accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [accumulo] EdColeman opened a new issue #1325: Is checking disk status in FileSystemMonitor necessary?
Date Tue, 13 Aug 2019 15:20:45 GMT
EdColeman opened a new issue #1325: Is checking disk status in FileSystemMonitor necessary?
URL: https://github.com/apache/accumulo/issues/1325
 
 
   At this point, hoping for a discussion to determine requirements / scenarios that may warrant
changes.  Scenario: When running on servers with multiple disks, the FileSystemMonitor will
kill a tserver if a disk fails (goes read-only)  
   - Is this necessary behavior - or could it be handled by terminating on appropriate IO
Exceptions? 
   - What are the Accumulo critical file / volume paths? Did this change when WALs moved to
hdfs?
   - How critical is "logging" and should a failure of slf4j/log4j to write a log cause a
termination of a process (or could log frame be configure with multiple volumes for fail-over?)
   - HDFS (data nodes) now supports allowing x volume (disk) failures before killing that
data node - should we allow the same?
   - Does containerization changes influence any other this? 

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services

Mime
View raw message