Zebra/Plack/Timeout Issue

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

Zebra/Plack/Timeout Issue

rfblanchard
I just started using plack yesterday and I ran the commands below to get started. Everything was fine. The next morning I found that plack daemon was not running. I needed to rerun both:
> sudo koha-plack --enable <instancename> 
> sudo koha-plack --start <instancename>
Also, for a while now each morning I need to re-start zebra as it seems to stop on a daily basis
> sudo koha-start-zebra peischool

Potentially related is that timeout preference is not respecting the value from administration. I set it to both 7d and 604800. No matter what I choose I need to login again daily to koha administration site.

Rebooting does not seem to be the cause of these problems. I just ran 'sudo reboot' and the server came back with all apache/mysql and daemons were running before and after the reboot. Also it didn't cause my session to timeout, which is good.

I'm Running Koha 16.05.02 and Debian 8.5
Any Ideas to why the zebra and plack continue to need manual intervention daily and why timeout is not respecting the value from administration?
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

Jonathan Druart
Hi, try to swith the system preference SessionStorage to mysql instead
of memcached

2016-10-04 19:31 GMT+02:00 rfblanchard <[hidden email]>:

> I just started using plack yesterday and I ran the commands below to get
> started. Everything was fine. The next morning I found that plack daemon was
> not running. I needed to rerun both:
>> sudo koha-plack --enable <instancename>
>> sudo koha-plack --start <instancename>
> Also, for a while now each morning I need to re-start zebra as it seems to
> stop on a daily basis
>> sudo koha-start-zebra peischool
>
> Potentially related is that timeout preference is not respecting the value
> from administration. I set it to both 7d and 604800. No matter what I choose
> I need to login again daily to koha administration site.
>
> Rebooting does not seem to be the cause of these problems. I just ran 'sudo
> reboot' and the server came back with all apache/mysql and daemons were
> running before and after the reboot. Also it didn't cause my session to
> timeout, which is good.
>
> I'm Running Koha 16.05.02 and Debian 8.5
> Any Ideas to why the zebra and plack continue to need manual intervention
> daily and why timeout is not respecting the value from administration?
>
>
>
> --
> View this message in context: http://koha.1045719.n5.nabble.com/Zebra-Plack-Timeout-Issue-tp5905919.html
> Sent from the Koha-general mailing list archive at Nabble.com.
> _______________________________________________
> Koha mailing list  http://koha-community.org
> [hidden email]
> https://lists.katipo.co.nz/mailman/listinfo/koha
_______________________________________________
Koha mailing list  http://koha-community.org
[hidden email]
https://lists.katipo.co.nz/mailman/listinfo/koha
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

Chris Cormack-7
In reply to this post by rfblanchard
Jonathan has covered the session timeouts, as for the restarting, you should only ever have to do the enable once.

Re the --start can you check if the time they are stopping is when logrotate is running? Or some other cron job

Chris

On 5 October 2016 6:31:31 AM NZDT, rfblanchard <[hidden email]> wrote:

>I just started using plack yesterday and I ran the commands below to
>get
>started. Everything was fine. The next morning I found that plack
>daemon was
>not running. I needed to rerun both:
>> sudo koha-plack --enable <instancename>
>> sudo koha-plack --start <instancename>
>Also, for a while now each morning I need to re-start zebra as it seems
>to
>stop on a daily basis
>> sudo koha-start-zebra peischool
>
>Potentially related is that timeout preference is not respecting the
>value
>from administration. I set it to both 7d and 604800. No matter what I
>choose
>I need to login again daily to koha administration site.
>
>Rebooting does not seem to be the cause of these problems. I just ran
>'sudo
>reboot' and the server came back with all apache/mysql and daemons were
>running before and after the reboot. Also it didn't cause my session to
>timeout, which is good.
>
>I'm Running Koha 16.05.02 and Debian 8.5
>Any Ideas to why the zebra and plack continue to need manual
>intervention
>daily and why timeout is not respecting the value from administration?
>
>
>
>--
>View this message in context:
>http://koha.1045719.n5.nabble.com/Zebra-Plack-Timeout-Issue-tp5905919.html
>Sent from the Koha-general mailing list archive at Nabble.com.
>_______________________________________________
>Koha mailing list  http://koha-community.org
>[hidden email]
>https://lists.katipo.co.nz/mailman/listinfo/koha

--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
_______________________________________________
Koha mailing list  http://koha-community.org
[hidden email]
https://lists.katipo.co.nz/mailman/listinfo/koha
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

rfblanchard
Upgraded to koha 16.05.04 and debian 8.6 today
-my sessionStorage was mysql. I tried to change it to memcached to see if any better.
-I checked and no crontab existed for /root/or the koha user on the server

Not sure how to determine exactly when plack and zebra are being killed.
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

Michael Kuhn
Hi R. F.

> -I checked and no crontab existed for /root/or the koha user on the server

Are you aware of the fact there are not only the crontabs for Linux user
"root" and the one of your Linux user for Koha but also the cronjobs in
the following files?

* /etc/cron.d/koha-common
* /etc/cron.hourly/koha-common
* /etc/cron.daily/koha-common
* /etc/cron.monthly/koha-common

Hope this helps.

Best wishes: Michael
--
Geschäftsführer · Diplombibliothekar BBS, Informatiker eidg. Fachausweis
Admin Kuhn GmbH · Pappelstrasse 20 · 4123 Allschwil · Schweiz
T 0041 (0)61 261 55 61 · E [hidden email] · W www.adminkuhn.ch
_______________________________________________
Koha mailing list  http://koha-community.org
[hidden email]
https://lists.katipo.co.nz/mailman/listinfo/koha
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

Chris Cormack-6
And also the one that I suspect is doing it

/etc/logrotate.d/koha-common

Chris

On 5 October 2016 at 08:20, Michael Kuhn <[hidden email]> wrote:

> Hi R. F.
>
>> -I checked and no crontab existed for /root/or the koha user on the server
>
>
> Are you aware of the fact there are not only the crontabs for Linux user
> "root" and the one of your Linux user for Koha but also the cronjobs in the
> following files?
>
> * /etc/cron.d/koha-common
> * /etc/cron.hourly/koha-common
> * /etc/cron.daily/koha-common
> * /etc/cron.monthly/koha-common
>
> Hope this helps.
>
> Best wishes: Michael
> --
> Geschäftsführer · Diplombibliothekar BBS, Informatiker eidg. Fachausweis
> Admin Kuhn GmbH · Pappelstrasse 20 · 4123 Allschwil · Schweiz
> T 0041 (0)61 261 55 61 · E [hidden email] · W www.adminkuhn.ch
>
> _______________________________________________
> Koha mailing list  http://koha-community.org
> [hidden email]
> https://lists.katipo.co.nz/mailman/listinfo/koha
_______________________________________________
Koha mailing list  http://koha-community.org
[hidden email]
https://lists.katipo.co.nz/mailman/listinfo/koha
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

Jonathan Druart
In reply to this post by rfblanchard
For the timeout, you can also try to switch off SessionRestrictionByIP.

Please keep us in touch if it fixes the issue.

2016-10-04 21:11 GMT+02:00 rfblanchard <[hidden email]>:

> Upgraded to koha 16.05.04 and debian 8.6 today
> -my sessionStorage was mysql. I tried to change it to memcached to see if
> any better.
> -I checked and no crontab existed for /root/or the koha user on the server
>
> Not sure how to determine exactly when plack and zebra are being killed.
>
>
>
> --
> View this message in context: http://koha.1045719.n5.nabble.com/Zebra-Plack-Timeout-Issue-tp5905919p5905923.html
> Sent from the Koha-general mailing list archive at Nabble.com.
> _______________________________________________
> Koha mailing list  http://koha-community.org
> [hidden email]
> https://lists.katipo.co.nz/mailman/listinfo/koha
_______________________________________________
Koha mailing list  http://koha-community.org
[hidden email]
https://lists.katipo.co.nz/mailman/listinfo/koha
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

Magnus Enger
On 5 October 2016 at 08:59, Jonathan Druart
<[hidden email]> wrote:
> For the timeout, you can also try to switch off SessionRestrictionByIP.

When I was having this problem it was unaffected by
SessionRestrictionByIP, but I was able to fix it by setting
SessionStorage = MySQL.

Best regards,
Magnus
_______________________________________________
Koha mailing list  http://koha-community.org
[hidden email]
https://lists.katipo.co.nz/mailman/listinfo/koha
Reply | Threaded
Open this post in threaded view
|

Re: Zebra/Plack/Timeout Issue

rfblanchard
In reply to this post by Chris Cormack-6
I set the SessionStorage back to MySQL as it seems more reliable even through reboots.

Thanks for letting me know about those other crons. I should have known about those.

I see why you suspected logrotate as it shuts down and restarts plack and zebra.
I manually triggered:
sudo logrotate -f /etc/logrotate.d/koha-common
It briefly took those items down but they came back up again no issue.

Typically its been in the morning when I notice the zebra/plack is down. It down yesterday but OK today which is almost worse as I did nothing to fix it. I'll continue to monitor.