[SERVER-10631] MongoDB periodically consumes excessive system cpu and refuses new connections Created: 27/Aug/13 Updated: 10/Dec/14 Resolved: 19/Mar/14 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | 2.2.6 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Michael Henson | Assignee: | Unassigned |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
Distributor ID: Ubuntu 5 member replica set (1 member is an arbiter) running on m1.large instances in AWS |
||
| Operating System: | ALL |
| Participants: |
| Description |
|
During the course of the day one of our replica-sets will appear to "lock" up on us. These locks occur every few minutes during our higher load times, but not necessarily at any regular interval. From what I've been able to determine they don't appear to be correlated with bursts in traffic or background syncs. Each of these events occur every few minutes and last anywhere between 2 and 5 seconds. During this time we see spikes in system cpu usage and the server does not accept any additional connections. I've provided some statistics from one of the most recent events below. We've run tcpdump and determined that while our clients are attempting to connect the server does not accept the connection until after the event has passed. This causes a build up in connection attempts over the course of the event many of which will be closed again upon the event's termination. We have another replica-set that is also running 2.2.6, but doesn't show the same issue. mpstat: 02:52:19 PM CPU %usr %nice %sys %iowait %irq %soft %steal %guest %idle 03:00:16 PM all 22.45 0.00 7.65 1.53 0.00 3.57 0.51 0.00 64.29 mongostat: sample connection attempt: |
| Comments |
| Comment by Stennie Steneker (Inactive) [ 19/Mar/14 ] |
|
Hi Michael, I'm going to close this issue due to inactivity. I would also note that the SERVER project is intended for reporting bugs or feature suggestions for the MongoDB server. For MongoDB-related support discussion please post on the mongodb-users group (http://groups.google.com/group/mongodb-user) or ask specific questions on Stack Overflow / ServerFault. Thanks, |
| Comment by Stennie Steneker (Inactive) [ 16/Dec/13 ] |
|
Hi Michael, Apologies for the delayed response on this issue. Are you still seeing the described behaviour? In order to investigate further we would need logs from the primary and a secondary during a period where MongoDB appears to be non-responsive. Regards, |