[OpenSIPS-Users] opensipsctl fifo commands hanging after an opensips service restart

[Răzvan Crainea]

I suspect that there was another opensipsctl process that was using the 
same fifo file as the new one - these are randomly generated and you 
might have run into a conflict.
Since you killed all the hung processes, there should be no conflicts, 
unless you get hung processes once again - if this happens, try to see 
if there are any overlapping fifo files (using losf or something similar).

Best regards,
Răzvan

On 12/17/19 1:21 PM, solarmon wrote:
> The issue was resolved, but not diagnosed.
> 
> I noticed that there were lots of opensipsctl process in the task list. 
> I have Nagios NCPA agent running scripts that uses opensipsctl and these 
> all seem to be hanging.
> 
> I killed all the opensipsctl processes using pkill and found that I can 
> now use opensipsctl fifo commands without it hanging. Note, I also had 
> to restart the NCPA listener service for Nagios monitoring to work 
> properly again.
> 
> I still would like to understand how it got into this situation in the 
> first place.
> 
> 
> On Tue, 17 Dec 2019 at 10:09, solarmon <solarmon at one-n.co.uk 
> <mailto:solarmon at one-n.co.uk>> wrote:
> 
>     Hi,
> 
>     I have an issue with opensipsctl command hanging when using fifo
>     commands.
> 
>     This occurred after I issued a 'systemctl reload opensips' command.
>     Actually, I subsequently had to issue a 'systemctl restart opensips'
>     command since the opensips service had stopped working.
> 
>     Since then, when using opensipsctl fifo command such as:
> 
>       opensipsctl fifo dlg_list_sharing_tags
> 
>     seems to hang. When I break out of it with CTRL-C I get the
>     following message:
> 
>     ^Crm: cannot remove ‘/tmp/osips_rply_bf3d191f’: No such file or
>     directory
> 
>     I noticed that "opensipsctl dispatcher show" does not hang, but
>     "opensipsctl dispatcher dump" does hang.
> 
>     Note: this is part of an opensips cluster (using clusterer) and the
>     same is happening with both opensips nodes. In fact, the issue is
>     happenning on three different sites that had the opensips nodes
>     service restart after a config change. (The config change was only
>     to change the fr_inv_timeout from 20 to 60).
> 
>     Please advise why this is happening and how to resolve it.
> 
> 
> _______________________________________________
> Users mailing list
> Users at lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
> 

-- 
Răzvan Crainea
OpenSIPS Core Developer
   http://www.opensips-solutions.com