We do not have a cluster. Housekeeping fixed the alert log delay, but not devices_perftimes. What part of housekeeping takes care of devices_perftimes ?
Regards Fredrik
-----Ursprungligt meddelande----- Från: observium [mailto:observium-bounces@observium.org] För Stef Renders Skickat: den 27 oktober 2016 15:57 Till: Observium Network Observation System observium@observium.org Ämne: Re: [Observium] Deleting device problems!
We use a galera cluster and getting locks on these tables are impossible on large installations. Even with housekeeping, so our housekeeping is done with a scheduled task to drop or rename the table & create a new one each day. Same goes for syslog:
CREATE DEFINER=`root`@`%` EVENT `devices_perftimes_cleanup` ON SCHEDULE EVERY 1 DAY STARTS '2016-09-09 01:00:00' ON COMPLETION PRESERVE DISABLE ON SLAVE COMMENT '' DO TRUNCATE devices_perftimes
We used to monitor our citrix netscaler farms with citrix command center, their way of fixing this was by having a table for each day (or week, or month) for collecting data. Could this ever be implemented as well in the future ? Especially for syslog.
Kind regards
-----Original Message----- From: observium [mailto:observium-bounces@observium.org] On Behalf Of Moerman, Maarten Sent: donderdag 27 oktober 2016 15:48 To: Observium Network Observation System observium@observium.org Subject: Re: [Observium] Deleting device problems!
I think it helps if you setup housekeeping first in this case, it will clean up your install, and will fix your issue.
Maarten
Sent from my iPhone
On 27 Oct 2016, at 15:19, Svensson Fredrik A fredrik.a.svensson@rjl.se wrote:
Hi there! When we are deleting a device we struggle with the time it takes, I have done a debug and this seems to be the problem:
SQL[DELETE FROM `devices_perftimes` WHERE `device_id` = '593'] ROWS[-1] SQL RUNTIME[97.04324698s] ERROR[Error in query: (The total number of locks exceeds the lock table size) 1206]
Can you help me with what I can do to fix this problem?
This also take some time: SQL[DELETE FROM `alert_log` WHERE `device_id` = '593'] ROWS[11480] SQL RUNTIME[15.23252392s]
Regards Fredrik
-----Ursprungligt meddelande----- Från: observium [mailto:observium-bounces@observium.org] För greg lambert Skickat: den 25 oktober 2016 09:41 Till: Observium Network Observation System observium@observium.org Ämne: [Observium] Acme Packet SBC 3820
Good Day,
I just loaded my SBC into observium - it doesn not graph the voice utilization. I was hoping to use it to graph voice channel useage etc.
Is there any suggested way to get this graphed?
Regards,
Greg
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
_______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
_______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium