Hi
I have a minor Zabbix set up:
ZABBIX server is running Yes
Number of hosts (monitored/not monitored/templates/deleted) 64(37/1/26/0)
Number of items (monitored/disabled/not supported)[trapper] 2076(1812/0/264)[30]
Number of triggers (enabled/disabled)[true/unknown/false] 845(845/0)[11/312/522]
Number of events 163076
Number of alerts 2261
I was looking around at some values today when I saw that they hadn't been updated for a while. Strange. I looked at the config, all items where supported, all where being monitored still no data for several days (the weekend).
I checked the zabbix_serverd. It was running alright, the load was a bit high, around 7 on a dual Xeon 2.8 GHz with 6 GB of ram, with only zabbix and mysql running. I tried sending the values in question manually with zabbix_sender from the server in question and got after about 5s:
zabbix_sender [7654]: Warning: Timeout while executing operation
sent: 0; failed: 1; total: 1
Wierd. Looked in the zabbix_server log (running with DebugLevel=2) and the last things there where:
2537:20080415:015102 JABBER: disconnecting
2547:20080415:113736 Error doing curl_easy_perform [server returned nothing (no headers, no data)]
which doesn't seems severe at all. So I tried killing zabbix_serverd and starting it again. And now everything worked just fine.
And now I ask myself, wtf was that!? Should I expect the server to just stop accepting commands randomly? When looking at the queue after the restart I sometimes have some in there under "ZABBIX agent" but they disapear if I do a refresh of the page so nothing alarming.
Does anyone have any hints or help to offer? Should I make a cronjob that kills and starts the server if it fails to push data? Was this a freak once in a lifetime thing, or is this common?
I have a minor Zabbix set up:
ZABBIX server is running Yes
Number of hosts (monitored/not monitored/templates/deleted) 64(37/1/26/0)
Number of items (monitored/disabled/not supported)[trapper] 2076(1812/0/264)[30]
Number of triggers (enabled/disabled)[true/unknown/false] 845(845/0)[11/312/522]
Number of events 163076
Number of alerts 2261
I was looking around at some values today when I saw that they hadn't been updated for a while. Strange. I looked at the config, all items where supported, all where being monitored still no data for several days (the weekend).
I checked the zabbix_serverd. It was running alright, the load was a bit high, around 7 on a dual Xeon 2.8 GHz with 6 GB of ram, with only zabbix and mysql running. I tried sending the values in question manually with zabbix_sender from the server in question and got after about 5s:
zabbix_sender [7654]: Warning: Timeout while executing operation
sent: 0; failed: 1; total: 1
Wierd. Looked in the zabbix_server log (running with DebugLevel=2) and the last things there where:
2537:20080415:015102 JABBER: disconnecting
2547:20080415:113736 Error doing curl_easy_perform [server returned nothing (no headers, no data)]
which doesn't seems severe at all. So I tried killing zabbix_serverd and starting it again. And now everything worked just fine.
And now I ask myself, wtf was that!? Should I expect the server to just stop accepting commands randomly? When looking at the queue after the restart I sometimes have some in there under "ZABBIX agent" but they disapear if I do a refresh of the page so nothing alarming.
Does anyone have any hints or help to offer? Should I make a cronjob that kills and starts the server if it fails to push data? Was this a freak once in a lifetime thing, or is this common?
Comment