Github user mikewalch commented on a diff in the pull request:

    https://github.com/apache/accumulo/pull/221#discussion_r102999845
  
    --- Diff: assemble/conf/log4j-service.properties ---
    @@ -15,42 +15,48 @@
     
     # Write out everything at the DEBUG level to the debug log
     log4j.appender.A2=org.apache.log4j.RollingFileAppender
    
-log4j.appender.A2.File=${org.apache.accumulo.core.dir.log}/${org.apache.accumulo.core.application}${accumulo.service.instance}_${org.apache.accumulo.core.ip.localhost.hostname}_fromprops.debug.log
    -log4j.appender.A2.MaxFileSize=1000MB
    -log4j.appender.A2.MaxBackupIndex=10
    
+log4j.appender.A2.File=${accumulo.log.dir}/${accumulo.application}${accumulo.service.instance}_${accumulo.local.hostname}.debug.log
    +log4j.appender.A2.MaxFileSize=100MB
    +log4j.appender.A2.MaxBackupIndex=5
     log4j.appender.A2.Threshold=DEBUG
     log4j.appender.A2.layout=org.apache.log4j.PatternLayout
     log4j.appender.A2.layout.ConversionPattern=%d{ISO8601} [%-8c{2}] %-5p: %m%n
     
     # Write out INFO and higher to the regular log
     log4j.appender.A3=org.apache.log4j.RollingFileAppender
    
-log4j.appender.A3.File=${org.apache.accumulo.core.dir.log}/${org.apache.accumulo.core.application}${accumulo.service.instance}_${org.apache.accumulo.core.ip.localhost.hostname}_fromprops.log
    -log4j.appender.A3.MaxFileSize=1000MB
    -log4j.appender.A3.MaxBackupIndex=10
    
+log4j.appender.A3.File=${accumulo.log.dir}/${accumulo.application}${accumulo.service.instance}_${accumulo.local.hostname}.log
    +log4j.appender.A3.MaxFileSize=100MB
    +log4j.appender.A3.MaxBackupIndex=5
    --- End diff --
    
    They were set to keep 20GB (10G for log & 10G for debug.log) of logs for 
each Accumulo service (tserver, master, gc, etc).  I understand that a user 
might want to do that on a beefy production node but I don't think it should be 
the default for users out of the box.  I have run into the problem of disks 
filling up due to logs when running a SSD VMs in AWS if these defaults are 
used.  This change will keep up to 1GB of logs per service on each node.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---

Reply via email to