interruption on zabbix server

[alucard95]

Hello all!!

I use Zabbix 1.6 (upgrade from 1.4) on ubuntu server with mysql database and since I install zabbix (before the upgrade) I have a problem.

My problem is that regularly I have interruption on my graphics and nothing in the logs:




I don't know why there are these interruptions, if anybody have idea..

Thanks

[BlouPhil]

Hi,

How many hosts are you monitoring? Items?

I've ran into this problem when my database was getting a beating from way too much data coming in...

zabbix_server, PHP frontend, MySQL were all running on the same Pentium4 server, at about 30 hosts I had graphs similar to yours

Solution was to get a dedicated powerfull MySQL server (dual core Xeon), installing it from source

What also helped was to disable some items that we didn't 'need'

Hope this helps

[alucard95]

Thanks Blouphil,

I agree, I think it can be the reason but I need to realize more test to confirm

[alucard95]

I still have the same problem with graphs

Hello,

I return to you because I still have the same problem however I use Zabbix with 2 server. One for the application and the other for the mysql database.

Sometimes, there is "hole" in my graph although I still have data.

I check the server and he is not overload, so is it possible that there is a specific configuration for the graphs??

[bbrendon]

Are you using passive or active items?

[alucard95]

Thanks to answer,

Can you explain what you call active or passive item?

[bbrendon]

When you configure items, the type is either Zabbix agent or Zabbix agent (passive)

The first is not labeled as active, though that's what I call it because it's the opposite of passive in my mind.

[alucard95]

Ok,

I use various type of agent, like snmp agent or zabbix agent but not passive items.

But, it is very strange, the hole appears at the same moment on all my graphics...

[MrKen]

Try editing php.ini and increasing memory_limit = 32M (or 64M or whatever).
You will probably have to restart httpd.

I think that should fix the problem, but if not, it might be helpful if you provide some info re: number of hosts/itmes, how much RAM, cpu, etc.

I notice that all your graphs are for 1 hour, are there still gaps when you view longer periods of time - 2 hr, 4 hr , 1 day ? No need to answer if the above works!

Hope this helps.

MrKen

[alucard95]

Thanks for your solution(s) MrKen.

I increase memory_limit to 64M. I test it few days and I give you feed back.

[tchjts1]

When you configure items, the type is either Zabbix agent or Zabbix agent (passive)

The first is not labeled as active, though that's what I call it because it's the opposite of passive in my mind.

What version are you using? Mine is labeled as active in 1.6.2

[bbrendon]

What version are you using? Mine is labeled as active in 1.6.2

I meant the first is not labeled as passive, which is what i call the non-active item.

BTW, a change to php.ini will take effect immediately, not in a few days. I don't see how memory is an issue here, though I could be wrong.

[bbrendon]

Ok,

I use various type of agent, like snmp agent or zabbix agent but not passive items.

But, it is very strange, the hole appears at the same moment on all my graphics...

This hole appears with SNMP, zabbix active, and zabbix [passive], items? The hole for all of the various items is exactly the same?

[tchjts1]

I am seeing the same issue here on a fresh 1.6.2 setup:
Zabbix app server - Ubuntu 8.0.4, DL380
Zabbix MySql server - Ubuntu 8.0.4, ML530

Number of hosts 3
Number of items 195
Number of triggers 73
Number of users 1
Required server performance, new values per second 7.0544 -

The memory_limit was at the default setting of 16M, I bumped it up to 64M.
I also tweaked the my.cnf settings as follows:

# * Fine Tuning
#
key_buffer              = 1500M
max_allowed_packet      = 16M
thread_stack            = 128K
thread_cache_size       = 8
#max_connections        = 100
table_cache             = 1800
#thread_concurrency     = 10
#
# * Query Cache Configuration
#
query_cache_limit       = 100M
query_cache_size        = 99M

Restarted everything. let it run for a few hours and am still seeing the gaps in the graphs. This is a 1 day view, but the gaps all the way to the right are reflective with the new changes in place. Bummer.

[alucard95]

This hole appears with SNMP, zabbix active, and zabbix [passive], items? The hole for all of the various items is exactly the same?

Yes, it is exactly the same on all my graphics.

So, with memory_limits=128M in php.ini, I still have the "hole".

So, I give you some precisions about zabbix architecture:

Zabbix application server: Ubuntu server 8.10, IBM xseries 335, 4Go RAM
Zabbix Mysql server: Ubuntu server 8.10, IBM netfinity 6000R, 8Go RAM

Hosts: 172
Items: 2295
triggers: 831
Required server performance, new values per second: 56.376


I check and none of Zabbix server appears overload.

Thanks for your help