[
https://issues.apache.org/jira/browse/CASSANDRA-8009?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14161740#comment-14161740
]
Andreas Schnitzerling commented on CASSANDRA-8009:
--------------------------------------------------
Hello,
I'm importing Gigabytes of logdata. I have 12 nodes, 2 running with C*
2.1-release and 10 with C* 2.0.10. Because of a failure in importing process
before, I delete the data and write again. I do not replace all data, so I
wouldn't delete the whole CF. The interesting part is, that the space of files
is growing in the same amount like on the 2.0.10-nodes but nodetool status is
showing me less and less load compared to the other nodes. I expect negative
values in the future like I read somewhere else with C* 2.1. Here is the output
of nodetool status:
UN 10.9.9.69 67,19 GB 256 11,1%
UN 10.6.8.107 62,06 GB 256 10,1%
UN 10.7.8.144 32,13 GB 128 4,3%
UN 10.6.8.9 2,47 GB 256 10,3%
UN 10.6.8.127 4,5 GB 256 10,7%
UN 10.6.8.212 60,88 GB 256 10,5%
UN 10.7.8.124 33,68 GB 128 3,8%
UN 10.7.8.140 60,58 GB 128 5,7%
UN 10.9.9.151 58,58 GB 256 10,0%
UN 10.9.9.241 54,84 GB 256 9,7%
UN 10.7.8.138 35,03 GB 128 4,6%
UN 10.9.9.240 58,83 GB 256 9,3%
Nodes 10.6.8.9 and 10.6.8.127 (both C* 2.1, rest C* 2.0.10) should have
something between 54GB and 61GB like the other nodes with 256 tokens.
I hope there's no impact to the data...
Env: Windows-7 32bit, Java 7 (1.7.0_67)
> nodetools -h host status is showing wrong load
> ----------------------------------------------
>
> Key: CASSANDRA-8009
> URL: https://issues.apache.org/jira/browse/CASSANDRA-8009
> Project: Cassandra
> Issue Type: Bug
> Components: Core, Tools
> Environment: Cassandra: Cassandra 2.1.0
> OS: Red Hat Enterprise Linux Server release 6.5
> Java: Java(TM) SE Runtime Environment (build 1.7.0_55-b13)
> Java: Java HotSpot(TM) 64-Bit Server VM (build 24.55-b03, mixed mode)
> Reporter: leandro moreira
> Fix For: 2.1.0
>
>
> When I use *nodetool status* it shows me a load very different from *df -h*
> outuput.
> {quote}
> $nodetool -h anyofmyhost status
> ===============
> Status=Up/Down
> |/ State=Normal/Leaving/Joining/Moving
> -- Address Load Tokens Owns Host ID
> Rack
> UN 1 1.67 TB 256 23.2% 6216e547-xxx RAC1
> UN 2 1.86 TB 256 24.3% 835c566d-xxx RAC1
> [user@anyofmyhost ~]$ df -h
> Filesystem Size Used Avail Use% Mounted on
> ...
> /dev/sda6 1.6T 70G 1.5T 5% /opt
> ...
> {quote}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
