ResponseNotReady

[peterboett]

Hello Supporter,

we are going to use SPLUNK. We have installed it on a machine and testet in our office and it worked fine. On weekend the server was installed in our cage in the datacenter. Now the server frontend won't speak with us. It throws the error 500 Internal Server Error (ResponseNotReady). In our office everything worked fine.

I've checked the web_access.log, web_service.log, splunkd_access.log and the splunkd.log, everything is nice in there and no errors are detectable. I've also checked it via TCP Dump, but the HTTP communication seems to be fine.

I'm using an SSL VPN connection to connect to our datacenter.

The distribution we use is a Debian (Kernel 2.6.32) and Splunk 5 (5.0-140868-linux-x86_64).
All i've found out is that this should be an error with phyton, but i'm not a programmer. Can you please give me a hint on what I should look? Any suggestions?
I've also done a fresh install because I do not really know if an ip change will cause trouble.
Let me know if you need more informations or maybe a logfile.

Thanks in advance

Peter

EDIT: The event is thrown in httplib.py on line 1018.

[Jason]

This error appears on Windows sometimes due to Splunk being unable to talk to itself. It's a known issue with Windows, see:

http://docs.splunk.com/Documentation/Splunk/latest/ReleaseNotes/Workaroundfornetworkaccessibilityiss...

I haven't heard of local ephemeral port exhaustion on linux systems though. The splunkd.log should contain a python stack trace containing more information related to the error.