You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The healthcheck does the following for all presets on all proxies: create-namespace, upload-object, delete-object, delete-namespace.
I think it would be a good idea to also do this for backends which have no proxies.
This can e.g. be the case for all flash backends which are only used for (fragment) caching.
The text was updated successfully, but these errors were encountered:
Please be aware of the fact that you also need to check the puts to that backend to another node. In that case you also check if the disks on that other node is reachable over the network.
Had this issue today at GIG. The cache backend wasn't reachable from the controller/volumedrivernodes due to non reachable nodes over the storage IP.
Another point of view is: the proposed test could be done from any node (wherever is mot convenient to run it).
And another check is added to the healthcheck to verify network connectivity between all nodes.
The healthcheck does the following for all presets on all proxies: create-namespace, upload-object, delete-object, delete-namespace.
I think it would be a good idea to also do this for backends which have no proxies.
This can e.g. be the case for all flash backends which are only used for (fragment) caching.
The text was updated successfully, but these errors were encountered: