cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ashic Mahtab (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-7239) Nodetool Status Reports Negative Load With VNodes Disabled
Date Tue, 29 Jul 2014 15:17:39 GMT


Ashic Mahtab commented on CASSANDRA-7239:

I set up a low powered cluster (2GB ram, shared disks) of four nodes and put opscenter on
a fifth. All running on vmware. It's using vnodes - 256 tokens per node. I got nodetool output
shown in nodetool.png. It shows negative load on a node. At times, we even saw two nodes with
negative load. Restarting the offfending node fixed the issue. We say a slight negative node
again at a different time. It self healed though.

There's no data in the nodes (as in, the cluster doesn't have any data put in). It's been
running blank for a few days...the data in there is just cassandra's / opscenter's internal

The negative loads have an interesting impact on opscenter, as can be seen in opscenter.png!!

All nodes are using 2.1.0-rc4. Opscenter is v5.0.


> Nodetool Status Reports Negative Load With VNodes Disabled
> ----------------------------------------------------------
>                 Key: CASSANDRA-7239
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Tools
>         Environment: 1000 Nodes EC2 m1.large ubuntu 12.04
>            Reporter: Russell Alexander Spitzer
>            Priority: Minor
>         Attachments: nodetool.png, opscenter.png
> When I run stress on a large cluster without vnodes (num_token =1 initial token set)
The loads reported by nodetool status are negative, or become negative after stress is run.
> {code}
> UN    -447426217 bytes  1       0.2%    8d40568c-044c-4753-be26-4ab62710beba
> UN     -447342449 bytes  1       0.2%    58e7f255-803d-493b-a19e-58137466fb78
> UN   -447298672 bytes  1       0.2%    ba29b1f1-186f-45d0-9e59-6a528db8df5d
> {code}

This message was sent by Atlassian JIRA

View raw message