[SERVER-3971] Assertion failures / segfaults querying an apparently corrupt index Created: 28/Sep/11  Updated: 08/Mar/13  Resolved: 17/Jan/13

Status: Closed
Project: Core Server
Component/s: Index Maintenance
Affects Version/s: 2.0.0
Fix Version/s: None

Type: Bug Priority: Minor - P4
Reporter: Leo Dirac Assignee: Unassigned
Resolution: Incomplete Votes: 0
Labels: indexing
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Mac OSX 10.6.8, 64-bit


Attachments: Text File mongolog.txt    
Operating System: OS X
Participants:

 Description   

I did a mongodump/mongorestore of a collection from another server onto my laptop, and then found on my laptop that basic queries were blowing up. I did a reindex and the problems went away.



 Comments   
Comment by Eliot Horowitz (Inactive) [ 17/Jan/13 ]

I think this was caused by running out of memory.
That is handled better now, and cannot reproduce this case anymore.

Comment by Leo Dirac [ 03/Oct/11 ]

Why yes, now that you mention it, I do see that the mongod process got killed for being out of memory a few times. But I always ran --repair when it asked, FWIW.

Comment by Eliot Horowitz (Inactive) [ 03/Oct/11 ]

Any chance the 32-bit sever ran out of memory/disk space?

Comment by Scott Hernandez (Inactive) [ 29/Sep/11 ]

The file format is the same on all systems, and the dump format is BSON. There is no need to dump, all you need to do is copy the underlying db files between machines.

Comment by Leo Dirac [ 29/Sep/11 ]

2.0.0 on both sides. But the server which I dumped from is 32-bit, and the laptop is 64-bit if that matters.

Comment by Scott Hernandez (Inactive) [ 28/Sep/11 ]

Was the data from a previous version of mongo?

Comment by Leo Dirac [ 28/Sep/11 ]

The log was from the interactive shell. The errors were totally repeatable before I did reIndex.

Generated at Thu Feb 08 03:04:35 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.