nodejs server leaves connections in CLOSE_WAIT state

Description

We saw this issue when TCP checks were performed on nodejs port. Looks like the TCP check process is not closing connections properly. This leaves a lots of connections in CLOSE_WAIT state, and the nodejs server becomes unresponsive.

Release Notes

None

Activity

Show:

Priyanka Nambiar December 5, 2014 at 9:03 PM

Unable to reproduce the issue as of now - waiting on more info. This is now lower priority for the 2.6.0 Release.

Poorna Chandra December 5, 2014 at 1:40 AM
Edited

I'm not sure what tool was used to make TCP checks. It is possible that the nagios plugin is closing connection after the check. I too was not able to reproduce this issue.

Is it possible to have a timeout for inactive connections on the server side, so that this does not happen in future?

Adrien Cahen December 4, 2014 at 6:03 AM

I tried repeatedly hitting standalone instance with TCP ping from https://www.nagios-plugins.org/ , then grepping netstat for CLOSE_WAIT:

```
$ bin/cdap.sh start
Starting Standalone CDAP ...................
Standalone CDAP started successfully.
Connect to the Console at http://ADs-MBP.local:9999

$ check_tcp 127.0.0.1 -p 9999
TCP OK - 0.000 second response time on port 9999|time=0.000227s;;;0.000000;10.000000

$ netstat -a | grep WAIT
$
```

cannot reproduce so far, it would be great to have steps to recreate the issue.

Cannot Reproduce
Pinned fields
Click on the next to a field label to start pinning.

Details

Assignee

Reporter

Affects versions

Components

Fix versions

Priority

Created November 29, 2014 at 11:26 PM
Updated March 27, 2015 at 7:30 PM
Resolved December 16, 2014 at 11:43 PM