[OpenAFS-devel] 1.4.0-rc4 weirdness

Christopher D. Clausen cclausen@acm.org
Tue, 1 Nov 2005 11:19:10 -0600


Robert Banz wrote:
>> Ok, don't know if this points to something that has already been
>> fixed
>> in some of the newer rc's, but here goes.
>>
>> I just had 3 fileservers 'busy out' on me in the space of a few
>>  minutes. Each of them was running the 1.4.0-rc4 fileserver and had
>> been up for
>> about 45 days each.
>>
>> I've attached the output of rxdebug on one of the fileservers.  I
>> didn't
>> see anything out of the ordinary, with the exception of the
>> tell-tale:
>>
>> 132 calls waiting for a thread
>> 2 threads are idle
>>
>> of utter total doom.

I had a similar thing happen yesterday with a mix of rc8 and rc7 
fileservers (on solaris 10 sparc.)  I had to pkill -9 the fileserver 
process on our main server to get the entire cell back up.  I figured it 
was a problem that was fixed in rc8 so I did not bother to report the 
issue at the time nor do I have rxdebug traces.  I will be more diligent 
in the future.

Could this possibly be caused by the daylight saving time change over 
the weekend?

<<CDC
-- 
Christopher D. Clausen
ACM@UIUC SysAdmin