[LU-11983] operation mds_connect to node failed: rc = -52 Created: 20/Feb/19  Updated: 02/Mar/19  Resolved: 28/Feb/19

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.10.6
Fix Version/s: None

Type: Bug Priority: Major
Reporter: Mahmoud Hanafi Assignee: Hongchao Zhang
Resolution: Fixed Votes: 0
Labels: None

Attachments: File mds.52.debug.gz     File oss.52.debug.gz    
Severity: 2
Rank (Obsolete): 9223372036854775807

 Description   

Upgrading from 2.10.5 to 2.10.6 we are seeing these messages about every 7 mins.

Feb 20 15:33:39 nbp8-oss22 kernel: [107794.374975] LustreError: 11-0: nbp8-MDT0000-lwp-OST007d: operation mds_connect to node 10.151.27.60@o2ib failed: rc = -52
Feb 20 15:33:39 nbp8-oss22 kernel: [107794.411340] LustreError: Skipped 299 previous similar messages

I tried to get debug dump after the error. Its attached to the case. But that errors doesn't show in the debug logs.

 

This is from dmesg

[113475.927136] Lustre: 33169:0:(mdt_handler.c:5340:mdt_connect_internal()) nbp8-MDT0000: client nbp8-MDT0000-lwp-OST0064_UUID does not support ibits lock, either very old or an invalid client: flags 0x2041401043000020
[113475.990096] Lustre: 33169:0:(mdt_handler.c:5340:mdt_connect_internal()) Skipped 4911 previous similar messages


 Comments   
Comment by Mahmoud Hanafi [ 21/Feb/19 ]

Is this a dup/related to LU-8402

 

Comment by Peter Jones [ 22/Feb/19 ]

Hongchao

Could you please advise?

Peter

Comment by Hongchao Zhang [ 22/Feb/19 ]

Hi Mahmoud,

Is the patch https://review.whamcloud.com/33977 in LU-11056 included in it?

Comment by Peter Jones [ 22/Feb/19 ]

No - https://github.com/jlan/lustre-nas/commits/nas-2.10.6?before=ac2bb5aa16b1c18e7ca014c6a5037717528755d5+35 

Comment by Jay Lan (Inactive) [ 22/Feb/19 ]

Yes. The nas-2.10.6 branch does include that commit. Peter's pointer did not go back enough. At end of the page if you press "Older" you would go next page and the commit would be there.

Comment by Mahmoud Hanafi [ 22/Feb/19 ]

Jay,

Does your last build 2.10.6-2nas include patch 33977?

Comment by Jay Lan (Inactive) [ 22/Feb/19 ]

Mahmoud,

Both 2.10.6-1nas and 2.10.6-2nas include commit #33977.

Comment by Hongchao Zhang [ 25/Feb/19 ]

Hi Jay,

the patch #33977 should be used along with the patch #34027 in LU-8402.
the branch nas-2.10.6-1nas doesn't contain the patch from LU-8402, but nas-2.10.6 contains it.
which branch were you using when this issue occurred?

Thanks!

Comment by Jay Lan (Inactive) [ 25/Feb/19 ]

It was 2.10.6-1nas that Mahmoud used in testing.

It is good to know that LU-8402 would address the issue. Thanks!

Comment by Mahmoud Hanafi [ 28/Feb/19 ]

This can be closed.

Comment by Peter Jones [ 28/Feb/19 ]

ok - thanks

Generated at Sat Feb 10 02:48:38 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.