[LU-8567] mdc_reint.c:57:mdc_reint()) error in handling -17 encountered on power8 node

• Status: Closed
• Project: Lustre
• Affects Version/s: Lustre 2.9.0
• Type: Bug
• Priority: Major
• Reporter: James A Simmons
• Assignee: Zhenyu Xu
• Resolution: Not a Bug

Environment

Power8 running RHEL7.2 with lustre version 2.8.56

• Severity: 3

Description

During my latest testing after I updated my test file system to the latest as well I found when running a IOR single shared file on one client the job alway fails. Nothing shows up in the kernel logs but I have gathered lustre debug logs to show what the problem is. I have attached those logs here.

Issue Links

• is related to: LU-6387 Add Power8 support to Lustre

Comments

[Peter Jones]

Bobijam

Could you please assist with this issue?

Thanks

Peter

[Zhenyu Xu]

What's the failed job output? Can you strace it?

The log shows that mkdir of "jsimmons" under 0x200000410:0x4:0x0] failed, since it exists.

00000080:00000001:1.0:1472488456.442910:1344:93034:0:(namei.c:1235:ll_mkdir()) Process entered
00000080:00200000:1.0:1472488456.442910:1344:93034:0:(namei.c:1238:ll_mkdir()) VFS Op:name=jsimmons, dir=[0x200000410:0x4:0x0](c0000007eb866c90)
00000080:00000001:1.0:1472488456.442912:1696:93034:0:(namei.c:982:ll_new_node()) Process entered
...
00000080:00000001:1.0:1472488456.444020:1792:93034:0:(namei.c:1008:ll_new_node()) Process leaving via err_exit (rc=18446744073709551599 : -17 : 0xffffffffffffffef)
...
00000080:00000001:1.0:1472488456.444031:1456:93034:0:(namei.c:1249:ll_mkdir()) Process leaving (rc=18446744073709551599 : -17 : ffffffffffffffef)

[James A Simmons]

Sorry this doesn't appear to be the source of the bug. The true source is LU-7321. Its just now that bug appears only on my servers. Sorry for the noise.