drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Deneche A. Hakim (JIRA)" <j...@apache.org>
Subject [jira] [Created] (DRILL-2692) Do not log errors more than once
Date Sat, 04 Apr 2015 17:11:33 GMT
Deneche A. Hakim created DRILL-2692:
---------------------------------------

             Summary: Do not log errors more than once
                 Key: DRILL-2692
                 URL: https://issues.apache.org/jira/browse/DRILL-2692
             Project: Apache Drill
          Issue Type: Improvement
    Affects Versions: 0.8.0
            Reporter: Deneche A. Hakim
            Assignee: Deneche A. Hakim
            Priority: Minor
             Fix For: 0.9.0


When a failure occurs in the execution stack it will be logged up to 6 times in the server
logs.

For example trying to query a bad json (extra ':') file will cause the following events:

- ScanBatch.next() logs a debug message "Failed to read the batch. Stopping..." and calls
FragmentContext.fail()
- FragmentContext.fail() logs an error message "Fragment Context received failure -- Fragment:
"
- FragmentExecutor.run() calls internalFail() and logs an error message when calling ErrorHelper.logAndConvertError()
  "Error <error id>: <error message>"
- Foreman.close() calls ErrorHelper.logAndConvertError() which logs an error message "Error
<error id>: <error message>"
- FragmentExecutor.closeOutResources() will throw a RuntimeException because it contains a
non null deferredException (the original exception). This exception is caught in run() and
a warning message is logged "Error while initializing or executing fragment"
- FragmentExecutor.run() calls FragmentContext.fail() when it catches the RuntimeException
which will log an error message "Fragment Context received failure -- Fragment: "



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

Mime
View raw message