[SERVER-46271] Crash MongoDB: Invalid access at address Created: 20/Feb/20 Updated: 27/Oct/23 Resolved: 16/Mar/20 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | 4.2.1 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Mikhael Grefon | Assignee: | Ian Boros |
| Resolution: | Gone away | Votes: | 0 |
| Labels: | qexec-team | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Attachments: |
|
| Operating System: | ALL |
| Sprint: | Query 2020-03-23 |
| Participants: |
| Description |
|
MongoDB several times shutdown. Logs:
|
| Comments |
| Comment by Ian Boros [ 16/Mar/20 ] | ||||
|
I'm sorry for the delay, unfortunately there is overhead in sorting through/triaging/scheduling these types of tickets. If you encounter problems like this again I encourage you to report them. The more data points we have, the easier it is to figure these problems out. IB | ||||
| Comment by Mikhael Grefon [ 15/Mar/20 ] | ||||
I dont run "mongod --repair", only "systemctl start mongod.service".
3 weeks have passed since the topic was opened until your reply. I have a working project and he could not wait so long! A few days after the topic was opened, I did not receive advice on how to fix the error and therefore began to experiment. After updating Mongo to the latest version and setting up replication to another server, the problem did not occur. | ||||
| Comment by Ian Boros [ 13/Mar/20 ] | ||||
|
mikhael@grefon.com A few questions:
Can you elaborate on what you mean by "repair didn't run"? Are you saying mongod started and shutdown cleanly? I'm also curious exactly what pipelines are running. The stack traces do give a clue, but it'd be good to have the exact query. Can you set the log verbosity of the "commands" component to 2? This should just involve changing the mongod conf file to have something like:
The fact that this crash is happening in so many different places makes me a bit suspicious there's an environmental issue. Have you seen this problem in different environments/configurations? Lastly, I will say that the more data points we have, the easy it will be to figure out what is going on. If you've seen more crashes like this recently, can you post the logs/stack traces? | ||||
| Comment by Carl Champain (Inactive) [ 24/Feb/20 ] | ||||
|
This issue appears to be a bug. We're passing this ticket along to the appropriate team. Updates will be posted on this ticket as they happen. Thank you!
| ||||
| Comment by Mikhael Grefon [ 21/Feb/20 ] | ||||
|
2020-02-09T11: 21:30 - read checksum error for 49152B block at offset 19829555 After that crash i run "mongod --repair". MongoDB started, but all data collection-0-6606871978163036226.wt was lost!
2020-02-19T12: 00: 55 - Invalid access at address: 0 / Got signal: 11 (Segmentation fault). "repair" did`t run. MongoDB started, data saved.
2020-02-19T17:53:33 - Invalid access at address: 0 / Got signal: 11 (Segmentation fault). "repair" did`t run. MongoDB started, data saved.
2020-02-20T13:56:10 - Invalid access at address: 0 / Got signal: 11 (Segmentation fault). "repair" did`t run. MongoDB started, data saved.
I attach logs: mongod.log Wondering why this happened and how to avoid it?
| ||||
| Comment by Carl Champain (Inactive) [ 21/Feb/20 ] | ||||
|
Can you please try to run mongod --repair? In the event that the --repair operation is unsuccessful, please provide:
Thank you, |