You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@accumulo.apache.org by GitBox <gi...@apache.org> on 2018/12/07 00:35:49 UTC

[GitHub] ctubbsii commented on issue #802: Failing to recover WAL leases

ctubbsii commented on issue #802: Failing to recover WAL leases
URL: https://github.com/apache/accumulo/issues/802#issuecomment-445081428
 
 
   I've seen this copy-into-place to fix stuck Hadoop before. IIRC, it happened before 2.6 also. This seems like a Hadoop bug. Maybe we can add complexity to Hadoop to fix it, but I'd hope that the fix can be pushed upstream into Hadoop, so we don't set a precedent of Accumulo taking responsibility for bugs in libraries Accumulo uses.
   
   I'm sympathetic to the need to have this fixed in production, and don't inherently have a problem with Accumulo doing more stuff to be fault-tolerant... I'm just worried slightly that the increasing complexity in Accumulo is not a scalable solution this category of problem. Without seeing the code needed to fix this, it's hard to say whether it would be complex or simple in this case.
   
   Regardless, I definitely think this should be brought to the attention of Hadoop (or whatever Hadoop vendor you're using).

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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