Comment 49 for bug 1253896

Revision history for this message
Salvatore Orlando (salvatore-orlando) wrote :

79 of the 93 failures observed in the last 24 hours happen in the newly merged load balacing scenario: http://logstash.openstack.org/#eyJmaWVsZHMiOltdLCJzZWFyY2giOiJmaWxlbmFtZTpcImNvbnNvbGUuaHRtbFwiIEFORCBtZXNzYWdlOlwidGVtcGVzdC5zY2VuYXJpby50ZXN0X2xvYWRfYmFsYW5jZXJfYmFzaWMuVGVzdExvYWRCYWxhbmNlckJhc2ljLnRlc3RfbG9hZF9iYWxhbmNlcl9iYXNpY1tjb21wdXRlLGdhdGUsbmV0d29yayxzbW9rZV0gLi4uIEZBSUxcIiIsInRpbWVmcmFtZSI6Ijg2NDAwIiwiZ3JhcGhtb2RlIjoiY291bnQiLCJvZmZzZXQiOjAsInRpbWUiOnsidXNlcl9pbnRlcnZhbCI6MH0sInN0YW1wIjoxMzkxNzcxNDg5Mzc0fQ==

the failure occurs while trying to connect to a worker VM to start a fake web server with netcat.
This time the connection happens over the private IP, so the l3 agent does not get into the picture. The port appears to be correctly wired in all analyzed failures. Therefore we need to look at potentiall sec group/dhcp issues.