[dev.icinga.com #420] Feature request: icinga online configuration #204
Comments
Updated by elagon on 2010-05-10 17:08:05 +00:00 For what I can see, you always need the core to reload the configuration. You can't just query the configuration every time you need to know a configuration variable. |
Updated by mfriedrich on 2010-05-16 17:11:33 +00:00 mh would be something like a polling loop, checking if there had been config changes, and then creating a mutex on the actual read config in memory. i'll put that in association with core re-design - #378 |
Updated by Meier on 2010-07-03 08:07:34 +00:00 Simply ignoring invalid config parts can bring up dependency problems. Same goes for live configuration changes, you have to rebuild all dependencies for the altered objects. |
Updated by mfriedrich on 2010-07-09 14:51:42 +00:00
|
Updated by mfriedrich on 2010-11-09 16:21:00 +00:00
|
Updated by mfriedrich on 2011-03-18 17:36:56 +00:00
maybe in association with a new core (api), but not in this special case. online config is always critical and can cause problem you will most likely not to be able to debug. |
This issue has been migrated from Redmine: https://dev.icinga.com/issues/420
Created by rbrinkmo on 2010-05-10 13:13:11 +00:00
Assignee: (none)
Status: Closed (closed on 2011-03-18 17:36:56 +00:00)
Target Version: (none)
Last Update: 2011-03-18 17:36:56 +00:00 (in Redmine)
It would be great to add an function for online configuration.
So, it should'nt be necessary to restart/reload nagios for config changes.
The reason: In big environments it could take a lot of time after reload, untill the next check of services. The more often you have to change something the time between the checks can take to long.
If a configuration is not complete or not correct it should ignored this parts/details instead of stopping the reconfiguration. There should be an info about incomplete or failed configurations (e.g.: self_config_check).
Relations:
The text was updated successfully, but these errors were encountered: