[rabbitmq-discuss] Rabbitmq server crash and no start more

[Gustavo Aquino]

I got this same problem using MacOS 10.5, that I'm using now, RHE and Debian
with RabbitMQ 1.7 never got it with 1.6.

Look I can reproduce it all the time, in all this environments just do
Crtl+C in console running RabbitMQ-server and try to start again, and booom
the problem appears. I just do it again to show the metrics that you request
me.


====== Fail process.
* ./start-rabbitmq.sh *

+---+   +---+
|   |   |   |
|   |   |   |
|   |   |   |
|   +---+   +-------+
|                   |
| RabbitMQ  +---+   |
|           |   |   |
|   v1.7.0  +---+   |
|                   |
+-------------------+
AMQP 8-0
Copyright (C) 2007-2009 LShift Ltd., Cohesive Financial Technologies LLC.,
and Rabbit Technologies Ltd.
Licensed under the MPL.  See http://www.rabbitmq.com/

node          : rabbit at 189-94-130-143
app descriptor:
/opt/local/lib/rabbitmq/bin/../lib/rabbitmq_server-1.7.0/sbin/../ebin/rabbit.app
home dir      : /opt/local/var/lib/rabbitmq
cookie hash   : diX16VkN3b6UrKJtDPVb1Q==
log           : /opt/local/var/log/rabbitmq/rabbit.log
sasl log      : /opt/local/var/log/rabbitmq/rabbit-sasl.log
database dir  : /opt/local/var/lib/rabbitmq/mnesia/rabbit

starting database             ...done
starting core processes       ...done
starting recovery             ...done
starting persister            ...done
starting guid generator       ...done
starting builtin applications ...done
starting TCP listeners        ...done
starting SSL listeners        ...done

broker running
*^C*
*BREAK: (a)bort (c)ontinue (p)roc info (i)nfo (l)oaded*
*       (v)ersion (k)ill (D)b-tables (d)istribution*
*a*

*$~ gustavoaquino$ ./start-rabbitmq.sh *
Password:

+---+   +---+
|   |   |   |
|   |   |   |
|   |   |   |
|   +---+   +-------+
|                   |
| RabbitMQ  +---+   |
|           |   |   |
|   v1.7.0  +---+   |
|                   |
+-------------------+
AMQP 8-0
Copyright (C) 2007-2009 LShift Ltd., Cohesive Financial Technologies LLC.,
and Rabbit Technologies Ltd.
Licensed under the MPL.  See http://www.rabbitmq.com/

node          : rabbit at 187-26-170-110
app descriptor:
/opt/local/lib/rabbitmq/bin/../lib/rabbitmq_server-1.7.0/sbin/../ebin/rabbit.app
home dir      : /opt/local/var/lib/rabbitmq
cookie hash   : diX16VkN3b6UrKJtDPVb1Q==
log           : /opt/local/var/log/rabbitmq/rabbit.log
sasl log      : /opt/local/var/log/rabbitmq/rabbit-sasl.log
database dir  : /opt/local/var/lib/rabbitmq/mnesia/rabbit

*starting database             ...{"init terminating in
do_boot",{{nocatch,{error,{cannot_start_application,rabbit,{{timeout_waiting_for_tables,[rabbit_user,rabbit_user_permission,rabbit_vhost,rabbit_config,rabbit_listener,rabbit_durable_route,rabbit_route,rabbit_reverse_route,rabbit_durable_exchange,rabbit_exchange,rabbit_durable_queue,rabbit_queue]},{rabbit,start,[normal,[]]}}}}},[{init,start_it,1},{init,start_em,1}]}}
*
*
*
*Crash dump was written to: erl_crash.dump*
*init terminating in do_boot ()*

=== Final fail process

About resources machine, the CPU, Memory and IO increase just 1 or 3 % of
start time, at start time my machine was using 10% of CPU, Memory dont grow
nothing about 20-30mb, and IO just continue in the same way I don't identify
any different in machine status.

To put my rabbit to work again after this test I just delete the schema.DAT
file in /opt/local/var/lib/rabbitmq/mnesia/rabbit, don't remove all the
files.

This is the file size before deleted.

-rw-r--r--   1 rabbitmq  rabbitmq  12K 24 Jan 16:36 schema.DAT


[]s


On Sun, Jan 24, 2010 at 5:07 PM, Matthew Sackman <matthew at lshift.net> wrote:

> Hi,
>
> On Sun, Jan 24, 2010 at 04:42:31PM -0200, Gustavo Aquino wrote:
> > It's works, I delete all rabbit files and schema.DAT and Rabbit come back
> to
> > work, but look that, i do one crtl+C in console and Rabbit crash again.
> >
> > I really would like to know why it's happened I'm working in a very big
> > project that planing to use RabbitMQ and this comportment seriously
> concern
> > us, principally that we will have persistent queues that can be removed.
> >
> > Can you explain more about the problem please ?
>
> Well I would if I could. I'm not sure why what's happening is happening
> for you - I've not really seen it before. If your database is empty(ish)
> then I have no idea why mnesia is taking so long to start up. Can you
> monitor disk activity and CPU activity? - is the startup process
> swamping lots of CPU or disk throughput as it tries to come up? Are you
> running in a clustered scenario? What OS are you using - is this "in the
> cloud" at all, or virtualised in some way?
>
> Normally, with an small database, mnesia should start up very quickly
> indeed, so I am puzzled as to why this timeout is occurring for you.
>
> Matthew
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20100124/32e10324/attachment.htm