As you know if the warning message from a check is acknowledged, it will stop alerting for that check.
if the text in the message changes, check still stays acknowledged.. Is there any way to change this so that every time the message changes from the check, it will clear the acknowledgement.
Current flow. warning is “File system is at 85%” the warning is then acknowledged the warning then changes to “File system is at 86%”… check stays acknowledged… /sad.gif” style=”vertical-align:middle” emoid=”:(” border=”0″ alt=”sad.gif” />
Desired flow. warning is “File system is at 85%” the warning is then acknowledged the warning then changes to “File system is at 86%”… check acknowledgement is cleared /smile.gif” style=”vertical-align:middle” emoid=”:)” border=”0″ alt=”smile.gif” />
Hi Daniel, We are reviewing how acknowledgements are handled. Having the ACK message clear when the metric changes makes sense for a monitor like the FSCAP monitor where the metric doesn't change very frequently but wouldn't make sense for something like a CPU check where the exact % can (and will) change on each check of the monitor. Perhaps it would be better if the ACK status cleared when a new threshold was crossed, for example if a monitor changed from WARN -> CRIT or a different file system crossed a threshold. How about if the ACK message reset itself after a set period of time? So if a monitor was in ACK state for 24 hours clear the ACK so that alerting would restart if the outage hadn't been resolved?
Thanks for your feedback,
Couldn't the auto clear just be setup as simple checkbox when typing in the acknowledgement? It sounds better to allow it to be a quick option instead of a standard because not everyone will want auto clear there for all their services.
We are also struggling with this ack feature.
AnÂ ack will disable the whole service instead of acknowledging one individual outage.
eg.. : We are sending SNMP messages from a certain source system to uptime.
This is a critical message. So far so good, but the system never sends an SNMP ok.
Till the end of time you will get uptime critical messages until you acknowledge.
Once you acknowledge you will never get any new messages anymore from that source system.
So it would be good to add the feature to acknowledge an outage and not the whole service.
As far as I can see, you can’t even undo the acknowledge manually.