[
https://issues.apache.org/jira/browse/HBASE-1058?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrew Purtell updated HBASE-1058:
----------------------------------
Attachment: hbase-1058-v3.patch
Attached as v3 is a patch that does not wedge under load. This will fail a
flush if there are too many store files, and will also request a compaction.
Works for me, tested under high stress.
I'm using my scan-and-heavy-write crawler/document parser scenario. It's a
mixed intensive upload and scanning workload -- very nasty.
> Prevent runaway compactions
> ---------------------------
>
> Key: HBASE-1058
> URL: https://issues.apache.org/jira/browse/HBASE-1058
> Project: Hadoop HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Blocker
> Fix For: 0.20.0
>
> Attachments: hbase-1058-v2.patch, hbase-1058-v3.patch,
> hbase-1058.patch
>
>
> A rabid upload will easily outrun our compaction ability dropping flushes
> faster than we can compact them up. Fix.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
