cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <>
Subject [jira] Commented: (CASSANDRA-715) HHOM goes into infinite loop, wasting cpu
Date Tue, 19 Jan 2010 15:42:54 GMT


Jonathan Ellis commented on CASSANDRA-715:

until this is fixed, running nodeprobe cleanup on each live node & restarting it should
fix this (by removing undelivered hints forcibly).  depending on your replication factor you
may have hints that are not removed by cleanup; in that case you can remove the hint files
from data/system/*Hint*.

> HHOM goes into infinite loop, wasting cpu
> -----------------------------------------
>                 Key: CASSANDRA-715
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.6
>         Environment: debian lenny amd64 OpenJDK 64-Bit Server VM (build 1.6.0_0-b11,
mixed mode)
>            Reporter: Brandon Williams
>             Fix For: 0.6
> To replicate: take a host down, cause hints to it, wait for HHOM to kick in
> The issue is line 201 of HHOM:
> startColumn =; // repeating the last as the first is fine since we just
deleted it
> That comment is false.  The column may not have been deleted, since the endpoint could
still be down.  This causes HHOM to go into an infinite loop trying to deliver hints to a
down host.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message