[DOCS-1862] Page in manual needed on how to deal with a possibly corrupt database Created: 29/Aug/13  Updated: 16/Mar/15  Resolved: 04/Mar/14

Status: Closed
Project: Documentation
Component/s: manual
Affects Version/s: None
Fix Version/s: v1.3.2, mongodb-2.6

Type: Task Priority: Critical - P2
Reporter: Ian Daniel Assignee: Bob Grabar
Resolution: Done Votes: 0
Labels: sprint-rollover
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
is depended on by SERVER-10017 Remove "run repair" recommendation fr... Closed
Duplicate
is duplicated by DOCS-1239 Pre-existing mongod.lock file is OK, ... Closed
Participants:
Days since reply: 9 years, 50 weeks, 5 days ago

 Description   

SERVER-10017 is a request to remove error messages that say to run repair, or to consider running repair. This is because running repair is usually a last resort; there are usually better ways to recover a node than running repair.

Eliot has approved SERVER-10017 with the approach of replacing the "run repair" recommendations in error messages with "action may be required", and with a link to a page in the manual that provides guidelines on what to do in potential corruption scenarios.

This DOCS ticket is for that page.

The page should describe that, if in a replica set, it is preferable to sync the node from scratch, or perhaps to seed it from the data files of another node (if there is no evidence of corruption in that node). Or if not in a replica set, using a recent backup. It should also discuss running validate to help determine if there is evidence of corruption, and the performance hit of doing so. Also repair as a last resort.

When complete, we need to edit the dochub link for the key "repair" (see DOCS-2776).



 Comments   
Comment by Sam Kleinman (Inactive) [ 28/Feb/14 ]

https://github.com/mongodb/docs/pull/1520

Comment by Githook User [ 17/Dec/13 ]

Author:

{u'username': u'Zackrobat', u'name': u'Zack Brown', u'email': u'zack.brown@10gen.com'}

Message: Minor rewrites

These need to be extracted from DOCS-1862 to avoid merge hell.

Signed-off-by: Sam Kleinman <samk@10gen.com>
Branch: v2.4
https://github.com/mongodb/docs/commit/3b73f408e64f93c099baaaa0d02ed063868e6b22

Comment by Githook User [ 17/Dec/13 ]

Author:

{u'username': u'Zackrobat', u'name': u'Zack Brown', u'email': u'zack.brown@10gen.com'}

Message: Minor rewrites

These need to be extracted from DOCS-1862 to avoid merge hell.

Signed-off-by: Sam Kleinman <samk@10gen.com>
Branch: master
https://github.com/mongodb/docs/commit/b7bd21f1712f8f4db8967ac9c15108dfe9a5da50

Comment by Githook User [ 05/Dec/13 ]

Author:

{u'username': u'tychoish', u'name': u'Sam Kleinman', u'email': u'samk@10gen.com'}

Message: DOCS-1862: edits to shutdown methods
Branch: master
https://github.com/mongodb/docs/commit/1ea4f550c91397c78925c65e4eac470817ee7cee

Comment by Githook User [ 05/Dec/13 ]

Author:

{u'username': u'tychoish', u'name': u'Sam Kleinman', u'email': u'samk@10gen.com'}

Message: DOCS-1862: edits to shutdown methods
Branch: v2.4
https://github.com/mongodb/docs/commit/c0b06c9ca9479cf5af518f7f1b700de280d574a1

Comment by Githook User [ 05/Dec/13 ]

Author:

{u'username': u'Zackrobat', u'name': u'Zack Brown', u'email': u'zack.brown@10gen.com'}

Message: Clean up shutdown procedures.

Take this out of DOCS-1862, which has gotten a lot of cruft that will otherwise cause merge issues.

Signed-off-by: Sam Kleinman <samk@10gen.com>
Branch: master
https://github.com/mongodb/docs/commit/0fc2c66445b1afbe3ffa31e9d963893aeaef0721

Comment by Githook User [ 05/Dec/13 ]

Author:

{u'username': u'Zackrobat', u'name': u'Zack Brown', u'email': u'zack.brown@10gen.com'}

Message: Clean up shutdown procedures.

Take this out of DOCS-1862, which has gotten a lot of cruft that will otherwise cause merge issues.

Signed-off-by: Sam Kleinman <samk@10gen.com>
Branch: v2.4
https://github.com/mongodb/docs/commit/90e393642986b7c6ac46076d1c66c7845eafe263

Comment by Ian Daniel [ 04/Sep/13 ]

Hi Sam,

Sorry for my delay in responding.

I agree that we should use that page rather than write another, however that page will need some changes in order for it to work.

The current page focusses on recovering after an unexpected shutdown, especially if journalling is not set. It would need to be generalised (including the title) to recovering from any cause of possible data corruption. If we do not change it, users referred to this page when seeing the warning message in a log file might think, "But there was no unexpected shutdown of the node, and I am running journalling, so this does not refer to me."

I also think that the page needs re-ordering so that we state at the start of the page what the recovery process is if you are running in a replica set. My reasoning is that it is a simpler procedure than running repair or validate, so users should not have to wade through the complicated scary procedure to find out that the replica set procedure is much simpler.

Kind regards,
Ian

Comment by Sam Kleinman (Inactive) [ 29/Aug/13 ]

Is the document you describe not the following document: http://docs.mongodb.org/manual/tutorial/recover-data-following-unexpected-shutdown/

Include the following dochub url http://dochub.mongodb.org/core/data-recovery in the server code...

Cheers,
sam

Generated at Thu Feb 08 07:42:05 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.