Werk #14839: oracle_dataguard_stats: Bufix for wrong Recovery Status on RAC Standby
| Component | Checks & agents | ||||||
| Title | oracle_dataguard_stats: Bufix for wrong Recovery Status on RAC Standby | ||||||
| Date | Nov 4, 2022 | ||||||
| Level | Trivial Change | ||||||
| Class | Bug Fix | ||||||
| Compatibility | Incompatible - Manual interaction might be required | ||||||
| Checkmk versions & editions |
|
The check results in false CRITs on RAC standby databases with multiple running instances. Some more changes are done, due to wrong design in old code.
The option to monitor the MRP process has been removed, because it is not needed. The apply_lag is more important than the recovery process itself. If recovery has been stopped, the monitoring of apply_lag will lead to WARN/CRIT, which is the most important metric in a Data-Guard environment.
BREAKING Change: The option mrp state has been removed, because it is not needed. A missing recovery process (MRP0) is not an issue, until the apply_lag has a proper WARN/CRIT definition - that is more important than a stopped recovery in the standby database.
You have to re-install the agent plugin "mk_oracle".