Re: threads/136345: Recursive read rwlocks in thread A cause deadlock with write lock in thread B

[ Available lists | Index of freebsd-threads | Month of Jul 2009 | Week of 16 Jul 2009 | Raw email | View thread | Wrap long lines | Reply | Tag ]
From
Nick Esborn <nick@desert.net>
Date
16 Jul 2009 17:56:58
Subject
Re: threads/136345: Recursive read rwlocks in thread A cause deadlock with write lock in thread B
Message-ID
C3A90BDA-0B42-4CDA-9193-AA27815CFE3B@desert.net

In reply to

[ Hide this part ]
 
On Jul 16, 2009, at 10:53 AM, Attilio Rao wrote:

> 2009/7/16 Nick Esborn <nick@desert.net>:
>>
>>
>> KDB, DDB, SCHED_ULE, and PREEMPTION are already turned on. I will
>> try
>> FULL_PREEMPTION, INVARIANT_SUPPORT, INVARIANTS, and WITNESS, but
>> when I
>> first upgraded to 8.0, this server was unable to handle its
>> workload with
>> the INVARIANTS and WITNESS options turned on.
>
> What do you mean with 'unable'? What was happening precisely?

System time would rise during periods of peak demand, and the system
would quickly fall behind on its workload of queries.

However, I have some hardware I can dedicate to this, and only run the
one MySQL data set which exhibits this problem. That should be enough
of a workload reduction to allow the server to keep up even with all
the above options turned on.

-nick

>
>> Also, it can take a while for it to become clear that the deadlock
>> has
>> occurred -- usually our monitoring picks it up when replication falls
>> behind. So it may be 15-20 minutes after the deadlock that I am
>> able to run
>> the above db commands. Of course the thread will still be
>> deadlocked.
>> Hopefully that doesn't reduce the value of the data obtained.
>
> It should be still fine.
>
> Thanks,
> Attilio
>
>
> --
> Peace can only be achieved by understanding - A. Einstein

--
nick@desert.net - all messages cryptographically signed



[ Show this part (application/pgp-signature) ]

Elapsed time: 0.141 seconds