Closing all indices before system recovery

[Waxman (Wax)]

Hi All,

I'd like to ask if there is a possibility to recover whole cluster without closing indices ?
In Kibana 7.11.2-> stack management -> snapshot and restore ->when trying restore "All data streams and indices, including system indices"
-> i'm getting:

`
Unable to restore snapshot

[snapshot_restore_exception] [es_backup:daily-snap-2021.10.20-xaq5a3leraymvisnlfttoa/n35AFb6bQOqNykET32yZAg] cannot restore index [.ds-ilm-history-5-2021.10.12-000001] because an open index with same name already exists in the cluster. Either close or delete the existing index or restore the index under a different name by providing a rename pattern and replacement name.
`

Is there a chance to close all indices without losing the system ? Because when I close all indices (not one by one because it's taking too long) I need an emergency user to operate cluster. Is there any other way to restore whole cluster ?

[Waxman (Wax)]

Look like the restore cannot be done when some data streams are open. Is there a way to use the snapshot from GUI or curl with open data stream ? Another thing is we cannot delete/close the data stream. Looks like some circle problem

[Waxman (Wax)]

I did some googling and find two options:
xpack.ilm.enabled: false
indices.lifecycle.history_index_enabled: false
Setting those parameters probably helped.

[sebastien]

Hi @Waxman

The .ds-ilm-history-5-2021.10.12-000001 is indeed the lifecycle history index. Were you able to create a snapshot after setting indices.lifecycle.history_index_enabled: false ?

Thanks
Sébastien