Hello,
We have a Zabbix install with server version 2.0.6 and agent version 2.0.9. We are thinking of upgrading the server to version 2.2.3 as it looks real promising but we would like to solve this issue first.
We use zabbix to check if any security packages are available for install. To do this, we run the following command :
It is defined in the agent configuration with the following line :
The problem we are having is that we will regularly (every 15 minutes or so) get an alert that says that there are 7 packages to be updated. Two minutes after this first alert, we will get a recovery message saying that there are no packages to be updated. This happens all the time.
To make sure that there was not a problem with this utility, I ran it in a loop for hours and it did not ever output anything else then "0" despite the alerts continuing being sent.
I even enabled Zabbix agent output to see what was going on and it seems that the correct data in parsed by the agent as show below :
I have also tried to tcpdump the zabbix agent traffic but I can't seem to be able to find the traffic associated to this specific check.
Despite all this, we keep getting this monitoring error where it is showing incorrect data for this check. It happens on a 3-4 hosts out of 50 Linux servers.
We are therefore looking for any pointers as to what we can do to see where this could be coming from ? Is this a known bug in a version of Zabbix ? Has anyone ever had similar issues ?
Thank you in advance for your help,
Antoine
We have a Zabbix install with server version 2.0.6 and agent version 2.0.9. We are thinking of upgrading the server to version 2.2.3 as it looks real promising but we would like to solve this issue first.
We use zabbix to check if any security packages are available for install. To do this, we run the following command :
Code:
/usr/lib/update-notifier/apt-check 2>&1 | cut -d';' -f2
Code:
UserParameter=zext.packages.security,/usr/lib/update-notifier/apt-check 2>&1 | cut -d';' -f2
To make sure that there was not a problem with this utility, I ran it in a loop for hours and it did not ever output anything else then "0" despite the alerts continuing being sent.
I even enabled Zabbix agent output to see what was going on and it seems that the correct data in parsed by the agent as show below :
Code:
1162:20140313:145728.272 In process_active_checks('127.0.4.5',10051)
1162:20140313:145728.272 In zbx_popen() command:'/usr/lib/update-notifier/apt-check 2>&1 | cut -d';' -f2'
1162:20140313:145728.272 End of zbx_popen():6
2719:20140313:145728.273 zbx_popen(): executing script
1158:20140313:145728.696 In update_cpustats()
1158:20140313:145728.696 End of update_cpustats()
1162:20140313:145729.305 In zbx_waitpid()
1162:20140313:145729.305 zbx_waitpid() exited, status:0
1162:20140313:145729.305 End of zbx_waitpid():2719
1162:20140313:145729.305 Run remote command [/usr/lib/update-notifier/apt-check 2>&1 | cut -d';' -f2] Result [1] [0]...
1162:20140313:145729.305 For key [zext.packages.security] received value [0]
1162:20140313:145729.305 In process_value() key:'app.my.domain.tld:zext.packages.security' value:'0'
Despite all this, we keep getting this monitoring error where it is showing incorrect data for this check. It happens on a 3-4 hosts out of 50 Linux servers.
We are therefore looking for any pointers as to what we can do to see where this could be coming from ? Is this a known bug in a version of Zabbix ? Has anyone ever had similar issues ?
Thank you in advance for your help,
Antoine