pcp
[Top] [All Lists]

Re: [pcp] pmdagfs2: updates

To: Paul Evans <pevans@xxxxxxxxxx>
Subject: Re: [pcp] pmdagfs2: updates
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Tue, 2 Sep 2014 01:04:25 -0400 (EDT)
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <11004827.42146753.1409632312287.JavaMail.zimbra@xxxxxxxxxx>
References: <53FF63C8.4020902@xxxxxxxxxx> <352630399.40713023.1409292921192.JavaMail.zimbra@xxxxxxxxxx> <5404A547.7040805@xxxxxxxxxx> <11004827.42146753.1409632312287.JavaMail.zimbra@xxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: cg+LFltCrf91+3zIZyy8+/y1wNxRlkbuxu3c
Thread-topic: pmdagfs2: updates
Hi Paul,

----- Original Message -----
> ----- Original Message -----
> > [...]
> > I've checked the qa output on my RHEL 6.6, RHEL 6.5, RHEL 7 and Fedora
> > 17 & 20 VM's and don't seem to seeing any issues with "No value(s)
> > available!" with the gfs2.glstats metrics on this end.
> > 
> > Are you still able to reproduce the failure? Also what distribution was
> > it popping up on? I'll continue to keep a close eye on my qa outputs
> > over the next couple of days to see if it does surfaces for me.
> 
> Yep, reproducible for me on RHEL 6.3.  Interestingly its only the glstats
> metrics, the others are all correctly reporting the loop devices as before.
> 
> I'll dig a little deeper into it here too.

The problem seems to be lack of kernel support for the glstats concept,
on this platform.  There are no /sys/kernel/debug/gfs2/loopN/gstats files
just "glocks" on this kernel.

Maybe this part should move into a new, separate test, and we could add a
_notrun() check to it, for the no glstats file case?   The remainder of
test qa/655 appears to be running correctly for this kernel FWIW -> (6.3)
2.6.32-279.46.1.el6.x86_64 #1 SMP Mon May 19 16:16:00 EDT 2014 x86_64.

cheers.

--
Nathan

<Prev in Thread] Current Thread [Next in Thread>