[OpenAFS] rxkad error=19270408

[Derrick Brashear]

On Tue, Apr 21, 2009 at 11:17 AM, Mircea Ciocan
<mircea.ciocan@cmosvision.com> wrote:
>
>
> Derrick Brashear wrote:
>>
>> On Tue, Apr 21, 2009 at 11:10 AM, Mircea Ciocan
>> <mircea.ciocan@cmosvision.com> wrote:
>>
>>>
>>> Derrick Brashear wrote:
>>>
>>>>
>>>> On Tue, Apr 21, 2009 at 11:06 AM, Mircea Ciocan
>>>> <mircea.ciocan@cmosvision.com> wrote:
>>>>
>>>>
>>>>>
>>>>> Derrick Brashear wrote:
>>>>>
>>>>>
>>>>>>
>>>>>> On Tue, Apr 21, 2009 at 10:56 AM, Ted Creedon
>>>>>> <tcreedon@easystreet.net>
>>>>>> wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>
>>>>>>> OpenAFS crashed my 8 processor Intel i7 using 16% of one cpu and 10=
0%
>>>>>>> of
>>>>>>> a
>>>>>>> single processor system too.
>>>>>>>
>>>>>>> The problem is time consuming due to the cold boots and the reset
>>>>>>> button...
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> Are you running 1.4.10? (The correct answer is yes. If you give the
>>>>>> wrong answer, fix it and try again)
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>> Yes, I'm running this on an OpenSUSE 11.1 x64:
>>>>> openafs-authlibs-1.4.10-13.2
>>>>> openafs-1.4.10-13.2
>>>>> openafs-client-1.4.10-13.2
>>>>> openafs-server-1.4.10-13.2
>>>>> openafs-kmp-default-1.4.10_2.6.27.21_0.1-13.2
>>>>> openafs-krb5-mit-1.4.10-13.2
>>>>>
>>>>> IMHO, no matter what kerberos key
>>>>>
>>>>>
>>>>
>>>> The kerberos key isn't causing the problem.
>>>>
>>>>
>>>>
>>>>>
>>>>> =A0b..s is happening this should not produce
>>>>> this miserable kernel loop that kills the most powerful machines
>>>>> available
>>>>> today, it either should have some slower cadence so that eventually
>>>>> some
>>>>> could stop the AFS processes or it should give up after some time, I
>>>>> this
>>>>> regard I consider this behavior a bug.
>>>>>
>>>>>
>>>>
>>>> it probably is.
>>>>
>>>> dumb question: are you using dynroot?
>>>>
>>>>
>>>
>>> Actually yes, I do, and it works like a charm, is that bad now ?!?!?!?
>>>
>>
>> sure. it means you're starting afs with no servers available to serve
>> root.afs.
>>
>> it's a bug. there's a ticket open for it. but it's easily avoidable:
>> don't do that.
>>
>
> Oh well, fashion changes, some while ago NOT USING dynroot was bad and
> obsolete, now is vice-versa ;), good to know that this was what was causi=
ng
> it
> but then again dynroot is quite convenient, I hope the bug gets fixed
> sometime.

wait, you're saying the problem stopped when you *stopped* using dynroot?
that's a new one.

--=20
Derrick