We currently have an active/active situation. But there are still a lot of
things that could improve.
This is what it looks like:
zenhub1 zenhub2
| |
zenprobe1 |
|
zenprobe2
All our devices are snmp-polled by the two zenprobes. In case one fails at
least the other one is still collecting performance data (we need to have
accurate performance data for our customer).
The 2 zenhubs are for realtime monitoring; meaning all the devices send their
snmptraps to both of the zenhubs.
Several problems I encountered are:
* Syncing event mappings between 2 zenhubs
* Syncing user accounts between 2 zenhubs
* The zenprobes are currently seen as one zenprobe by the hub. If one fails I
have to manually switch the renderUrl
* When a zenprobe fails I get gaps in the graphs. So when pulling out data
for our customers I have to take into account where these gaps occured and use
the other zenprobe for that period of time.
+ shared storage could be a solution but is also a redundancy risk
There are more things I've come across, I might put them here later on.
Good luck
-------------------- m2f --------------------
Read this topic online here:
http://community.zenoss.com/forums/viewtopic.php?p=19491#19491
-------------------- m2f --------------------
_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users
