Email Alerts delay

[sateesh]

Hi Team,

Huge time gap between Event trigger and Message actions. because of this alert emails are delivering after long time.

Ex:
Event list time:
2016-12-21 08:49:00 PROBLEM 20m 2s 20m 2s Yes (1)

Message actions time:

2016-12-21 09:01:33 Email sent host1 ([email protected]) Subject:
PROBLEM: Free Disk Space is critical on volume I: : 16.84 % alert on host "host1 "


I have checked SMTP server it working fine no queue and load in SMTP server.


Kindly find zabbix_server performance data.

Zabbix value cache misses 2016-12-21 09:27:480 vps -0.03 vps
Zabbix value cache hits 2016-12-21 09:27:452.8 Kvps +702.49 vps
Zabbix value cache, % free 2016-12-21 09:27:4089.45 % +0.02 %
Zabbix trend write cache, % free 2016-12-21 09:28:0695.83 %-
Zabbix text write cache, % free 2016-12-21 09:27:55100 %-
Zabbix queue over 10m 2016-12-21 09:27:516106 +728
Zabbix queue 2016-12-21 09:27:529739 +827
Zabbix history write cache, % free 2016-12-21 09:27:54 100 %-
Zabbix configuration cache, % free 2016-12-21 09:27:53 94.32 %-
Zabbix busy unreachable poller processes, in % 2016-12-21 09:27:50 30.54 %+14.27 %
Zabbix busy trapper processes, in % 2016-12-21 09:27:49 33.28 %-4.17 %
Zabbix busy timer processes, in % 2016-12-21 09:27:48 0.34 %+0.01 %
Zabbix busy self-monitoring processes, in % 2016-12-21 09:27:46 0 %-
Zabbix busy proxy poller processes, in % 2016-12-21 09:27:45 11.02 %+0.03 %
Zabbix busy poller processes, in % 2016-12-21 09:27:44 15.32 %-0.68 %
Zabbix busy icmp pinger processes, in % 2016-12-21 09:27:40 22.78 %-0.33 %
Zabbix busy http poller processes, in % 2016-12-21 09:27:39 0.05 %-0.01 %
Zabbix busy housekeeper processes, in % 2016-12-21 09:27:38 100 %-
Zabbix busy history syncer processes, in % 2016-12-21 09:27:37 11.66 %+3.09 %
Zabbix busy escalator processes, in % 2016-12-21 09:27:36 100 %-
Zabbix busy discoverer processes, in % 2016-12-21 09:27:35 0 %-
Zabbix busy db watchdog processes, in % 2016-12-21 09:27:34 0.03 %-
Zabbix busy configuration syncer processes, in % 2016-12-21 09:27:33 19.51 %+5.06 %
Zabbix busy alerter processes, in % 2016-12-21 09:27:32 0 %-0.01 %
Values processed by Zabbix server per second 2016-12-21 09:28:06 1393.01-148.94

Kindly help me how to fix the issue.

Thanks & Regards
Sateesh

[DmitryL]

Check how bussy your escalator is.
Check your actions, maybe you have too sensitive conditions and they are flooding escalations table.
If you have 3.0 zabbix - maybe you need to increase count of escalators.

[sateesh]

hi

i am useing zabbix 2.4 and escalator process is always 100%, but it was not consuming cpu power, is there a way to increase escalator process.

Thanks & Regards
Sateesh

[DmitryL]

Not in 2.4 version unfortunately.
Please show result from database
select status, count(*) from escalations group by status;

[sateesh]

Please find out of escalations count status.

select status, count(*) from zabbix.escalations group by status;

STATUS COUNT(*)
---------- ----------
2 159323
0 11145

Thanks & Regards
Sateesh

[DmitryL]

Without deep debugging i would say
Review your actions. Make sure to generate only needed information.
truncate table escalations;
Also check alerts table.
select count(*) from alerts where status 0;

[sateesh]

Hi,

SQL> select count(*) from zabbix.alerts where status=0;

COUNT(*)
----------
0

Thanks you for the help, any recommendations.

Thanks & Regards
Sateesh