-
All of the other checks similar to swift-recon.py use their client libraries so this should as well. That said, the swift team is moving away from actively developing python-swiftclient so we should a…
-
The rabbitmq_status.py assumes cluster size of 3 hosts which is a hard coded number. In cases where we rabbitmq cluster > 3 nodes we will run into monitoring issues. Please either determine that from …
-
Please design a monitoring for all relevant Openstack log files, so we are getting notified if logfiles are stale due to a process crashed etc. This functionality should be integrated into rpc-maas
-
In my case the check did not work properly if galera is in the following state:
wsrep_local_state 2
wsrep_local_state_comment Donor/Desynced
wsrep_cluster_size 3
wsrep_cluster_status Primary…
-
Dell script openmanage.py does not seem to be catching predictive drive failures. Unfortunately predictive failures do not "degrade" virtual disks or show up under omreport chassis. The failure predic…
kstev updated
8 years ago
-
disk_utilisation.py is not excluding nb block devices and bails out on 30 metric limit:
error Maximum of 30 metrics per checkerror Maximum of 30 metrics per check
I will create a pull request to remo…
-
Currently we do not check important memcache metric like:
- bytes
- evictions
- curr_connections
Since memcache can affect the performance of openstack negatively we should have those checks in place…
-
We had disks filling up on compute nodes and that caused maas to store corrupt service catalog cache files (took some time to find that). In such cases, I would like to add a parsing of file this file…
-
Started seeing this in 10.1.18...
```
PLAY [rabbit] *****************************************************************
GATHERING FACTS ***************************************************************…
-
rabbitmq_status.py (10.1.9) does not provide the 'rabbitmq_max_channels_per_conn--' metric, which causes the MaaS checks (installed from os-ansible-deployment/rpc_deployment/playbooks/monitoring/) to …