-
Notifications
You must be signed in to change notification settings - Fork 24.8k
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
Dangling indices documentation #58751
Dangling indices documentation #58751
Conversation
Further docs fixes. Update gateways docs to cover the new `gateway.gateway.auto_import_dangling_indices` setting, and rework text since the default behaviour is now not to import automatically.
Pinging @elastic/es-docs (>docs) |
Pinging @elastic/es-distributed (:Distributed/Distributed) |
The test failures look related, holding off on a proper review for now. |
@elasticmachine run elasticsearch-ci/docs |
@DaveCTurner CI is now green. |
We discussed this in today's distrib team sync and agreed to default auto-importing to A common (unsafe) source of dangling indices is using |
Yes, on reflection I should have raised a separate PR - curse my laziness 😅 |
I split the server changes into #58898. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good, I left a few nits.
Co-authored-by: David Turner <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM; I believe @lockewritesdocs is also taking a look too.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Overall, LGTM. I left some editorial comments that you can choose to incorporate.
One suggestion: consider adding an Examples section to the Import dangling index API page. This addition provides users with a more comprehensive example and shows the response structure.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've left two small comments, places where I think we should add a few more clarifications. Thank you
When a node joins the cluster, if it finds any shards stored in its local data | ||
directory that do not already exist in the cluster, it will consider those | ||
shards to be "dangling". You can automatically import dangling indices | ||
into the cluster using the following setting, however this is not safe, and |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm a bit confused by the "this is not safe". Without further qualification, this would leave a reader puzzled as well.
to `false`. | ||
|
||
WARNING: The auto-import functionality was intended as a best effort to help users | ||
who lose all master nodes. For example, if a new master node were to be |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm wondering if we should add that that scenario (losing all master nodes) should be avoided at all cost, as it puts the cluster's metadata and data at risk. Also worth pointing out above (when talking about "this is not safe") that importing dangling indices does not provide any guarantees as to whether the data that's imported truly represents the latest state of the data as it was when the index was still part of the cluster.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@ywelsch I've reworked the text to and address your points.
Co-authored-by: Adam Locke <[email protected]>
its use is strongly discouraged in favour of the | ||
<<dangling-indices-api,dedicated API>. | ||
|
||
WARNING: Losing all master nodes is a situation that should be avoided at |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@lockewritesdocs Is it overkill to have two WARNING:
sections? I felt that they were making separate points, and all the content here ought to be scary.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. Can you re-request a review next time please (click button next to my name in Github UI)? The PR will then appear in my list of PRs to review. Makes it easier to stay on top of things.
Part of #48366. Add documentation for the dangling indices API added in #58176. Co-authored-by: David Turner <[email protected]> Co-authored-by: Adam Locke <[email protected]>
Backported to |
Part of #48366. This PR adds documentation for the dangling indices API added in #58176.