[SERVER-25365] MongoDB 3.2.1 crash during resync Created: 30/Jul/16 Updated: 30/Jul/16 Resolved: 30/Jul/16 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Admin |
| Affects Version/s: | 3.2.1 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Tinco Andringa | Assignee: | Kelsey Schubert |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Operating System: | ALL | ||||||||
| Participants: | |||||||||
| Description |
|
Hi, One of our MongoDB shards last replica just crashed while resyncing to its replica which had crashed due to a known 3.2.1 bug. This caused us some downtime, crashing during a resync is rather painful.. It's rebooting now, and will be rebooting in to 3.2.3, but booting takes a long time (it's a fairly large wiredtiger db) so we'll be down for a while. Please let me know if you need more info or if it's a dupe or something.
|
| Comments |
| Comment by Tinco Andringa [ 30/Jul/16 ] | |
|
Thanks! We'll be rolling our cluster over to 3.2.3 asap. | |
| Comment by Kelsey Schubert [ 30/Jul/16 ] | |
|
Hi tinco, Thank you for uploading the complete logs for the log. After review, this issue appears to be a duplicate of Kind regards, | |
| Comment by Ramon Fernandez Marina [ 30/Jul/16 ] | |
|
Yep, the upload worked. Here's the interesting line:
I don't recall seeing this before, but we'll investigate. Please continue to watch the ticket for updates. | |
| Comment by Tinco Andringa [ 30/Jul/16 ] | |
|
Just a note about our database. We have a database per customer, and currently have a collection in that database per day. This is because in the old database engine we used to get much better delete performance this way. We have not tested if delete performance still is better that way now, I've seen WiredTiger has a significantly different file layout, so maybe our architecture is a bit crazy now. | |
| Comment by Tinco Andringa [ 30/Jul/16 ] | |
|
Thanks Ramon, I've uploaded the logs, though it didn't go to a 'success' page, did it work? | |
| Comment by Ramon Fernandez Marina [ 30/Jul/16 ] | |
|
tinco, can you please upload full logs for this node from the last startup until the fassert()? I think we'll need more than the snippet above to track this down. I've created a safe, secure upload portal so your logs can only be accessed by us for the purpose of debugging this isse. Thanks, |