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
There are "rare" occasions when the cluster state may be manually deleted. For example, if a new trial license is wanted, or if the cluster state is red and cannot be recovered. This is not a recommended way to recover, but might be the last ditch option to try in some circumstances.
If the cluster state is removed rm -rf _state then upon an elasticsearch restart, the indices are recreated based upon files seen on disk.
If this occurs, ML job and datafeed configurations are not recreated (amongst other things). However the results and state indices are. It would not be possible to view these ML jobs without some black magic in play, and knowledge of the exact ML configuration details.
There are no plans yet to address or mitigate this, nor any easy resolution. Noting for informational purposes.
The text was updated successfully, but these errors were encountered:
Original comment by @sophiec20:
There are "rare" occasions when the cluster state may be manually deleted. For example, if a new trial license is wanted, or if the cluster state is red and cannot be recovered. This is not a recommended way to recover, but might be the last ditch option to try in some circumstances.
If the cluster state is removed
rm -rf _state
then upon an elasticsearch restart, the indices are recreated based upon files seen on disk.If this occurs, ML job and datafeed configurations are not recreated (amongst other things). However the results and state indices are. It would not be possible to view these ML jobs without some black magic in play, and knowledge of the exact ML configuration details.
There are no plans yet to address or mitigate this, nor any easy resolution. Noting for informational purposes.
The text was updated successfully, but these errors were encountered: