ZFS and deadlock with {nullfs,NFS}

[Dan Mack]

On Wed, 20 Jun 2007, John Nielsen wrote:

> On Wednesday 20 June 2007 12:30:23 pm Dan Mack wrote:
>> On Wed, 20 Jun 2007, Kris Kennaway wrote:
>>
>> <snip>
>>
>>> 404 at the moment, but look for processes involving zil* in the
>>> backtrace.  I had to disable zil (vfs.zfs.zil_disable=1 tunable) to
>>> prevent low-memory deadlocks on my machines.  Since then it's been
>>> fine.
>>>
>>> You may also wish to use my patches (see the archives) to improve
>>> performance and low-memory behaviour.
>>>
>>> Kris
>>
>> Does someone have these recommended sysctls embodied in an example
>> /boot/loader.conf yet?   Here is mine, does it look reasonable to keep ZFS
>> from running into the kmem_ memory panics?  I have no idea if I found all
>> of your recommendations so it would be nice if they were summarized in one
>> place.
>>
>>   |  # /boot/loader.conf  i386 / 1GB memory / SMP
>>   |  kern.maxvnodes="50000"
>
> Note that kern.maxvnodes has to be set in /etc/sysctl.conf to have any effect.
>
> JN

FWIW ... I'm still experiencing panics; this time with when running cvsup 
to update /usr/ports:

borg# grep -i Panic /var/crash/info*
/var/crash/info.0:  Panic String: kmem_malloc(16384): kmem_map too small: 268419072 total allocated
/var/crash/info.1:  Panic String: kmem_malloc(131072): kmem_map too small: 266354688 total allocated
/var/crash/info.2:  Panic String: kmem_malloc(4096): kmem_map too small: 268435456 total allocated
/var/crash/info.3:  Panic String: kmem_malloc(28672): kmem_map too small: 268435456 total allocated


Anyone ideas on other tunables that might keep the system running?  I'm 
not too concerned about performance.   Otherwise I'll have to go spend 
some money on an AMD64 mainboard, memory, and cpu :-(

Dan