== recovery-small test 10a: finish request on server after client eviction (bug 1521) ========================================================== 13:08:44 (1726852124)
ldlm.namespaces.MGC192.168.201.110@tcp.early_lock_cancel=0
ldlm.namespaces.lustre-MDT0000-mdc-ffff8800b6610000.early_lock_cancel=0
ldlm.namespaces.lustre-OST0000-osc-ffff8800b6610000.early_lock_cancel=0
ldlm.namespaces.lustre-OST0001-osc-ffff8800b6610000.early_lock_cancel=0
fail_loc=0x305
fail_loc=0
fail_val=0
ldlm.namespaces.MGC192.168.201.110@tcp.early_lock_cancel=1
ldlm.namespaces.lustre-MDT0000-mdc-ffff8800b6610000.early_lock_cancel=1
ldlm.namespaces.lustre-OST0000-osc-ffff8800b6610000.early_lock_cancel=1
ldlm.namespaces.lustre-OST0001-osc-ffff8800b6610000.early_lock_cancel=1
/home/green/git/lustre-release/lustre/tests/test-framework.sh: line 4072: /mnt/lustre/recon: Input/output error
Connected clients: oleg110-client.virtnet
/mnt/lustre has perms 0777 OK
[  445.588655] Lustre: ll_ost00_004: service thread pid 18852 was inactive for 40.029 seconds. Watchdog stack traces are limited to 3 per 300 seconds, skipping this one.
[  443.924661] Lustre: ll_ost00_002: service thread pid 8413 was inactive for 40.043 seconds. The thread might be hung, or it might only be slow and will resume later. Dumping the stack trace for debugging purposes:
[  441.364661] Lustre: mdt00_002: service thread pid 7071 was inactive for 40.063 seconds. The thread might be hung, or it might only be slow and will resume later. Dumping the stack trace for debugging purposes:
[  513.400718] Lustre: mdt00_002: service thread pid 7071 completed after 112.099s. This likely indicates the system was overloaded (too many service threads, or not enough hardware resources).