Project

General

Profile

Bug #4690

nfs4: panic while shutting down

Added by Simon Klinkert over 5 years ago. Updated over 5 years ago.

Status:
New
Priority:
Normal
Assignee:
-
Category:
nfs - NFS server and client
Start date:
2014-03-20
Due date:
% Done:

0%

Estimated time:
Difficulty:
Medium
Tags:
needs-triage

Description

With a 1/10 chance I get this pretty panic stack after typing init 6:

Mar 20 14:30:38 kobaia nfs: [ID 168492 kern.warning] WARNING: nfs_idmap_str_uid: Can't communicate with mapping daemon nfsmapid
Mar 20 14:30:45 kobaia unix: [ID 836849 kern.notice] 
Mar 20 14:30:45 kobaia ^Mpanic[cpu5]/thread=ffffff0040d27c40: 
Mar 20 14:30:46 kobaia genunix: [ID 103648 kern.notice] mutex_enter: bad mutex, lp=ffffff9a81af6000 owner=ffffff0040d27c40 thread=ffffff0040d27c40
Mar 20 14:30:46 kobaia unix: [ID 100000 kern.notice] 
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27820 unix:mutex_panic+73 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27890 unix:mutex_vector_enter+367 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27910 nfs:nfs4_ephemeral_unmount_engine+fc ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d279c0 nfs:nfs4_ephemeral_umount+104 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27a30 nfs:nfs4_unmount+d9 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27a60 genunix:fsop_unmount+1b ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27ab0 genunix:dounmount+57 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27b00 genunix:umount2_engine+96 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27b40 nfs:nfs4_ephemeral_record_umount+39 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27bf0 nfs:nfs4_ephemeral_harvest_forest+1db ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27c20 nfs:nfs4_ephemeral_harvester+77 ()
Mar 20 14:30:46 kobaia genunix: [ID 655072 kern.notice] ffffff0040d27c30 unix:thread_start+8 ()

I think this bug is at least one year old.

History

#1

Updated by Marcel Telka over 5 years ago

Please share the crash dump file. Thanks.

#2

Updated by Simon Klinkert over 5 years ago

Sorry, but I can't share the dump because it happened on a customer machine. I can provide some mdb output if you like to know something special.

I will try to reproduce it on a test machine to get a crash dump without customer data. I guess the preconditions are a lot of nfs4 mountpoints (I have > 100).

Also available in: Atom PDF