[SERVER-15351] Index data corruption on primary Created: 22/Sep/14 Updated: 20/Mar/15 Resolved: 20/Mar/15 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Index Maintenance |
| Affects Version/s: | 2.4.8 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Critical - P2 |
| Reporter: | Dharshan Rangegowda | Assignee: | Bruce Lucas (Inactive) |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Attachments: |
|
| Backwards Compatibility: | Fully Compatible |
| Operating System: | ALL |
| Participants: |
| Description |
|
Hi, My primary has hit an error updating an index. No action was performed on our end other than inserting data. Please let me know what other data you need.
|
| Comments |
| Comment by Bruce Lucas (Inactive) [ 20/Mar/15 ] | ||||||||||||
|
Hi Dharshan, We weren't able to reproduce this issue, and there wasn't enough information in the db files themselves to reconstruct what happened - for example whether it was a MongoDB issue or a platform issue (hardware, operating system). Since this occurred on an older version of MongoDB, I'll go ahead and close this ticket now; if you encounter this again, particularly on a newer version of MongoDB (we are currently shipping 3.0.1), please re-open the ticket. Thanks, | ||||||||||||
| Comment by Dharshan Rangegowda [ 24/Sep/14 ] | ||||||||||||
|
Hi Bruce, The DB has been running for well over 8 months now. As far as I know there were no administrative actions (e.g. Index building) happening on the server. I have attached the last three log files for the server. This is all the log files we have. Hopefully it will help. | ||||||||||||
| Comment by Bruce Lucas (Inactive) [ 24/Sep/14 ] | ||||||||||||
|
Thanks Dharshan. The .ns file is in an inconsistent state in at least two ways:
To try to understand how the .ns file may have gotten into that state I'll need some more information about the history of the instance:
Thanks, | ||||||||||||
| Comment by Dharshan Rangegowda [ 23/Sep/14 ] | ||||||||||||
|
Hi Bruce, The .ns file is attached. Please let me know what you find. | ||||||||||||
| Comment by Bruce Lucas (Inactive) [ 23/Sep/14 ] | ||||||||||||
|
Hi Dharshan, Can you share the edspringload.ns file with us so we can take a closer look at what's going wrong? It's small so you can just compress it and attach it to this ticket. It only contains metadata, so shouldn't reveal any more information than has already been posted on this ticket, but if you would be more comfortable we can provide a private location for you to upload it; just let me know. Thanks, | ||||||||||||
| Comment by Dharshan Rangegowda [ 22/Sep/14 ] | ||||||||||||
|
Hi Ramon, Ran your script. The collection which is having the index problems has 9 indexes. edspringload.edspringload.transferCrosswalk : 9 However we have not added any indexes or have not experienced index addition failures - Is this the same issue? cleanup.cleanup.assessmentRestoreIgdis20140128 : 6 edspringload.edspringload.users : 20 | ||||||||||||
| Comment by Ramon Fernandez Marina [ 22/Sep/14 ] | ||||||||||||
|
Does any of your collection have 10 or more indexes? I'm asking because you may be running into You can use the following code snippet to see how many indexes are in each collection for all your databases:
|