statsenderd sometimes fails to read nailgun settings.yaml
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Fix Released
|
High
|
Matthew Mosesohn |
Bug Description
Source: https:/
Symptoms in /var/log/
2015-03-11 08:45:04.547 DEBUG [7fb2f50c0700] (settings) Looking for settings.yaml package config using old style __file__
2015-03-11 08:45:04.549 DEBUG [7fb2f50c0700] (settings) Trying to read config file /usr/lib/
2015-03-11 08:45:04.683 DEBUG [7fb2f50c0700] (settings) Trying to read config file /etc/nailgun/
2015-03-11 08:45:04.683 ERROR [7fb2f50c0700] (settings) Error while reading config file /etc/nailgun/
2015-03-11 08:45:04.684 DEBUG [7fb2f50c0700] (settings) Trying to read config file /etc/fuel/
2015-03-11 08:45:04.687 INFO [7fb2f50c0700] (settings) DEVELOPMENT MODE ON:
2015-03-11 08:45:04.687 INFO [7fb2f50c0700] (settings) Static dir is /usr/lib/
2015-03-11 08:45:04.687 INFO [7fb2f50c0700] (settings) Template dir is /usr/lib/
2015-03-11 08:45:05.112 INFO [7fb2f50c0700] (statsenderd) Starting standalone stats sender...
2015-03-11 08:45:05.223 ERROR [7fb2f50c0700] (statsenderd) Stats sender exception: (OperationalError) could not connect to server: Connection refused
Is the server running on host "localhost" (::1) and accepting
TCP/IP connections on port 5432?
FATAL: password authentication failed for user "nailgun"
The other daemons seem to run okay except statsenderd. This bug also seems to be quite a rare floating bug.
Changed in fuel: | |
importance: | Undecided → High |
assignee: | Fuel Python Team (fuel-python) → Artem Roma (aroma-x) |
status: | New → Confirmed |
Changed in fuel: | |
assignee: | Artem Roma (aroma-x) → Matthew Mosesohn (raytrac3r) |
The issue is not connected with functionality of the code of statsenderd but with error while reading configurations files ("/etc/ nailgun/ settings. yaml" to be precisely) in nailgun.settings module. And such error (after analyzing of traceback message) can occur in case the file is present but is empty. So I suppose we should switch focus to finding causes of that happening.