Status of some hosts and triggers stuck
In the continued series of adventures of the server migration (from a server on our LAN running Ubuntu 11.04 with Zabbix 1.8.3 to a dedicated server with Debian Stable 6.03 and Zabbix 1.8.6 manually backported from Debian Testing, all with a PostgreSQL 8.4 database which was entirely migrated from one server to the other), i've hit a couple more snags:
How do i get those unstuck? Is there some query i can send to Postgres to manually reset those states and leave them to work as they're supposed to?
Also, is there a way to programatically check if there's more inconsistencies like this to fix them? or is this something one of those things you must check manually?
- There's a server where the majority of triggers stay in "Unknown" state, shown in the triggers page with the red X and "Zabbix was restarted" on the mouseover. It's been like this since yesterday.
- There's a few hosts, happily collecting information and working fine, but the hosts list shows them as unavailable (the red Z to the right). A mouseover reveals "Got empty string from [IPADDRESS]. Assuming that agent dropped connection because of access permissions", despite the fact that it's collecting information and is not shown as unavailable in the dashboard.
How do i get those unstuck? Is there some query i can send to Postgres to manually reset those states and leave them to work as they're supposed to?
Also, is there a way to programatically check if there's more inconsistencies like this to fix them? or is this something one of those things you must check manually?
Comment