You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@sentry.apache.org by "Lenni Kuff (JIRA)" <ji...@apache.org> on 2015/11/03 08:17:27 UTC

[jira] [Created] (SENTRY-949) Cleanup logging of major HDFS sync events

Lenni Kuff created SENTRY-949:
---------------------------------

             Summary: Cleanup logging of major HDFS sync events 
                 Key: SENTRY-949
                 URL: https://issues.apache.org/jira/browse/SENTRY-949
             Project: Sentry
          Issue Type: Task
    Affects Versions: 1.6.0
            Reporter: Lenni Kuff


I recently was debugging an HDFS sync issue and found the existing logging is not super clear for important events. Most of this can be interpreted from the logging that's there, but it would be good to have very clear messages for when significant events happen:

# Be clear on whether HDFS sync is enabled/disabled: "Sentry HDFS Sync is: [ENABLED|DISABLED]" in the NN plugin, HMS plugin, and Sentry service.
# Log a message "Completed loading of {# of paths} HMS paths in {X} seconds. Sending full update to Sentry Service {hostname}." in the HMS plugin.
# Log the total number of failed heartbeat attempts and time the metadata cache has been stale in the NN plugin (once a minute or so).
# Remove the " !!" suffixes from a lot of the log messages.

I'm sure there are other things, but these seem like a good place to start. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)