Exports do not purge

Summary

Purging old exports takes a long time and eventually times out

Environment

HipChat Server 2.2.2 and older

Steps to Reproduce

1. Run a purge of exports older than, say 3 days

   hipchat export -u 3
   

2. The purge errors out with a message like:

   Couldn't delete 251c4087-d524-48c3-88e4-478bbac18ecb, please look at the hipchat logs for more information: ('Connection aborted.', BadStatusLine("''",))
   

3. Purge the export by posting to the service directly:

   time curl -vvv -XDELETE http://localhost:7777/export/251c4087-d524-48c3-88e4-478bbac18ecb
   

4. Curl will run for 30 seconds and time-out:

   * Hostname was NOT found in DNS cache
   *   Trying 127.0.0.1...
   * Connected to localhost (127.0.0.1) port 7777 (#0)
   > DELETE /export/251c4087-d524-48c3-88e4-478bbac18ecb HTTP/1.1
   > User-Agent: curl/7.35.0
   > Host: localhost:7777
   > Accept: */*
   >
   * Empty reply from server
   * Connection #0 to host localhost left intact
   curl: (52) Empty reply from server
   
   real	0m30.066s
   user	0m0.024s
   sys	0m0.013s
   

Expected Results

Old exports should be purged and archive removed from /file_store/cumulus/posixdata/f/

Actual Results

The below exception is thrown in the hcapp.log file:

2017-02-22T14:20:32.289289+00:00 hipchat.rgoffice.net gunicorn.scissortail-web.error: [7308] WORKER TIMEOUT (pid:7378)
2017-02-22T14:20:32.289289+00:00 hipchat.rgoffice.net gunicorn.scissortail-web.error: [7378] Worker exiting (pid: 7378)

Workaround

Increase the worker timeout to 120s in /hipchat-scm/scissortail/gunicorn.conf.py:

timeout = int(os.environ.get('WEB_WORKER_TIMEOUT', 120))