[lustre-discuss] Issue with lnet showing network down lustre 2.5.3

Patrick Farrell paf at cray.com
Thu May 28 08:50:32 PDT 2015


Dmesg should have more info.
________________________________________
From: lustre-discuss [lustre-discuss-bounces at lists.lustre.org] on behalf of Kurt Strosahl [strosahl at jlab.org]
Sent: Thursday, May 28, 2015 10:48 AM
To: Edward Wahl
Cc: lustre-discuss at lists.lustre.org
Subject: Re: [lustre-discuss] Issue with lnet showing network down lustre       2.5.3

Hi,

I've been loading the lnet modules with service lnet start, when I try loading the lustre modules I get the following...

# modprobe lustre
WARNING: Error inserting fid (/lib/modules/2.6.32-220.7.1.el6.x86_64/extra/kernel/fs/lustre/fid.ko): Input/output error
WARNING: Error inserting mdc (/lib/modules/2.6.32-220.7.1.el6.x86_64/extra/kernel/fs/lustre/mdc.ko): Input/output error
WARNING: Error inserting osc (/lib/modules/2.6.32-220.7.1.el6.x86_64/extra/kernel/fs/lustre/osc.ko): Input/output error
WARNING: Error inserting lov (/lib/modules/2.6.32-220.7.1.el6.x86_64/extra/kernel/fs/lustre/lov.ko): Input/output error
FATAL: Error inserting lustre (/lib/modules/2.6.32-220.7.1.el6.x86_64/extra/kernel/fs/lustre/lustre.ko): Input/output error

The lnet options should be fine, I'm using the lnet.conf from a system that is already up and running with 2.5.3

w/r,
Kurt

----- Original Message -----
From: "Edward Wahl" <ewahl at osc.edu>
To: "Kurt Strosahl" <strosahl at jlab.org>, lustre-discuss at lists.lustre.org
Sent: Thursday, May 28, 2015 10:40:48 AM
Subject: RE: Issue with lnet showing network down lustre 2.5.3

Were you loading both lustre and lnet with a modprobe/insmod?  Try just modprobe on lustre.  I saw something similar from 1.8 to 2.4 clients.  JUST could not get IB up with modprobe lnet and then realized that only loading lustre fixed it.  I did not look into the details, but assume I'd done something silly/incorrect in my /etc/modprobe or client build option for o2ib=. (RHEL ofed on clients, not Melanox)  I never looked too closely due to time restrictions and another move to 2.5.x quickly afterwards.

What client compile options did you use for RHEL 6?

Ed Wahl
OSC

________________________________________
From: lustre-discuss [lustre-discuss-bounces at lists.lustre.org] on behalf of Kurt Strosahl [strosahl at jlab.org]
Sent: Thursday, May 28, 2015 9:27 AM
To: lustre-discuss at lists.lustre.org
Subject: [lustre-discuss] Issue with lnet showing network down lustre 2.5.3

Good Morning,

   I've encountered an unusual issue when upgrading a lustre client from 1.8.7 to 2.5.3.  The client installed, but when I start the lnet service it does not come up (all the modules load).  When I run a service lnet status it shows the following error:

sudo service lnet status
cat: /proc/sys/lnet/routes: No such file or directory
loaded

The box has a qdr ib card and a 10 gig ethernet card.  It is running centos 6.2 with the lustre client compiled on the box.

~> rpm -qa | grep lustre
lustre-iokit-2.5.3-2.6.32_220.7.1.el6.x86_64.x86_64
lustre-client-2.5.3-2.6.32_220.7.1.el6.x86_64.x86_64
lustre-client-tests-2.5.3-2.6.32_220.7.1.el6.x86_64.x86_64
lustre-client-modules-2.5.3-2.6.32_220.7.1.el6.x86_64.x86_64
lustre-client-source-2.5.3-2.6.32_220.7.1.el6.x86_64.x86_64
~> uname -r
2.6.32-220.7.1.el6.x86_64

w/r,
Kurt
_______________________________________________
lustre-discuss mailing list
lustre-discuss at lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
_______________________________________________
lustre-discuss mailing list
lustre-discuss at lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org


More information about the lustre-discuss mailing list