[rabbitmq-discuss] A rabbitmq error in my cluster

[Simon MacMullen]

On 11/01/2014 06:54, Si-li Liu wrote:
> I found my rabbitmq node is down in the cluster.

There are two different issues here.

> =ERROR REPORT==== 9-Jan-2014::03:17:27 ===
> ** Generic server rabbit_disk_monitor terminating
> ** Last message in was update

The problem here is that the disk space monitor can't parse the output 
of "/bin/df -kP /var/lib/rabbitmq/mnesia/". What does that command 
produce on your machine? What OS are you running?

> After a while, the log reports
>
> =ERROR REPORT==== 9-Jan-2014::03:20:15 ===
> Mnesia(rabbitmq at localhost): ** ERROR ** (core dumped to file:
> "/var/lib/rabbitmq/MnesiaCore.rabbitmq at localhost_1389_237615_166554")
>   ** FATAL ** {error,{"Cannot rename disk_log file",latest_log,
>
>   "/var/lib/rabbitmq/mnesia/rabbitmq at localhost/PREVIOUS.LOG",
>                       {log_header,trans_log,"4.3","4.8",rabbitmq at localhost,
>                                   {1389,237615,128040}},

This is more serious - the "Cannot rename disk_log file" is causing real 
problems. Could something be deleting or moving files in 
/var/lib/rabbitmq/? Or changign their permissions?

Cheers, Simon