Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add important warning note to backup .kibana index before upgrading #25658

Closed
ppf2 opened this issue Nov 14, 2018 · 7 comments · Fixed by #50525
Closed

Add important warning note to backup .kibana index before upgrading #25658

ppf2 opened this issue Nov 14, 2018 · 7 comments · Fixed by #50525
Assignees
Labels
Team:Docs Team:Operations Team label for Operations Team

Comments

@ppf2
Copy link
Member

ppf2 commented Nov 14, 2018

Kibana version: 6.5

Given that we are doing some major migration/rewrite of the .kibana index automatically (https://www.elastic.co/guide/en/kibana/current/upgrade-migrations.html#upgrade-migrations-old-indices) when upgrading to 6.5 and we are not automatically backing up the original .kibana index for troubleshooting or unexpected failures, it is important for the end users to backup the .kibana index before performing the upgrade.

"When changes are necessary, a new incremental .kibana_N index is created with updated mappings, then the saved objects are loaded in batches from the existing index, transformed to whatever extent necessary, and added to this new index."

"Saved objects are stored in an index named .kibana_N, where N is a number that increments over time as Kibana is upgraded."

While we keep kibana_N around, the above statements suggest that .kibana_1 is already a modified version of the original so isn't a good rollback index to the original.

Currently in our upgrade guide, it simply says "Backup your data using the Elasticsearch snapshots feature. "
https://www.elastic.co/guide/en/kibana/current/upgrade.html#upgrade

image

We should explicitly call out taking a snapshot of the .kibana index before upgrading (not just the end user's data indices).

@ppf2 ppf2 added the Team:Docs label Nov 14, 2018
@ppf2
Copy link
Member Author

ppf2 commented Nov 14, 2018

@AlonaNadler

@jpcarey
Copy link

jpcarey commented Nov 14, 2018

I don't feel that just documenting a snapshot backup (not all users have this setup) is enough, it should be automatic.
#20674

@ppf2
Copy link
Member Author

ppf2 commented Nov 14, 2018

If .kibana_1 is actually the "original" backup (without any modifications), let's clarify that in the documentation :)

@AlonaNadler
Copy link

Adding @elastic/kibana-operations which is now working on the upgrade assistant

@epixa
Copy link
Contributor

epixa commented Nov 15, 2018

.kibana_1 is a straight reindex (no migrations) from .kibana, so it's the backup. We created it that way intentionally. So folks shouldn't need to manually back anything up, though doing so before upgrades is and has always been a good idea.

We can clarify docs about this. There is a note in https://www.elastic.co/guide/en/kibana/current/upgrade-migrations.html about the initial migration, but it's probably worth clarifying a bit.

@tylersmalley tylersmalley added the Team:Operations Team label for Operations Team label Nov 15, 2018
@gchaps
Copy link
Contributor

gchaps commented Oct 9, 2019

@jbudz Can you please provide some guidance on how to handle this issue in the docs?

@jbudz
Copy link
Member

jbudz commented Oct 10, 2019

Probably something at the end of the first paragraph to the tune of what court `mentioned.

It's always recommended to have backups on hand prior to upgrading (something about Kibana upgrades automatically when starting a new version). https://www.elastic.co/guide/en/kibana/current/snapshot-repositories.html can be used to backup Kibana's data by targeting .kibana* indices.

I don't think we need to add other scenarios, they may help, but the end result should be to always have backups on hand.

Unrelated but related I think the general approach in the document is at odds with user friendly automatic upgrades. I think the troubleshooting section should match exact errors with resolution steps, and the internals description should be removed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Team:Docs Team:Operations Team label for Operations Team
Projects
None yet
Development

Successfully merging a pull request may close this issue.

7 participants