[LU-3300] Restore missing proc information for LMT

• Type: Task
• Resolution: Unresolved
• Priority: Major
• Fix Version/s: Lustre 2.4.0
• Affects Version/s: Lustre 2.4.0
• Assignee: Emoly Liu
• Reporter: Christopher Morrone (Inactive)

Description

Lustre's proc seems to have had a number of regressions. LMT's ltop is no longer able to find many of the values it used to display.

In particular, brw_stats from obdfilter is gone, and does not appear to have been replaced after the OSD work. At minimum, that was used by ltop to report the number of bulk rpcs handled.

The MDS display is also missing a number of values.

We don't necessarily need to put them back exactly how they were before, but we need to export them in some way that will make them usable for folks.

It would be best to decide on interfaces before 2.4.0 is locked in.

Issue Links

• duplicates: LU-3106 create symlinks in procfs from ofd to osd
• is related to: LU-3355 proc regression testing
• is related to: LU-2396 ofd brw_stats not maintained
• is related to: LU-3296 fs/lustre/mdt/*/md_stats not showing any stats after remount
• is related to: LU-2096 name ofd device type "ofd" instead of "obdfilter"
• is related to: LU-4259 No brw_stats on ZFS

Comments

[Emoly Liu]

Chris, is there any more work you would like us to do for this ticket? Or can we mark it as resolved? Thanks.

[Emoly Liu]

Chris, I add a symlink of osd brw_stats to lod. Hope that can meet your requirement.
Please see http://review.whamcloud.com/7663 .

[Peter Jones]

Emoly

Could you please see what work remains on this ticket?

Thanks

Peter

[Christopher Morrone (Inactive)]

The problem that you fixed we hadn't even seen yet. We did then see it. Now we don't.

So now brw_stats, or something equivalent just needs to be added, I believe.

[John Hammond]

Chris, any updates here? We were waiting on a response from LLNL.

[Robert Read]

With that patch landed I'm getting metadata in LMT again, thanks! Still waiting to hear from LLNL...

[John Hammond]

The patch for the osd statfs proc handlers has landed to master. But I'm leaving this ticket open waiting for a response from LLNL.

[John Hammond]

Please see http://review.whamcloud.com/6385.

[John Hammond]

Robert, I broke the osd statfs handlers. A patch is forthcoming.

[Robert Read]

I noticed with recent 2.4 builds that lmt is failing to capture metrics on the MDS because several files are empty, however this worked in the 2.4.63 builds that I was using for my LUG testing, so this is a recent regression. These files are in both the lod and osd-ldiskfs directories and they're empty in both:

[ec2-user@mds0 ~]$ head /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/*
==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/activeobd <==
8

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/blocksize <==

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/desc_uuid <==
scratch-MDT0000-mdtlov_UUID

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/filesfree <==

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/filestotal <==

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/kbytesavail <==

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/kbytesfree <==

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/kbytestotal <==

==> /proc/fs/lustre/lod/scratch-MDT0000-mdtlov/numobd <==
8