I have a DDB session open to a crashed ZFS server

Nikolay Denev ndenev at gmail.com
Wed Oct 17 14:57:11 UTC 2012


On Oct 17, 2012, at 5:35 PM, Andriy Gapon <avg at freebsd.org> wrote:

> on 17/10/2012 10:04 Nikolay Denev said the following:
>> I'm running with the patch from here : http://thread.gmane.org/gmane.os.freebsd.devel.file-systems/16000/focus=16017
>> And there were no deadlocks since it's applied.
>> If you're hitting the same issue as I was, this should help.
> 
> I am going to commit that change soon-ish, if nobody expressly objects to it.
> 
> BTW, Nikolay, is your system (that used to deadlock) amd64?
> Could you please capture 'sysctl vm | fgrep kmem' output when it is under load?
> 
> -- 
> Andriy Gapon

Yep, It's a amd64, currently I'm doing some iozone throughput testing over NFS and I can say it's loaded :

last pid:  9818;  load averages: 13.26, 13.98, 15.14                                                                                  up 4+00:09:55  16:56:10
44 processes:  1 running, 43 sleeping
CPU:  0.2% user,  0.1% nice, 63.5% system,  4.0% interrupt, 32.1% idle
Mem: 193M Active, 1689M Inact, 163G Wired, 70M Cache, 22G Free   
ARC: 147G Total, 6884M MRU, 128G MFU, 2824K Anon, 8674M Header, 3896M Other

vm.kmem_map_free: 46525120512                   
vm.kmem_map_size: 149978828800                  
vm.kmem_size_scale: 1
vm.kmem_size_max: 329853485875
vm.kmem_size_min: 0
vm.kmem_size: 199974834176



More information about the freebsd-fs mailing list