-
Type:
Bug
-
Resolution: Fixed
-
Priority:
Major - P3
-
None
-
Affects Version/s: 3.0.0, 3.0.1, 3.0.2, 3.0.3, 3.0.4, 3.0.5
-
Component/s: Native
-
Environment:any platform
-
(copied to CRM)
-
None
-
None
-
None
-
None
-
None
-
None
-
None
While node is running, connections to mongos increase continuously after mongos server restarted.
When I use '"mongodb": "^3.0.0"'~'"mongodb": "^3.0.6", connections do not destroyed after mongos restarted.
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
0 0 0
[ID@MONGOS_SERVER ~]$ date
Wed Apr 11 17:49:37 KST 2018
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
7 42 623
[ID@MONGOS_SERVER ~]$ pkill mongos
[ID@MONGOS_SERVER ~]$ numactl --interleave=all mongos -f /workspace/db/config/mongos.conf &
[1] 155819
[ID@MONGOS_SERVER ~]$ about to fork child process, waiting until server is ready for connections.
forked process: 155821
child process started successfully, parent exiting
[1]+ Done numactl --interleave=all mongos -f /workspace/db/config/mongos.conf
[ID@MONGOS_SERVER ~]$ date
Wed Apr 11 17:49:56 KST 2018
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
14 84 1246
[ID@MONGOS_SERVER ~]$ date
Wed Apr 11 17:52:45 KST 2018
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
14 84 1246
[ID@MONGOS_SERVER ~]$ pkill mongos
[ID@MONGOS_SERVER ~]$ numactl --interleave=all mongos -f /workspace/db/config/mongos.conf &
[1] 156799
[ID@MONGOS_SERVER ~]$ about to fork child process, waiting until server is ready for connections.
forked process: 156801
[ID@MONGOS_SERVER ~]$ child process started successfully, parent exiting
[1]+ Done numactl --interleave=all mongos -f /workspace/db/config/mongos.conf
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
21 126 1869
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
27 162 2403
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
27 162 2403
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
27 162 2403
However, when I use '"mongodb": "^2.2.35", it's normal after mongos restarted.
[ID@MONGOS_SERVER ~]$ date
Wed Apr 11 17:59:12 KST 2018
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
7 42 623
[ID@MONGOS_SERVER ~]$ pkill mongos
[ID@MONGOS_SERVER ~]$ numactl --interleave=all mongos -f /workspace/db/config/mongos.conf &
[1] 158835
[ID@MONGOS_SERVER ~]$ about to fork child process, waiting until server is ready for connections.
forked process: 158837
[ID@MONGOS_SERVER ~]$ child process started successfully, parent exiting
[1]+ Done numactl --interleave=all mongos -f /workspace/db/config/mongos.conf
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
0 0 0
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
7 42 623
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
7 42 623
[ID@MONGOS_SERVER ~]$ date
Wed Apr 11 17:59:52 KST 2018
[ID@MONGOS_SERVER ~]$ pkill mongos
[ID@MONGOS_SERVER ~]$
[ID@MONGOS_SERVER ~]$ numactl --interleave=all mongos -f /workspace/db/config/mongos.conf &
[1] 159900
[ID@MONGOS_SERVER ~]$ about to fork child process, waiting until server is ready for connections.
forked process: 159902
[ID@MONGOS_SERVER ~]$ child process started successfully, parent exiting
[1]+ Done numactl --interleave=all mongos -f /workspace/db/config/mongos.conf
[ID@MONGOS_SERVER ~]$
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
0 0 0
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
6 36 534
[ID@MONGOS_SERVER ~]$ netstat | grep ESTABLISHED | grep 64 | grep 144 | wc
7 42 623
[ID@MONGOS_SERVER ~]$ date
Wed Apr 11 18:00:11 KST 2018
- is caused by
-
NODE-1595 Failure to clean up underlying topology when failing to initially connect
-
- Closed
-