cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rana Aich <aichr...@gmail.com>
Subject Re: UnavailableException when data grows
Date Thu, 30 Sep 2010 19:45:46 GMT
I have arranged my initial tokens and get this result:

Address       Status     Load          Range
     Ring

170141183460469231731687303715880000000
192.168.202.1 Up         208.39 GB
34028236692093846346337460743177000000     |<--|
192.168.202.2 Up         333.52 GB
68056473384187692692674921486354000000     |   ^
192.168.202.3 Up         333.49 GB
102084710076281539039012382229530000000    v   |
192.168.202.4 Up         333.59 GB
136112946768375385385349842972710000000    |   ^
192.168.202.201Up         208.41 GB
144620005941398846971934208158500000000    v   |
192.168.202.202Up         83.32 GB
 153127065114422308558518573344300000000    |   ^
192.168.202.203Up         83.3 GB
161634124287445770145102938530090000000    v   |
192.168.202.204Up         83.38 GB
 170141183460469231731687303715880000000    |-->

The the range is calculated as follows:

1st Initial Token 2^127 * 0.2,
2nd Initial Token 2^127 * 0.4,
3rd Initial Token 2^127 * 0.6,
4th Initial Token 2^127 * 0.8,
5th Initial Token 2^127 * 0.85,
6th Initial Token 2^127 * 0.9,
7th Initial Token 2^127 * 0.95
8th Initial Token 2^127 * 1.0

With the top four boxes having 80% of disk space. And the bottom 4 having
20% of the total.

However as the nodetool shows my 5th box (Lowend Box) has 208.41 GB
(expected result ~ 83 GB) with 73% of the capacity already utilized. Other 3
Lowend boxes (~83GB) are behaving properly.

Does anyone have any clue?

>
On Mon, Sep 27, 2010 at 11:50 PM, Oleg Anastasyev <oleganas@gmail.com>wrote:

> Rana Aich <aichrana <at> gmail.com> writes:
>
> >
> > Yet my nodetool shows the following:
> >
> > 192.168.202.202Down       319.94 GB
> 7200044730783885730400843868815072654
>      |<--|
> > 192.168.202.4 Up         382.39 GB
> 23719654286404067863958492664769598669
>     |   ^
> > 192.168.202.2 Up         106.81 GB
> 36701505058375526444137310055285336988
>     v   |
> > 192.168.202.3 Up         149.81 GB
> 65098486053779167479528707238121707074
>     |   ^
> > 192.168.202.201Up         154.72 GB
> 79420606800360567885560534277526521273
>     v   |
> > 192.168.202.204Up         72.91 GB
>  85219217446418416293334453572116009608
>     |   ^
> > 192.168.202.1 Up         29.78 GB
>  87632302962564279114105239858760976120
>     v   |
> > 192.168.202.203Up         9.35 GB
> 87790520647700936489181912967436646309
>     |-->|
>
> It seems you did not configured tokens in your ring at all - automatic
> configuration is useless in practice.
> If you used random partitioning, for 8 nodes cluster these token
> configuration
> seems more reasonable:
> 1:      0
> 2:      21267647932558653966460912964485513216
> 3:      42535295865117307932921825928971026432
> 4:      63802943797675961899382738893456539648
> 5:      85070591730234615865843651857942052864
> 6:      106338239662793269832304564822427566080
> 7:      127605887595351923798765477786913079296
> 8:      148873535527910577765226390751398592512
>
> You can generate token configuration for any cluster size by using tool
> http://people.apache.org/~pquerna/cassandra_tokens.py
>
>
>
>
>
>
>
>
>
>

Mime
View raw message