I have just run into this issue, and googling around I havent found anyone who clearly provides a fix to this. Therefore I'm posting a work around I have found, so others wont suffer the panic i have had over the last few hours.
Issue: Zabbix alerter process at 100% and all items in the alert log are stuck in pending. This happens after one of the users has enabled jabber media.
Cause:
I found the problem was that the default setup was to send jabber messages to [email protected]. The alerter process opens up a connection to to this and stalls. I'm not sure if this a problem with the way the socket is opened in the server c code, but basically the connection never times out so the alerter process is stuck
root@monitoring:/opt/zabbix/alerts# ps auxwww| grep zabb | grep alert
zabbix 28976 0.0 0.2 271044 2112 ? S 13:49 0:00 /usr/sbin/zabbix_server: db watchdog [synced alerts config in 0.000386 sec, idle 60 sec]
zabbix 29009 0.0 0.2 273708 2852 ? S 13:49 0:00 /usr/sbin/zabbix_server: alerter [sending alerts]
root@monitoring:/opt/zabbix/alerts# strace -p 29009
Process 29009 attached
connect(6, {sa_family=AF_INET, sin_port=htons(5222), sin_addr=inet_addr("74.122.237.76")}, 16
Work around:
Anyway the simple fix is to put the following entry in the hosts file and then restart zabbix_server. The alert queue will then rapidly process .
127.0.0.1 company.com
Not the best way to do things but it may help people out of the hole until its fixed correctly.
Issue: Zabbix alerter process at 100% and all items in the alert log are stuck in pending. This happens after one of the users has enabled jabber media.
Cause:
I found the problem was that the default setup was to send jabber messages to [email protected]. The alerter process opens up a connection to to this and stalls. I'm not sure if this a problem with the way the socket is opened in the server c code, but basically the connection never times out so the alerter process is stuck
root@monitoring:/opt/zabbix/alerts# ps auxwww| grep zabb | grep alert
zabbix 28976 0.0 0.2 271044 2112 ? S 13:49 0:00 /usr/sbin/zabbix_server: db watchdog [synced alerts config in 0.000386 sec, idle 60 sec]
zabbix 29009 0.0 0.2 273708 2852 ? S 13:49 0:00 /usr/sbin/zabbix_server: alerter [sending alerts]
root@monitoring:/opt/zabbix/alerts# strace -p 29009
Process 29009 attached
connect(6, {sa_family=AF_INET, sin_port=htons(5222), sin_addr=inet_addr("74.122.237.76")}, 16
Work around:
Anyway the simple fix is to put the following entry in the hosts file and then restart zabbix_server. The alert queue will then rapidly process .
127.0.0.1 company.com
Not the best way to do things but it may help people out of the hole until its fixed correctly.