"Failure: 408 Request Timeout" is sent to remote machine in response to different actions

[clotterm]

I use typica to control an Eucalyptus cloud. In some cases when I try to instantiate a new image or terminate an instance in the cloud, I'll get a

HTTP/1.1 408 Request Timeout
Content-Type: text/plain; charset=UTF-8

Failure: 408 Request Timeout

error message which causes some additional errors since typica is not able to parse this plaintext message. The strange thing is that this message is not returned in all cases, just sometimes (seems to happen randomly).

I checked the log files, nothing special which would give any hint to the error. (see the cloud-debug at http://pastebin.com/UymrpB72 for more details).

I sniffed the TCP packets with wireshark to check where the error comes from, and it is definitely the CC:
http://pastebin.com/wa8wagtk

The configuration of the whole Eucalyptus cloud is:

CC/CLC/Walrus is one machine

NC is another machine

Both machines run Ubuntu 10.04 with eucalyptus 1.6.2-0ubuntu30

[Andy Grimm (agrimm)]

This issue is now being tracked upstream at http://eucalyptus.atlassian.net/browse/EUCA-2699

Please watch that issue for further updates.