tynyttie commented on pull request #1803:
URL: https://github.com/apache/accumulo/pull/1803#issuecomment-755036001
Process I used for debugging this issue. Note doing this process I tailed
the logs for new info.
1. In RecoveryPath.java, I hard coded a string and notice that RestartIt
passed with no issues.
Example: walPath = Path(walPath, FileType.RECOVERY.getDirectory() + '-' +
'addedFileName');
2. In RecoveryPath.java., I used StringBuffer, StringWriter, and normal
String creation to build a new string.
Example: testSting = FileType.RECOVERY.getDirectory() + '-'
'namer.getNextName();
Path newPath = new Path(walPath, testString); ------
RestartIT failed with this approach.
3. In RecoveryPath.java, instead of using UniqueNameAllocator to generate
the name, I create a new name generate in order to make sure the name was
present when needed. --- RestartIT failed with this approach.
4. I used Jstack to help track down the cause of threads stuck in wait
state. With this approach, I couldn't pinpoint the true origin.
5. I connected to the server and added break statement to walk through the
code. This approach was unsuccessful. None of my break statements was picked up
by the Tserver.
6. In LogSorter, I added a sleep where the issue is to make sure the process
doesn't end. This approach was unsuccessful.
The issues are on: (LogSorter.java:197) and (LogSorter.java:167)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
