[SERVER-16118] MongoD 2.8rc0 terminates with "WiredTigerRecordStore::insertRecord 12: Cannot allocate memory" Created: 13/Nov/14 Updated: 24/Nov/14 Resolved: 13/Nov/14 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Storage, Tools |
| Affects Version/s: | 2.8.0-rc0 |
| Fix Version/s: | 2.8.0-rc1 |
| Type: | Bug | Priority: | Critical - P2 |
| Reporter: | rohit.nijhawan@10gen.com | Assignee: | Matt Kangas |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Attachments: |
|
||||||||||||||||||||
| Issue Links: |
|
||||||||||||||||||||
| Operating System: | ALL | ||||||||||||||||||||
| Participants: | |||||||||||||||||||||
| Description |
|
I tried to mongorestore an existing dump. The dump was taken with the older (non-Go) version of mongodump with an older mongod. However this time mongod and mongorestore were both version 2.8.0-rc0 And then mongod terminated with this in the log
It's possible that something is wrong in my environment but please try to reproduce by running mongod with wiredtiger and then: mongorestore -d=en -c=messages ~/enron/messages.bson I was able to do it 3 times. |
| Comments |
| Comment by Matt Kangas [ 13/Nov/14 ] | ||||||||||||||||||||||||||||||||||
|
WT commit b4df6ce merged in https://github.com/mongodb/mongo/commit/f31a38be3ce39267382e59b4fe74be1129b96679 | ||||||||||||||||||||||||||||||||||
| Comment by Michael Cahill [ 13/Nov/14 ] | ||||||||||||||||||||||||||||||||||
|
The cause was that this workload generated a log record that was not serialized correctly by WiredTiger. | ||||||||||||||||||||||||||||||||||
| Comment by Michael Cahill [ 13/Nov/14 ] | ||||||||||||||||||||||||||||||||||
|
This issue should be fixed by https://github.com/wiredtiger/wiredtiger/commit/b4df6ce9d74be9841d631657cd09cdedb562c007 on WiredTiger's develop branch. | ||||||||||||||||||||||||||||||||||
| Comment by Alex Gorrod [ 13/Nov/14 ] | ||||||||||||||||||||||||||||||||||
|
I can reproduce this as well. I've seen several different stack traces and with a debug build I see: memory allocation bug: object at 0x7fffe6104020 has been already deallocated; or else a word before the object has been corrupted (memory stomping bug) | ||||||||||||||||||||||||||||||||||
| Comment by Daniel Pasette (Inactive) [ 13/Nov/14 ] | ||||||||||||||||||||||||||||||||||
|
Seeing a slightly different trace after repro.
|