[SERVER-1628] Excessive CPU usage by available replicaset nodes when one node is down Created: 13/Aug/10 Updated: 12/Jul/16 Resolved: 14/Aug/10 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Performance, Replication |
| Affects Version/s: | 1.6.0 |
| Fix Version/s: | 1.6.1, 1.7.0 |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | AndrewK | Assignee: | Dwight Merriman |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
Windows 7 64bit |
||
| Attachments: |
|
| Operating System: | ALL |
| Participants: |
| Description |
|
In a replicaset of 3 idle nodes, when one node is unavailable, the remaining two nodes' processes spike to about 60%-80% CPU usage. The spikes seem to occur in an alternating fashion once a second. Bringing the missing node back online results in all nodes' processes returning to a consistent <1% cpu usage. The attached graph shows the processes spiking on CPU #2 (right most graph) and then returning to normal at the end when the 3rd node is brought back online. |
| Comments |
| Comment by auto [ 16/Aug/10 ] |
|
Author: {'login': 'erh', 'name': 'Eliot Horowitz', 'email': 'eliot@10gen.com'}Message: BackgroundJob::wait bug Conflicts: util/background.cpp |
| Comment by Dwight Merriman [ 13/Aug/10 ] |
|
1.6.1? |
| Comment by auto [ 13/Aug/10 ] |
|
Author: {'login': 'dwight', 'name': 'dwight', 'email': 'dwight@10gen.com'}Message: BackgroundJob::wait bug |
| Comment by auto [ 13/Aug/10 ] |
|
Author: {'login': 'dwight', 'name': 'dwight', 'email': 'dwight@10gen.com'}Message: make BackgroundJob less goofy in its implementation. towards |
| Comment by Dwight Merriman [ 13/Aug/10 ] |
|
reproduced. |