[erlang-questions] Process hanging with huge message queue , with current_function, {mnesia_tm, get_dirty_reply, 2}

[Ahmed Omar]

Sorry i expressed it wrong. i mean it's on top of the list, which is the 1st
on the front (1st message arrived). but the replies are from only some of
the nodes, not all. And get_dirty_reply will wait for a specific node (no
matter if replies comes also from other nodes that are involved in the
commit)

On Fri, May 27, 2011 at 6:49 PM, Ahmed Omar <spawn.think@REDACTED> wrote:

> Actually looking into the message queue, the replies are there but they are
> at the top of list (which is the end of the queue). So i guess the process
> doesn't have enough time to scan the whole queue to pick up the results.
>
>
> On Fri, May 27, 2011 at 6:30 PM, Ahmed Omar <spawn.think@REDACTED> wrote:
>
>> Hi everyone,
>> I have a cluster of nodes, and on some of the nodes i have processes
>> hanging with current_function,{mnesia_tm,get_dirty_reply,2}
>> Checking the code of mnesia_tm:get_dirt_reply/2 , this function will wait
>> forever. It checks every 1 second if the node that is supposed to reply is
>> still a member of running_db_nodes. If yes, it will try to receive either
>> the result, exit message or mnesia down message  . And it will keep on doing
>> that forever.
>> So it seems to cover all cases, however the process ends up in an infinite
>> loop.
>> Any ideas?
>>
>> Erlang version : R13B03 on Linux Debian Lenny 5.0.6
>>
>> --
>> Best Regards,
>> - Ahmed Omar
>> http://nl.linkedin.com/in/adiaa
>> Follow me on twitter
>> @spawn_think <http://twitter.com/#!/spawn_think>
>>
>>
>
>
> --
> Best Regards,
> - Ahmed Omar
> http://nl.linkedin.com/in/adiaa
> Follow me on twitter
> @spawn_think <http://twitter.com/#!/spawn_think>
>
>


-- 
Best Regards,
- Ahmed Omar
http://nl.linkedin.com/in/adiaa
Follow me on twitter
@spawn_think <http://twitter.com/#!/spawn_think>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20110527/25e84e36/attachment.htm>