[LU-17301] Client mount hangs for several minutes

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.16.0
• Severity: 3
• Assignee: Shaun Tancheff
• Reporter: Tim Day

Description

After patch https://review.whamcloud.com/c/fs/lustre-release/+/51329, client mount hangs for several minutes when using `llmount.sh`. When this patch is reverted, the mount completes in a normal amount of time.

 

Setup:

Kernel - 5.15.94
openZFS - 2.1.99-1735_gdc72c60ec
Distro - CentOS Linux release 8.5.2111

 

Kernel messages:

[  480.629670] Lustre: mdt_rdpg01_001: service thread pid 52665 was inactive for 40.946 seconds. The thread might be hung, or it might only be slow and will resume later. Dumping the stack trace for debugging purposes:
[  480.632948] Pid: 52665, comm: mdt_rdpg01_001 5.15.94 #1 SMP Fri Feb 17 01:05:48 UTC 2023
[  480.634352] Call Trace TBD:
[  500.084741] LustreError: 52665:0:(upcall_cache.c:251:upcall_cache_get_entry()) acquire for key 0: error -110
[  500.086740] Lustre: Mounted lustre-client

The mount hangs for several minutes even after the 'Mounted' message gets printed.

Issue Links

• is related to: LU-16901 Provide l_getidentity_nss identity provider