Example failure:
http://cautotest.corp.google.com/afe/#tab_id=view_job&object_id=58263974

Reading from the suite's DEBUG log, I see:

03/29 03:49:44.822 INFO |        server_job:0128| INFO	----	Start sanity	timestamp=1459248584	localtime=Mar 29 03:49:44	
03/29 03:49:44.822 DEBUG|             suite:0824| Scheduling dummy_PassServer.sanity
03/29 03:49:45.950 DEBUG|             suite:1078| Adding job keyval for dummy_PassServer.sanity=58264253-chromeos-test
03/29 03:49:45.951 DEBUG|     dynamic_suite:0613| Waiting on suite.
03/29 03:49:46.185 INFO |         discovery:0190| URL being requested: https://monorail-prod.appspot.com/_ah/api/discovery/v1/apis/monorail/v1/rest

The last thing it says is that it's making some monorail requeset. Looks like the request never returned, and thus the suite was hung in the create-bug-filing-handle stage. The builder ended up timing out and killing the suite.

Can we add a timeout to this bug filing stuff so that if the request is frozen, we don't freeze the entire suite?