Noel Dellofano wrote:

> hmm.. you should be able to unload and load the zfs module if  
> everything is unmounted/exported and all the consumers of libzfs 
> aren't  running, this includes fmd as well as zoneadm.  Did you try 
> disabling  zoneadm as well?
> You can also check that all three of these variables should be zero:
>     zfs_active_fs_count
>     spa_active_count
>     and zvol_minors
>

One other step that i sometimes have to do, which may not be possible 
for you, is destorying all pools on the system.  Then modunload'ing 
works fine.

eric

> Noel :)
>
>
> ************************************************************************ 
> **
>
> "Question all the answers"
> On May 10, 2006, at 12:09 AM, Darren J Moffat wrote:
>
>> Is it supposed to be possible to unload the zfs module ?
>>
>> There are no pools configured on this machine or they have all been  
>> exported.
>>
>> It doesn't seem to work.  I'd rather not have to reboot time and 
>> time  again just to test new bits (at least the ones that don't 
>> cause  panics!).
>>
>> I saw that fmd had /dev/zfs open so I did:
>>     svcadm disable -t fmd
>>
>> That resulted in fuser not reporting that anyone had /dev/zfs open 
>> but  still the module wouldn't unload.
>>
>> How do other ZFS developers deal with getting new kernel bits on to  
>> the system to test ?
>>
>> -- 
>> Darren J Moffat
>> _______________________________________________
>> zfs-code mailing list
>> zfs-code at opensolaris.org
>> http://opensolaris.org/mailman/listinfo/zfs-code
>
>
> _______________________________________________
> zfs-code mailing list
> zfs-code at opensolaris.org
> http://opensolaris.org/mailman/listinfo/zfs-code



Reply via email to