<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi Amy,<div><br></div><div>You could try first unmount all ost, mgs, etc, and redo a tunefs on each relevant disk:</div><div><div><br></div><div>tunefs.lustre --writeconf --mgs --mdt --fsname=lufs DISKNAME</div><div>tunefs.lustre --erase-param --mgsnode=10.0.38.102@tcp0 --writeconf DISKNAME</div><div><br></div><div>Best Regards,</div><div>Jiawei</div><div><br></div><div><div>On Aug 10, 2009, at 3:56 PM, Lee Amy wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div>On Mon, Aug 10, 2009 at 9:32 AM, Lee Amy<<a href="mailto:openlinuxsource@gmail.com">openlinuxsource@gmail.com</a>> wrote:<br><blockquote type="cite">---------- Forwarded message ----------<br></blockquote><blockquote type="cite">From: Lee Amy <<a href="mailto:openlinuxsource@gmail.com">openlinuxsource@gmail.com</a>><br></blockquote><blockquote type="cite">Date: Mon, Aug 10, 2009 at 9:32 AM<br></blockquote><blockquote type="cite">Subject: Re: [Lustre-discuss] Help: NIC Changed Error<br></blockquote><blockquote type="cite">To: Rhys McMurdo <<a href="mailto:rhys@mcmurdo.id.au">rhys@mcmurdo.id.au</a>><br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">On Mon, Aug 10, 2009 at 6:14 AM, Rhys McMurdo<<a href="mailto:rhys@mcmurdo.id.au">rhys@mcmurdo.id.au</a>> wrote:<br></blockquote><blockquote type="cite"><blockquote type="cite">Hi Amy,<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">You may want to try the following options in your /etc/modprobe.conf<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">options lnet networks=tcp0(eth1)<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">Regards,<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">Rhys<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">2009/8/8 Lee Amy <<a href="mailto:openlinuxsource@gmail.com">openlinuxsource@gmail.com</a>><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Hi,<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">I'm a Lustre newbie. The server I set up is combined MGS/MDT file<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">system on a block device. And set up OST on a block device. I set up<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">MGS/MDT and OST in the same machine by using 2 disks. The NID is<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">10.0.38.102@tcp, and the address 10.0.38.102 was assigned to eth0. One<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">day I noticed the eth0 is broken so I use another NIC eth1 then assign<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">IP address 10.0.38.102 to this card.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Then I use client the mount the server Lustre FS by following command.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">mount -t lustre 10.0.38.102@tcp:/ericlfs /mnt/foobar<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">It reported following error messages.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Lustre: Request x1310428982411274 sent from MGC10.0.38.102@tcp to NID<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">10.0.38.102@tcp 5s ago has timed out (limit 5s).<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">LustreError: 4397:0:(client.c:792:ptlrpc_import_delay_req()) @@@<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">IMP_INVALID  req@ffff81002cb7d800 x1310428982411276/t0<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">o501-><a href="mailto:MGS@MGC10.0.38.102">MGS@MGC10.0.38.102</a>@tcp_0:26/25 lens 264/432 e 0 to 1 dl 0 ref 1<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">fl Rpc:/0/0 rc 0/0<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">LustreError: 15c-8: MGC10.0.38.102@tcp: The configuration from log<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">'ericlfs-client' failed (-108). This may be the result of<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">communication errors between this node and the MGS, a bad<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">configuration, or other errors. See the syslog for more information.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">LustreError: 4397:0:(llite_lib.c:1169:ll_fill_super()) Unable to<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">process log: -108<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Lustre: client ffff81002bd17400 umount complete<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">mount.lustre: mount 10.0.38.102@tcp:/ericlfs at /mnt failed: Cannot<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">send after transport endpoint shutdown<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">So I feel a little confused. Is this problem caused by I replace the<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">NIC card? And furthermore, how do I fix that problem?<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Thank you very much.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Best Regards,<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Amy<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">_______________________________________________<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Lustre-discuss mailing list<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><a href="mailto:Lustre-discuss@lists.lustre.org">Lustre-discuss@lists.lustre.org</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><a href="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">_______________________________________________<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">Lustre-discuss mailing list<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><a href="mailto:Lustre-discuss@lists.lustre.org">Lustre-discuss@lists.lustre.org</a><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><a href="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</a><br></blockquote></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Thanks very much. Anyway, my nid is 10.0.38.102@tcp, not<br></blockquote><blockquote type="cite">10.0.38.102@tcp0. If I add the above item in /etc/modprobe.conf I<br></blockquote><blockquote type="cite">don't know whether it will affect something wrong.<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Could you tell me what's the difference between tcp and tcp?<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Thank you very much.<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Regards,<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Amy<br></blockquote><blockquote type="cite"><br></blockquote>Hi,<br><br>It seems this method cannot solve my problem. My NID is<br>10.0.38.102@tcp, and furthermore when I add the item<br><br>options lnet network=tcp0(eth1)<br><br>I still encountered the same problem and after this failure I change<br>this item back to<br><br>options lnet network=tcp<br><br>That still got failure. So I really feel very confused about that.<br>When I installed Lustre the NID is 10.0.68.102@tcp. not tcp0 suffix.<br><br>Could someone tell me how to fix that problem?<br><br>Thank you very much.<br><br>Regards,<br><br>Amy<br>_______________________________________________<br>Lustre-discuss mailing list<br><a href="mailto:Lustre-discuss@lists.lustre.org">Lustre-discuss@lists.lustre.org</a><br>http://lists.lustre.org/mailman/listinfo/lustre-discuss<br></div></blockquote></div><br></div></body></html>