cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jérémy Sevellec (JIRA) <>
Subject [jira] [Commented] (CASSANDRA-2961) Expire dead gossip states based on time
Date Thu, 08 Sep 2011 07:21:08 GMT


Jérémy Sevellec commented on CASSANDRA-2961:

I found why i have make 2 call of addExpireTimeIfFound into SS instead of calling it into
excise : 
There is 3 calls to excise in SS : handleStateLeft, handleStateRemoving and... removeToken.
In removeToken, we don't have the "pieces" of the VV which contain expireTime. So we can't
extract an expireTime.

so there is 3 possibilities : 
- modify excise to add "pieces" parameter, set it to null in the removeToken call and manage
the case if pieces is null. I find this solution not so beautiful but will work.
- refactor and create 2 method signature of excise : one with pieces parameter and one without.
- keep as it is.

It's as you want. Tell me what is your preference (or another).

> Expire dead gossip states based on time
> ---------------------------------------
>                 Key: CASSANDRA-2961
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>    Affects Versions: 1.0
>            Reporter: Brandon Williams
>            Priority: Minor
>             Fix For: 1.0
>         Attachments: trunk-2961-v2.patch, trunk-2961.patch
> Currently dead states are held until aVeryLongTime, 3 days.  The problem is that if a
node reboots within this period, it begins a new 3 days and will repopulate the ring with
the dead state.  While mostly harmless, perpetuating the state forever is at least wasting
a small amount of bandwidth.  Instead, we can expire states based on a ttl, which will require
that the cluster be loosely time synced; within the quarantine period of 60s.

This message is automatically generated by JIRA.
For more information on JIRA, see:


View raw message