<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7235.2">
<TITLE>RE: Root file system access though telnet</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<BR>

<P><FONT SIZE=2>Hi,<BR>
I want to know how root file system can be accessed through telnet.Actualy I connected to server through telnet and I want to access my root file system. can anyone tell me, how to access it.<BR>
<BR>
Thanks and Regards,<BR>
Ashok Bharat<BR>
<BR>
<BR>
-----Original Message-----<BR>
From: lustre-discuss-bounces@lists.lustre.org on behalf of lustre-discuss-request@lists.lustre.org<BR>
Sent: Fri 2/29/2008 1:56 PM<BR>
To: lustre-discuss@lists.lustre.org<BR>
Subject: Lustre-discuss Digest, Vol 25, Issue 59<BR>
<BR>
Send Lustre-discuss mailing list submissions to<BR>
        lustre-discuss@lists.lustre.org<BR>
<BR>
To subscribe or unsubscribe via the World Wide Web, visit<BR>
        <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
or, via email, send a message with subject or body 'help' to<BR>
        lustre-discuss-request@lists.lustre.org<BR>
<BR>
You can reach the person managing the list at<BR>
        lustre-discuss-owner@lists.lustre.org<BR>
<BR>
When replying, please edit your Subject line so it is more specific<BR>
than "Re: Contents of Lustre-discuss digest..."<BR>
<BR>
<BR>
Today's Topics:<BR>
<BR>
   1. Re: Configuration of lustre FS on single node (Amit Sharma)<BR>
   2. Re: Configuration of lustre FS on single node<BR>
      (ashok bharat bayana)<BR>
   3. Re: Configuration of lustre FS on single node (Amit Sharma)<BR>
<BR>
<BR>
----------------------------------------------------------------------<BR>
<BR>
Message: 1<BR>
Date: Fri, 29 Feb 2008 12:17:52 +0530<BR>
From: Amit Sharma <Amit.Sharma@Sun.COM><BR>
Subject: Re: [Lustre-discuss] Configuration of lustre FS on single<BR>
        node<BR>
To: ashok bharat bayana <ashok.bharat.bayana@iiitb.ac.in><BR>
Cc: lustre-discuss@lists.lustre.org<BR>
Message-ID: <47C7AA98.9010603@sun.com><BR>
Content-Type: text/plain; format=flowed; charset=ISO-8859-1<BR>
<BR>
Ashok,<BR>
<BR>
Just looking at the command you tried :<BR>
 > $ mkfs.lustre --fsname spfs --mdt --mgs /dev/sda<BR>
<BR>
You may have wrecked your root file system. Or is /dev/sda some other<BR>
device ?<BR>
<BR>
It may be simpler for you to try llmount.sh to get a simple setup of<BR>
lustre on a single node (with loop back mounts)<BR>
<BR>
thanks,<BR>
Amit<BR>
<BR>
ashok bharat bayana wrote:<BR>
><BR>
> Hello,<BR>
> I successfully build lustre(1.6.4.2) on my system for a patchless client<BR>
> But I dont know how to proceed in configuring the file system.<BR>
><BR>
> I'm trying to build all MDS,client and MDT's on a single node.<BR>
>  >From the tutorials I came to know that<BR>
><BR>
> First, create an MDT for the "spfs" file system that uses the /dev/sda<BR>
> disk. This MDT also acts as the MGS for the site.<BR>
><BR>
> $ mkfs.lustre --fsname spfs --mdt --mgs /dev/sda<BR>
><BR>
> But by triggering this command my system crashed in which the whole data<BR>
> is lost and again I need to install the OS.<BR>
> I want help in proceeding of mounting a lustre file system.<BR>
><BR>
> Thanks and Regards,<BR>
> Ashok Bharat<BR>
><BR>
> -----Original Message-----<BR>
> From: lustre-discuss-bounces@lists.lustre.org on behalf of<BR>
> lustre-discuss-request@lists.lustre.org<BR>
> Sent: Wed 2/27/2008 10:30 PM<BR>
> To: lustre-discuss@lists.lustre.org<BR>
> Subject: Lustre-discuss Digest, Vol 25, Issue 56<BR>
><BR>
> Send Lustre-discuss mailing list submissions to<BR>
>         lustre-discuss@lists.lustre.org<BR>
><BR>
> To subscribe or unsubscribe via the World Wide Web, visit<BR>
>         <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
> or, via email, send a message with subject or body 'help' to<BR>
>         lustre-discuss-request@lists.lustre.org<BR>
><BR>
> You can reach the person managing the list at<BR>
>         lustre-discuss-owner@lists.lustre.org<BR>
><BR>
> When replying, please edit your Subject line so it is more specific<BR>
> than "Re: Contents of Lustre-discuss digest..."<BR>
><BR>
><BR>
> Today's Topics:<BR>
><BR>
>    1. Multiple NICs per OST (Joshua Bower-Cooley)<BR>
>    2. obd error on MGS/MDT node (Jeremy Mann)<BR>
><BR>
><BR>
> ----------------------------------------------------------------------<BR>
><BR>
> Message: 1<BR>
> Date: Tue, 26 Feb 2008 11:55:19 -0800<BR>
> From: Joshua Bower-Cooley <jbowercooley@lcogt.net><BR>
> Subject: [Lustre-discuss] Multiple NICs per OST<BR>
> To: lustre-discuss@lists.lustre.org<BR>
> Message-ID: <200802261155.19907.jbowercooley@lcogt.net><BR>
> Content-Type: text/plain;  charset="utf-8"<BR>
><BR>
> Hi-<BR>
> I've had lustre running happily for some time over a single 10g ethernet NIC<BR>
> per node. After switching to dual 10g and creating a new filesystem<BR>
> (1.6.4.2), I'm seeing nothing but keep-alive packets with bad checksums.<BR>
><BR>
> What is the current "correct" way to do this now? The manual suggest not<BR>
> using<BR>
> bonding, but several list postings now reccommend it. Without bonding, do I<BR>
> need to have my 2 switches stacked, or will Lustre recognize the division in<BR>
> my subnet?<BR>
><BR>
> LNET module options I've tried are:<BR>
> 1) networks="tcp0(eth2,eth3)"<BR>
> 2) ip2nets="tcp(eth2,eth3); tcp(eth2) 10.9.[1-4].*; tcp(eth3) 10.9.[5-8].*;"<BR>
> and many other variations<BR>
><BR>
> All other services are functional on both interfaces.<BR>
><BR>
> Thanks in advance,<BR>
> J Bower-Cooley<BR>
> Scientific Systems Engineer<BR>
> Las Cumbres Observatory Global Telescope<BR>
><BR>
><BR>
> ------------------------------<BR>
><BR>
> Message: 2<BR>
> Date: Wed, 27 Feb 2008 09:50:08 -0600 (CST)<BR>
> From: "Jeremy Mann" <jeremy@biochem.uthscsa.edu><BR>
> Subject: [Lustre-discuss] obd error on MGS/MDT node<BR>
> To: lustre-discuss@lists.lustre.org<BR>
> Message-ID:<BR>
>         <62901.24.173.62.147.1204127408.squirrel@biochem.uthscsa.edu><BR>
> Content-Type: text/plain;charset=iso-8859-1<BR>
><BR>
> Today we are starting to see obd_change_cbdata errors on our MGS/MDT node.<BR>
> What does this error mean?<BR>
><BR>
> Version of lustre and kernel is:<BR>
><BR>
> 2.6.9-42.0.10.EL_lustre-1.6.0.1smp X86_64<BR>
><BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
> previous similar message<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
> previous similar message<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 2<BR>
> previous similar messages<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 4<BR>
> previous similar messages<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 5<BR>
> previous similar messages<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
> previous similar message<BR>
><BR>
> --<BR>
> Jeremy Mann<BR>
> jeremy@biochem.uthscsa.edu<BR>
><BR>
> University of Texas Health Science Center<BR>
> Bioinformatics Core Facility<BR>
> <A HREF="http://www.bioinformatics.uthscsa.edu">http://www.bioinformatics.uthscsa.edu</A><BR>
> Phone: (210) 567-2672<BR>
><BR>
><BR>
><BR>
> ------------------------------<BR>
><BR>
> _______________________________________________<BR>
> Lustre-discuss mailing list<BR>
> Lustre-discuss@lists.lustre.org<BR>
> <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
><BR>
><BR>
> End of Lustre-discuss Digest, Vol 25, Issue 56<BR>
> **********************************************<BR>
><BR>
><BR>
> ------------------------------------------------------------------------<BR>
><BR>
> _______________________________________________<BR>
> Lustre-discuss mailing list<BR>
> Lustre-discuss@lists.lustre.org<BR>
> <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
<BR>
<BR>
--<BR>
Amit Sharma<BR>
Lustre Engineering<BR>
Sun Microsystems, Bangalore.<BR>
<BR>
<BR>
------------------------------<BR>
<BR>
Message: 2<BR>
Date: Fri, 29 Feb 2008 12:51:31 +0530<BR>
From: "ashok bharat bayana" <ashok.bharat.bayana@iiitb.ac.in><BR>
Subject: Re: [Lustre-discuss] Configuration of lustre FS on single<BR>
        node<BR>
To: "Amit Sharma" <Amit.Sharma@Sun.COM><BR>
Cc: lustre-discuss@lists.lustre.org<BR>
Message-ID: <8626C1B7EB748940BCDD7596134632BE85022B@jal.iiitb.ac.in><BR>
Content-Type: text/plain; charset="iso-8859-1"<BR>
<BR>
hi,<BR>
/dev/sda is a non ext3 FS, that's the reason i think, my system crashed.<BR>
my linux fs is on /dev/sda8 and an ext3 one So should I go with<BR>
<BR>
$ mkfs.lustre --fsname spfs --mdt --mgs /dev/sda8<BR>
<BR>
and proceed?<BR>
<BR>
and Moreover, to run the llmount.sh or local.sh we should have lmc,lconf and lctl utilities, but these are not provided in lustre1.6.4.2, so that's why I'm going in this way.<BR>
<BR>
Thanks and Regards,<BR>
Ashok Bharat<BR>
<BR>
<BR>
-----Original Message-----<BR>
From: Amit Sharma [<A HREF="mailto:Amit.Sharma@Sun.COM">mailto:Amit.Sharma@Sun.COM</A>]<BR>
Sent: Fri 2/29/2008 12:17 PM<BR>
To: ashok bharat bayana<BR>
Cc: lustre-discuss@lists.lustre.org<BR>
Subject: Re: [Lustre-discuss] Configuration of lustre FS on single node<BR>
<BR>
Ashok,<BR>
<BR>
Just looking at the command you tried :<BR>
 > $ mkfs.lustre --fsname spfs --mdt --mgs /dev/sda<BR>
<BR>
You may have wrecked your root file system. Or is /dev/sda some other<BR>
device ?<BR>
<BR>
It may be simpler for you to try llmount.sh to get a simple setup of<BR>
lustre on a single node (with loop back mounts)<BR>
<BR>
thanks,<BR>
Amit<BR>
<BR>
ashok bharat bayana wrote:<BR>
><BR>
> Hello,<BR>
> I successfully build lustre(1.6.4.2) on my system for a patchless client<BR>
> But I dont know how to proceed in configuring the file system.<BR>
><BR>
> I'm trying to build all MDS,client and MDT's on a single node.<BR>
>  >From the tutorials I came to know that<BR>
><BR>
> First, create an MDT for the "spfs" file system that uses the /dev/sda<BR>
> disk. This MDT also acts as the MGS for the site.<BR>
><BR>
> $ mkfs.lustre --fsname spfs --mdt --mgs /dev/sda<BR>
><BR>
> But by triggering this command my system crashed in which the whole data<BR>
> is lost and again I need to install the OS.<BR>
> I want help in proceeding of mounting a lustre file system.<BR>
><BR>
> Thanks and Regards,<BR>
> Ashok Bharat<BR>
><BR>
> -----Original Message-----<BR>
> From: lustre-discuss-bounces@lists.lustre.org on behalf of<BR>
> lustre-discuss-request@lists.lustre.org<BR>
> Sent: Wed 2/27/2008 10:30 PM<BR>
> To: lustre-discuss@lists.lustre.org<BR>
> Subject: Lustre-discuss Digest, Vol 25, Issue 56<BR>
><BR>
> Send Lustre-discuss mailing list submissions to<BR>
>         lustre-discuss@lists.lustre.org<BR>
><BR>
> To subscribe or unsubscribe via the World Wide Web, visit<BR>
>         <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
> or, via email, send a message with subject or body 'help' to<BR>
>         lustre-discuss-request@lists.lustre.org<BR>
><BR>
> You can reach the person managing the list at<BR>
>         lustre-discuss-owner@lists.lustre.org<BR>
><BR>
> When replying, please edit your Subject line so it is more specific<BR>
> than "Re: Contents of Lustre-discuss digest..."<BR>
><BR>
><BR>
> Today's Topics:<BR>
><BR>
>    1. Multiple NICs per OST (Joshua Bower-Cooley)<BR>
>    2. obd error on MGS/MDT node (Jeremy Mann)<BR>
><BR>
><BR>
> ----------------------------------------------------------------------<BR>
><BR>
> Message: 1<BR>
> Date: Tue, 26 Feb 2008 11:55:19 -0800<BR>
> From: Joshua Bower-Cooley <jbowercooley@lcogt.net><BR>
> Subject: [Lustre-discuss] Multiple NICs per OST<BR>
> To: lustre-discuss@lists.lustre.org<BR>
> Message-ID: <200802261155.19907.jbowercooley@lcogt.net><BR>
> Content-Type: text/plain;  charset="utf-8"<BR>
><BR>
> Hi-<BR>
> I've had lustre running happily for some time over a single 10g ethernet NIC<BR>
> per node. After switching to dual 10g and creating a new filesystem<BR>
> (1.6.4.2), I'm seeing nothing but keep-alive packets with bad checksums.<BR>
><BR>
> What is the current "correct" way to do this now? The manual suggest not<BR>
> using<BR>
> bonding, but several list postings now reccommend it. Without bonding, do I<BR>
> need to have my 2 switches stacked, or will Lustre recognize the division in<BR>
> my subnet?<BR>
><BR>
> LNET module options I've tried are:<BR>
> 1) networks="tcp0(eth2,eth3)"<BR>
> 2) ip2nets="tcp(eth2,eth3); tcp(eth2) 10.9.[1-4].*; tcp(eth3) 10.9.[5-8].*;"<BR>
> and many other variations<BR>
><BR>
> All other services are functional on both interfaces.<BR>
><BR>
> Thanks in advance,<BR>
> J Bower-Cooley<BR>
> Scientific Systems Engineer<BR>
> Las Cumbres Observatory Global Telescope<BR>
><BR>
><BR>
> ------------------------------<BR>
><BR>
> Message: 2<BR>
> Date: Wed, 27 Feb 2008 09:50:08 -0600 (CST)<BR>
> From: "Jeremy Mann" <jeremy@biochem.uthscsa.edu><BR>
> Subject: [Lustre-discuss] obd error on MGS/MDT node<BR>
> To: lustre-discuss@lists.lustre.org<BR>
> Message-ID:<BR>
>         <62901.24.173.62.147.1204127408.squirrel@biochem.uthscsa.edu><BR>
> Content-Type: text/plain;charset=iso-8859-1<BR>
><BR>
> Today we are starting to see obd_change_cbdata errors on our MGS/MDT node.<BR>
> What does this error mean?<BR>
><BR>
> Version of lustre and kernel is:<BR>
><BR>
> 2.6.9-42.0.10.EL_lustre-1.6.0.1smp X86_64<BR>
><BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
> previous similar message<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
> previous similar message<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 2<BR>
> previous similar messages<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 4<BR>
> previous similar messages<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 5<BR>
> previous similar messages<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
> obd_change_cbdata: NULL export<BR>
> LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
> previous similar message<BR>
><BR>
> --<BR>
> Jeremy Mann<BR>
> jeremy@biochem.uthscsa.edu<BR>
><BR>
> University of Texas Health Science Center<BR>
> Bioinformatics Core Facility<BR>
> <A HREF="http://www.bioinformatics.uthscsa.edu">http://www.bioinformatics.uthscsa.edu</A><BR>
> Phone: (210) 567-2672<BR>
><BR>
><BR>
><BR>
> ------------------------------<BR>
><BR>
> _______________________________________________<BR>
> Lustre-discuss mailing list<BR>
> Lustre-discuss@lists.lustre.org<BR>
> <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
><BR>
><BR>
> End of Lustre-discuss Digest, Vol 25, Issue 56<BR>
> **********************************************<BR>
><BR>
><BR>
> ------------------------------------------------------------------------<BR>
><BR>
> _______________________________________________<BR>
> Lustre-discuss mailing list<BR>
> Lustre-discuss@lists.lustre.org<BR>
> <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
<BR>
<BR>
--<BR>
Amit Sharma<BR>
Lustre Engineering<BR>
Sun Microsystems, Bangalore.<BR>
<BR>
-------------- next part --------------<BR>
An HTML attachment was scrubbed...<BR>
URL: <A HREF="http://lists.lustre.org/pipermail/lustre-discuss/attachments/20080229/b7713b02/attachment-0001.html">http://lists.lustre.org/pipermail/lustre-discuss/attachments/20080229/b7713b02/attachment-0001.html</A><BR>
<BR>
------------------------------<BR>
<BR>
Message: 3<BR>
Date: Fri, 29 Feb 2008 13:56:31 +0530<BR>
From: Amit Sharma <Amit.Sharma@Sun.COM><BR>
Subject: Re: [Lustre-discuss] Configuration of lustre FS on single<BR>
        node<BR>
To: ashok bharat bayana <ashok.bharat.bayana@iiitb.ac.in><BR>
Cc: lustre-discuss@lists.lustre.org<BR>
Message-ID: <47C7C1B7.8050806@sun.com><BR>
Content-Type: text/plain; format=flowed; charset=ISO-8859-1<BR>
<BR>
Ashok,<BR>
<BR>
If you are trying to setup lustre (with all the servers MDS, OSS.. etc)<BR>
building only the patchless client wont help you.<BR>
<BR>
You will have to build lustre as described here :<BR>
<A HREF="http://wiki.lustre.org/index.php?title=Lustre_Howto">http://wiki.lustre.org/index.php?title=Lustre_Howto</A><BR>
<BR>
thanks,<BR>
Amit.<BR>
<BR>
ashok bharat bayana wrote:<BR>
><BR>
> Hi,<BR>
> the kernel version I used is 2.6.20.15 and the distro is Ubuntu 7.10<BR>
> The steps so far I have done are:<BR>
> 1)I have downloaded the sources lustre(1.6.4.2) from CFS<BR>
> 2)I tried to build lustre for a patchless client in which i triggered<BR>
> the command,<BR>
> [lustre]$ ./configure --with-linux=/unpatched/kernel/source<BR>
> 3)then I build it using make and then installed the modules using make<BR>
> install commands<BR>
> (I'm not sure whether the modules are installed correctly or not.Isn't<BR>
> the modules are installed using make install cmd)<BR>
> 4)Now in /usr/lib there is a lustre folder created.<BR>
> (Can we setup all MDS,client and OST's on a single system? my system<BR>
> configuration is Intel Pentium 512 MB RAM 1.6 GHz)<BR>
> Could you please tell me how to mount lustre fs in my laptop.<BR>
><BR>
> Thanks and Regards,<BR>
> Ashok Bharat<BR>
> -----Original Message-----<BR>
> From: Amit Sharma [<A HREF="mailto:Amit.Sharma@Sun.COM">mailto:Amit.Sharma@Sun.COM</A>]<BR>
> Sent: Fri 2/29/2008 12:03 PM<BR>
> To: ashok bharat bayana<BR>
> Cc: lustre-discuss@lists.lustre.org<BR>
> Subject: Re: [Lustre-discuss] Configuration of lustre FS on single node<BR>
><BR>
> Ashok,<BR>
><BR>
> Could you let us know what are the modules that you have installed on<BR>
> the system. The linux version on your system. And the steps that you<BR>
> have done so far.<BR>
><BR>
> thanks,<BR>
> Amit<BR>
> ashok bharat bayana wrote:<BR>
>  ><BR>
>  > Hello,<BR>
>  > I successfully build lustre(1.6.4.2) on my system for a patchless client<BR>
>  > But I dont know how to proceed in configuring the file system.<BR>
>  ><BR>
>  > I'm trying to build all MDS,client and MDT's on a single node.<BR>
>  >  >From the tutorials I came to know that<BR>
>  ><BR>
>  > First, create an MDT for the "spfs" file system that uses the /dev/sda<BR>
>  > disk. This MDT also acts as the MGS for the site.<BR>
>  ><BR>
>  > $ mkfs.lustre --fsname spfs --mdt --mgs /dev/sda<BR>
>  ><BR>
>  > But by triggering this command my system crashed in which the whole data<BR>
>  > is lost and again I need to install the OS.<BR>
>  > I want help in proceeding of mounting a lustre file system.<BR>
>  ><BR>
>  > Thanks and Regards,<BR>
>  > Ashok Bharat<BR>
>  ><BR>
>  > -----Original Message-----<BR>
>  > From: lustre-discuss-bounces@lists.lustre.org on behalf of<BR>
>  > lustre-discuss-request@lists.lustre.org<BR>
>  > Sent: Wed 2/27/2008 10:30 PM<BR>
>  > To: lustre-discuss@lists.lustre.org<BR>
>  > Subject: Lustre-discuss Digest, Vol 25, Issue 56<BR>
>  ><BR>
>  > Send Lustre-discuss mailing list submissions to<BR>
>  >         lustre-discuss@lists.lustre.org<BR>
>  ><BR>
>  > To subscribe or unsubscribe via the World Wide Web, visit<BR>
>  >         <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
>  > or, via email, send a message with subject or body 'help' to<BR>
>  >         lustre-discuss-request@lists.lustre.org<BR>
>  ><BR>
>  > You can reach the person managing the list at<BR>
>  >         lustre-discuss-owner@lists.lustre.org<BR>
>  ><BR>
>  > When replying, please edit your Subject line so it is more specific<BR>
>  > than "Re: Contents of Lustre-discuss digest..."<BR>
>  ><BR>
>  ><BR>
>  > Today's Topics:<BR>
>  ><BR>
>  >    1. Multiple NICs per OST (Joshua Bower-Cooley)<BR>
>  >    2. obd error on MGS/MDT node (Jeremy Mann)<BR>
>  ><BR>
>  ><BR>
>  > ----------------------------------------------------------------------<BR>
>  ><BR>
>  > Message: 1<BR>
>  > Date: Tue, 26 Feb 2008 11:55:19 -0800<BR>
>  > From: Joshua Bower-Cooley <jbowercooley@lcogt.net><BR>
>  > Subject: [Lustre-discuss] Multiple NICs per OST<BR>
>  > To: lustre-discuss@lists.lustre.org<BR>
>  > Message-ID: <200802261155.19907.jbowercooley@lcogt.net><BR>
>  > Content-Type: text/plain;  charset="utf-8"<BR>
>  ><BR>
>  > Hi-<BR>
>  > I've had lustre running happily for some time over a single 10g<BR>
> ethernet NIC<BR>
>  > per node. After switching to dual 10g and creating a new filesystem<BR>
>  > (1.6.4.2), I'm seeing nothing but keep-alive packets with bad checksums.<BR>
>  ><BR>
>  > What is the current "correct" way to do this now? The manual suggest not<BR>
>  > using<BR>
>  > bonding, but several list postings now reccommend it. Without<BR>
> bonding, do I<BR>
>  > need to have my 2 switches stacked, or will Lustre recognize the<BR>
> division in<BR>
>  > my subnet?<BR>
>  ><BR>
>  > LNET module options I've tried are:<BR>
>  > 1) networks="tcp0(eth2,eth3)"<BR>
>  > 2) ip2nets="tcp(eth2,eth3); tcp(eth2) 10.9.[1-4].*; tcp(eth3)<BR>
> 10.9.[5-8].*;"<BR>
>  > and many other variations<BR>
>  ><BR>
>  > All other services are functional on both interfaces.<BR>
>  ><BR>
>  > Thanks in advance,<BR>
>  > J Bower-Cooley<BR>
>  > Scientific Systems Engineer<BR>
>  > Las Cumbres Observatory Global Telescope<BR>
>  ><BR>
>  ><BR>
>  > ------------------------------<BR>
>  ><BR>
>  > Message: 2<BR>
>  > Date: Wed, 27 Feb 2008 09:50:08 -0600 (CST)<BR>
>  > From: "Jeremy Mann" <jeremy@biochem.uthscsa.edu><BR>
>  > Subject: [Lustre-discuss] obd error on MGS/MDT node<BR>
>  > To: lustre-discuss@lists.lustre.org<BR>
>  > Message-ID:<BR>
>  >         <62901.24.173.62.147.1204127408.squirrel@biochem.uthscsa.edu><BR>
>  > Content-Type: text/plain;charset=iso-8859-1<BR>
>  ><BR>
>  > Today we are starting to see obd_change_cbdata errors on our MGS/MDT<BR>
> node.<BR>
>  > What does this error mean?<BR>
>  ><BR>
>  > Version of lustre and kernel is:<BR>
>  ><BR>
>  > 2.6.9-42.0.10.EL_lustre-1.6.0.1smp X86_64<BR>
>  ><BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
>  > previous similar message<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
>  > obd_change_cbdata: NULL export<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
>  > previous similar message<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
>  > obd_change_cbdata: NULL export<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 2<BR>
>  > previous similar messages<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
>  > obd_change_cbdata: NULL export<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 4<BR>
>  > previous similar messages<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
>  > obd_change_cbdata: NULL export<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 5<BR>
>  > previous similar messages<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata())<BR>
>  > obd_change_cbdata: NULL export<BR>
>  > LustreError: 65:0:(obd_class.h:1171:obd_change_cbdata()) Skipped 1<BR>
>  > previous similar message<BR>
>  ><BR>
>  > --<BR>
>  > Jeremy Mann<BR>
>  > jeremy@biochem.uthscsa.edu<BR>
>  ><BR>
>  > University of Texas Health Science Center<BR>
>  > Bioinformatics Core Facility<BR>
>  > <A HREF="http://www.bioinformatics.uthscsa.edu">http://www.bioinformatics.uthscsa.edu</A><BR>
>  > Phone: (210) 567-2672<BR>
>  ><BR>
>  ><BR>
>  ><BR>
>  > ------------------------------<BR>
>  ><BR>
>  > _______________________________________________<BR>
>  > Lustre-discuss mailing list<BR>
>  > Lustre-discuss@lists.lustre.org<BR>
>  > <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
>  ><BR>
>  ><BR>
>  > End of Lustre-discuss Digest, Vol 25, Issue 56<BR>
>  > **********************************************<BR>
>  ><BR>
>  ><BR>
>  > ------------------------------------------------------------------------<BR>
>  ><BR>
>  > _______________________________________________<BR>
>  > Lustre-discuss mailing list<BR>
>  > Lustre-discuss@lists.lustre.org<BR>
>  > <A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
><BR>
><BR>
> --<BR>
> Amit Sharma<BR>
> Lustre Engineering<BR>
> Sun Microsystems, Bangalore.<BR>
><BR>
<BR>
<BR>
--<BR>
Amit Sharma<BR>
Lustre Engineering<BR>
Sun Microsystems, Bangalore.<BR>
<BR>
<BR>
------------------------------<BR>
<BR>
_______________________________________________<BR>
Lustre-discuss mailing list<BR>
Lustre-discuss@lists.lustre.org<BR>
<A HREF="http://lists.lustre.org/mailman/listinfo/lustre-discuss">http://lists.lustre.org/mailman/listinfo/lustre-discuss</A><BR>
<BR>
<BR>
End of Lustre-discuss Digest, Vol 25, Issue 59<BR>
**********************************************<BR>
<BR>
</FONT>
</P>

</BODY>
</HTML>