Hi Zabbix Experts,
We have a medium-sized setup (I don't know if thats the right characterization)
Here's our Dashboard for you to get some context
Parameter Value Details
Zabbix server is running Yes
localhost:10051
Number of hosts (enabled/disabled/templates)
368 304 / 1 / 63
Number of items (enabled/disabled/not supported)
22022 17752 / 214 / 4056
Number of triggers (enabled/disabled [problem/ok])
5046 1703 / 3343 [284 / 1419]
Number of users (online) 12 3
Required server performance, new values per second
318.21 -
As you can see we just over 300 hosts, 20k items and a dozen users.
The NVP is just over 315 !
What we have been seeing is that when we navigate to "Latest data" page under Monitoring tab...
I see that zabbix is generating a lot of HUGE queries and sending it to our PostgreSQL database.
Our setup for almost (95%) of the items is to collect the metric every minute, store the History data for 7 days and trends data for 365 days.
We have a Screen refresh for each user set at 300 sec. and at any time there are no more than 3 to 5 users looking.
For the following workloads,
0> Is this EXPECTED that we see many such BIG queries ?
1> I also see that there's a select count(*) from history query which is by far THE MOST expensive query that gets fired VERY often. I see it being fired almost every 10 secs (and many times even quicker)
2> Is there a way to stop the count query above fired or reduce its frequency.
Please advice.
We have a medium-sized setup (I don't know if thats the right characterization)
Here's our Dashboard for you to get some context
Parameter Value Details
Zabbix server is running Yes
localhost:10051
Number of hosts (enabled/disabled/templates)
368 304 / 1 / 63
Number of items (enabled/disabled/not supported)
22022 17752 / 214 / 4056
Number of triggers (enabled/disabled [problem/ok])
5046 1703 / 3343 [284 / 1419]
Number of users (online) 12 3
Required server performance, new values per second
318.21 -
As you can see we just over 300 hosts, 20k items and a dozen users.
The NVP is just over 315 !
What we have been seeing is that when we navigate to "Latest data" page under Monitoring tab...
I see that zabbix is generating a lot of HUGE queries and sending it to our PostgreSQL database.
Our setup for almost (95%) of the items is to collect the metric every minute, store the History data for 7 days and trends data for 365 days.
We have a Screen refresh for each user set at 300 sec. and at any time there are no more than 3 to 5 users looking.
For the following workloads,
0> Is this EXPECTED that we see many such BIG queries ?
1> I also see that there's a select count(*) from history query which is by far THE MOST expensive query that gets fired VERY often. I see it being fired almost every 10 secs (and many times even quicker)
2> Is there a way to stop the count query above fired or reduce its frequency.
Please advice.
Comment