[Spce-user] Fwd: Issues with web panel crashing Firefox, Mac kernel panic

[Andy Bradford]

On Tue, May 24, 2016 at 9:21 AM, Alex Lutay <alutay at sipwise.com> wrote:

> Dear Andy,
>
> Lets continue a talk in public spce-users@ mailing list,
> Sipwise doesn't support SPCE users directly.
>

No worries - apologies - didn't realize I hadn't replied all at the time.


>
> I have checked boot log (it always said started successfully):
>
> > Sun May 15 14:30:12 2016: [....] Starting NGCP-Panel Webapp: [ ok ]
> > Sun May 22 21:33:59 2016: [....] Starting NGCP-Panel Webapp: [ ok ]
> > Sun May 22 23:30:25 2016: [....] Starting NGCP-Panel Webapp: [ ok ]
>

Probably since the last reboot it has. Haven't had it lock up since we
rebooted after some changes over the weekend. Next time it happens, I'll
look here and share it out.


>
> So, I still have no answer why it did not start in result.
> Maybe 'oom-killer' was involved to kill it due to lack of RAM, try:
> > grep -i oom-killer /var/log/syslog
>

Didn't get anything in return on that command. Will wait for it to happen
again and will try that.

>
> As for 2GB of RAM you have: it is really a minimum to fly on minimal
> settings, You need to use "--low-performance" mode then:
>
> > root at sp1:~# ngcp-toggle-performance-config
> > /usr/sbin/ngcp-toggle-performance-config - tool to adjust sip:provider
> configuration for low/high performance
> >
> >   --help                Display this usage information
> >   --high-performance    Adjust configuration for system with normal/high
> performance
> >   --low-performance     Adjust configuration for system with low
> performance (e.g. VMs)
> >
> > root at sp1:~#
>
> You have high-performance setup by default
> (10 heavy panel threads are starting, running and eating RAM).
> To play with NGCP three threads are more then enough.
> See the option "fastcgi_workers" in /etc/ngcp-config/config.yml
>

Thanks - I did in fact forget about low performance mode. Since we're only
processing a few concurrent calls, and really just have sipwise sitting
between our SIP provider and Asterisk (because of some weird issues with
Asterisk handling inbound calls, and sometimes matching outbound peers for
incoming calls - causing all kinds of fun things to happen), I figured 2GB
would be enough for us in the cloud for the time being. I've taken your
suggestion of enabling low performance and pushed that through now. I'll
see if that helps things.

>
> On 05/24/2016 03:04 PM, Andy Bradford wrote:
> [...skipped...]
>
> --
> Alex Lutay
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sipwise.com/pipermail/spce-user_lists.sipwise.com/attachments/20160524/35372d7c/attachment-0001.html>