[SERVER-20125] ReplicaSet Cluster log lots of "9001 SocketException" Created: 26/Aug/15 Updated: 27/Apr/18 Resolved: 03/Sep/15 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Replication |
| Affects Version/s: | 3.0.4 |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | YANG Chenghu | Assignee: | Sam Kleinman (Inactive) |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
Linux Redhat6 |
||
| Issue Links: |
|
||||||||
| Participants: | |||||||||
| Description |
|
Deployed a ReplicaSet Cluster, watch mongod.log, log "9001 socket exception [CLOSED]" every 30 seconds log content:
Check source code, about cleanUpOlderThan_inlock. MongoDB ConnectionPool will close the connection that before 30 seconds creating, but, it's not pretty. it should be it not used in 30 seconds |
| Comments |
| Comment by YANG Chenghu [ 25/Sep/15 ] | ||||||||
|
thanks for you reply, I think mongod closed the socket connection every 30 seconds is not sensible.
| ||||||||
| Comment by Sam Kleinman (Inactive) [ 03/Sep/15 ] | ||||||||
|
It looks like you have set systemLog.traceAllExceptions to true. You can change the presence of the backtrace data at runtime using the traceExceptions parameter. You may also want to consider configuring or modifying your own logrotation system to control how you persist and store log data. The SERVER project is for tracking bugs and issue requests related to the core MongoDB server product. Questions like this are probably best addressed to one of our support channels. Regards, |