[rabbitmq-discuss] Rabbitmq 3.2.2 Keeps crashing in different ways :(

Emile Joubert emile at rabbitmq.com
Wed Jan 15 10:40:35 GMT 2014


Hi Peter,

Thanks for the detailed information. I've not been able to access the
large log, but find some interim suggestions/observations inline:

On 14/01/14 15:15, Peter Kopias wrote:

>  He acknowledged this as a real bug
>  ("{gm,find_prefix_common_suffix,2,[]}" in the stacktrace)

It may well end up being related to this error:

> "Absurdly large distribution output data buffer (2427628186 bytes) passed.
> Aborted (core dumped)"

Did these errors appear  near the same time?



> http://corpweb.dunakanyar.net/petrosdump/rabbitmqoutofmemory20140113.tar.bz2

Can you please confirm the hash of this file? I get a bunzip2 data
integrity error when decompressing.


>  I don't think high availability has a part of "if (last_err==E_NOSPC)
> die();"

You probably need a larger disk free limit.

>  I've got 16Gbytes of ram in each node. The watermark is set to 7.8Gbyte,
> so we should never have memory problems.

You could try a lower vm_memory_high_watermark. If you have very fast
publishers then it is possible for available RAM to be consumed faster
than the broker can react.


>  - Which is the "pivotal preferred stable version"? And which is the
> "pivotal preferred stable linux distribution"?

The latest version is generally preferred.

>  - Does rabbitmq have a ticket system somewhere?
>  - Where to report bugs to?

To the mailing list, as you have done. Thank you.


>   As these are testing machines currently it is possible to give a
> developer access to them if it helps.

Thanks, but it is even more useful to be able to reproduce issues. The
error starting "Absurdly large distribution" has been reproduced. Please
see if my suggestions solve the other problems.



-Emile




More information about the rabbitmq-discuss mailing list