no_root_squash)" [19:54] they are not strictly necessary [19:55] try to change it into "/srv/fai/nfsroot 10.0.6.*(async,ro,subtree_check,no_root_squash)"!!! [19:55] nfsroot is better set to ro [19:55] restart nfsd [19:57] ok give me a sec [20:00] nope still same problem on the install client [20:00] pls check the logs again [20:01] do they tell about any unauthenticated requests or the like? [20:01] no [20:01] then, about any authenticated requests? [20:02] no it doesn't say anything like that [20:04] on the client in live.log it says begin: trying nfsmount -o nolock -o ro 10.0.6.69:/srv/fai/nfsroot /live/image [20:04] then it has a whole bunch of lines that say rpc failed : 2 [20:05] what about rpcinfo -p [20:05] on the server? [20:08] what am I looking for in there [20:09] basically just check whether this works out [20:09] yeah [20:09] it has a couple ports one for tcp and udp for nfs and mountd [20:09] could you paste the output (to paste.debian.net) [20:09] Cyear, also the output of 'ifconfig -a' on both boxes [20:18] ok I posted both of them on there under my name Corey [20:18] url? [20:19] the latest two on paste.debian.net ... [20:19] the paste.deian.net [20:19] yeah [20:19] Corey, any firewall rules? [20:19] would be nice of you to paste them here, it savers work for your helpers [20:20] hmm, as you get to this shell [20:20] fai-guy (~fai-guy@p54BD7A08.dip.t-dialin.net) left irc: Ping timeout: 480 seconds [20:20] for every keystreoke you do you save 4 for us. [20:20] try nfsmount [20:21] nfsmount 10.0.6.69:/srv/fai/nfsroot [20:21] oz sorry MT said to put it on the paste.debian.net so that's where I put it [20:21] on the server or install client [20:22] on the client, in this shell you get [20:22] http://paste.debian.net/45208 [20:23] strange network of your unusual [20:23] strange network of yours, it's quite unusual [20:23] so. :) [20:24] your rpc services are not running... [20:24] hmm - which ones would you expect? [20:25] it's a university network. We are running our own subnet inside our university's network so we can do computing experiments and such for the business department [20:26] I tried the nfsmount and it says no such file or directory [20:27] can you paste the _exact_ error msg on paste.debian.net? [20:27] hmm, maybe mount works instead [20:31] sepski (~sep@217.17.211.51) left irc: Quit: Leaving [20:31] oz do you want the live.log or screen print out when it stops [20:31] whatever you can provide probably helps us understand your situation [20:32] MTyes, something like that [20:32] it would be nice if you could paste the output..."or similar" makes debugging quite hard for us [20:34] I tried the mount 10.0.6.69:/srv/fai/nfsroot and it says failed can't find /etc/fstab no such file or directory on the client [20:34] ah [20:34] :-) [20:35] mount 10.0.6.69:/srv/fai/nfsroot /tmp [20:35] (or the like) [20:35] try 'nfsmount 10.0.6.69:/srv/fai/nfsroot /tmp' [20:35] :-) [20:35] Corey said there is no nfsmount [20:35] using mount instead [20:35] Corey: what FAI version are you using? [20:36] MT: uh, i read it the other way round. :) [20:36] I'd try to fix it one the server, first [20:36] 3.2.4 [20:37] funny, how come that he has no nfsmount in the nfsroot? [20:37] I tried the mount 10.0.6.69:/srv/fai/nfsroot /tmp and it says failed no such file or directory on the client [20:37] fai-setup worked fine on the server [20:37] can you mount it on the server? [20:38] Corey, just check for some directory that exists on the client [20:38] and try to mount it there [20:38] no I tried that and it said permission denied [20:38] the client does have /tmp [20:38] I think it meant that the server doesn't have nfsroot directory but it does [20:39] Corey, pls check the logs of the server (once more...) [20:39] Corey: does the mount work on the server...? [20:39] (once more, 2 :-) ) [20:40] I guess Corey would need to add an additional export for 127.0.0.1 ... [20:41] no, he should not if exported to *, or should it? [20:44] on the server I tried mount 10.0.6.69:/srv/fai/nfsroot /mnt/testnfs and it says failed, reason given by server: permission denied. this is on the server [20:45] do the logs then tell anything? [20:45] Corey: the logs one the server.... [20:45] you want me to check the daemon log [20:45] in daemon.log [20:46] it says nfs mount of /srv/fai/nfsroot attempted from 10.0.6.69 [20:47] then unauthorized access by nfs client 10.0.6.69 [20:47] Corey: please _paste_ your /etc/exports [20:48] then blocked attempt of 10.0.6.69 to mount /srv/fai/nfsroot [20:48] the error very probably is located by a typo in there [20:48] anything about your tries from the client? [20:48] s/located/caused/ [20:49] year, 10.0.6.69 is your server's ip, isn't it? [20:49] same permission denied [20:49] yes [20:49] Corey: please _paste_ your /etc/exports [20:49] Last message repeated 1 time(s). [20:49] :-) [20:52] # /etc/exports: the access control list for filesystems which may be exported [20:52] # to NFS clients. See exports(5). [20:52] # [20:52] # Example for NFSv2 and NFSv3: [20:52] # /srv/homes hostname1(rw,sync) hostname2(ro,sync) [20:52] # [20:52] # Example for NFSv4: [20:52] # /srv/nfs4 gss/krb5i(rw,sync,fsid=0,crossmnt) [20:52] # /srv/nfs4/homes gss/krb5i(rw,sync) [20:52] # [20:52] .o0(on paste.debian.net ... :) ) [20:53] it's there [20:53] and that's all? [20:53] yes [20:53] there is nothing more in it? that's all commented out stuff. [20:53] http://paste.debian.net/45210 [20:53] yes [20:54] please change that back to 10.0.0.0/255.0.0.0 [20:54] do exportfs -r [20:55] and retry [20:55] thx [20:56] what does "exportfs" on the server tell you? [20:57] exportfs says -bash: exportfs: command not found [20:58] Corey: what does 'dpkg -l nfs-kernel-server' tell you? [20:58] that command should be there [20:59] ii nfs-kernel-ser 1.0.10-6+etch. Kernel NFS server support <- if you don't get this do: [20:59] apt-get install nfs-kernel-server [21:00] I uninstalled the nfs kernel server and installed nfs user server as a suggestion from one of the guys here [21:00] hm. I think that was a wrong choice... [21:00] !?!?!?! [21:00] seconded. [21:00] MT: who? [21:00] no idea.... [21:01] seconded? [21:01] I second your opinion [21:01] that nfs-user-server is a very very bad choice [21:01] Corey, I think nfs-user-server doesn't even support nfsv3, is pretty much unmaintained and the like... [21:02] oh [21:02] it's a big "no-no-no" :-) [21:02] so I'll switch it back [21:03] Action: oz_ places a cool bottle of champagner - Finish comes near [21:03] or so :-) [21:04] this also explains the no_subtree_check problem ... [21:04] right [21:04] but it does not yet (at least not in an obvious way) solve the main problem, but let's see [21:04] I wonder that the crap understands a "subtree_check" option. :) [21:05] MT: i think it should also explain the missing rpc [21:05] Action: oz_ still is grinning at the champagne [21:06] ok i reinstalled the kernel server and nfsmount says it's an unknown command but mount worked on the server [21:06] nfsmount on the server? [21:06] now try the client [21:07] oh yes, nfsmount is specific to some kernel utils, if mount is available that's fine [21:08] homer:~# apt-file search nfsmount -> yaird: usr/lib/yaird/exec/nfsmount [21:08] it's just in the initrd, it's very fine. [21:08] ??? [21:09] On the client nfsmount 10.0.6.69:/srv/fai/nfsroot /tmp says rpc failed: 2 [21:10] On the client mount 10.0.6.69:/srv/fai/nfsroot /tmp says failed no such file or directory [21:10] then, any entries in the server log? [21:11] Corey: what does "dpkg -l rstatd" tell you? [21:17] there is nothing from the client in the daemon log on the server [21:17] hmm, strange [21:17] Corey: what does "dpkg -l rstatd" tell you? [21:17] it says rstatd version [21:18] nor is there anything in the syslog or messages [21:18] apt-get install rstatd [21:18] ok [21:19] version is 4.0.1-2 [21:19] restart nfsd [21:19] then check again [21:20] ok [21:21] I wonder why some intelligent dependecy did not prevent this [21:23] same errors on client [21:25] what does daemon.log tell you? [21:25] still a missing rpc? [21:26] how does the output of 'rpcinfo -p' on the server look like? [21:28] btw...can you ping the client from the server? [21:29] yes ping works [21:30] how does the output of 'rpcinfo -p' on the server look like? [21:34] ok pasted rpcinfo on paste.debian [21:35] I don't think that you acutally need rstatd for the basic things, but that's just IMHO [21:35] what's IMHO [21:35] Corey, you might want to try some of the options to mount [21:35] In My Humble Opinion [21:35] oh [21:36] like what options [21:36] mom ... [21:37] would you guys want to try secure shell into the machine and take a look [21:37] try mount -t nfs -o nfsvers=3 [21:38] well, I usually don't spy on other people's hosts :-) [21:39] fai-guy (~fai-guy@p54BD7A08.dip.t-dialin.net) joined #fai. [21:39] lol [21:40] mount -t..... says a whole lot of I'm using the mount command wrong and this is how it's used [21:40] ah, did you add the 10. ... and /tmp to the above? [21:41] no [21:41] there is one thing you could to though [21:42] please paste /var/log/fai/make-fai-nfsroot.log to paste.debian.net [21:43] ok [21:44] I have to get going but you can e-mail me if you find anything at heco0701@stcloudstate.edu [21:44] thanks for all the help oz_ and MT I really appreciate it. [21:45] Corey, hopefully we can get that going later on ... [21:45] I'll be back on tomorrow around 10 or 11am [21:46] GMT? [21:46] Central zulu -6 [21:46] it's 3:45pm here not [21:46] now [21:47] thanks again [21:47] Corey (~Corey@199.17.59.129) left irc: Quit: Leaving [21:56] fai-guy (~fai-guy@p54BD7A08.dip.t-dialin.net) left irc: Quit: leaving [22:01] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds [23:36] ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds [00:00] --- Thu Dec 20 2007