I'm also quite interested in this question. Here's my understanding on this
problem.
1. If your workload is append-only, doing a major compaction shouldn't
affect the read performance too much, because each row appears in one
sstable anyway.
2. If your workload is mostly updating existing rows, t
The tpstats shows there was no read request pending on this node. Maybe you
should have a look at the other nodes first.
But my suggestion is to upgrade hector to 1.0.x if possible. Hector 0.7 is
for cassandra 0.7. Just FYI, we had some issues when upgrading cassandra
from 0.8 to 1.0. And the prob
It needs to stream all its data to the other node, which may take a long
time. You can use nodetool netstats to see the progress of streaming.
Best,
Cheng
On Sun, Apr 22, 2012 at 4:11 AM, Aniket Chakrabarti <
chakr...@cse.ohio-state.edu> wrote:
> Hi,
>
> I am trying to use nodetool decommission
Hi. We got the same problem here. Even the wordcount map/reduce example in
the source tar works fine with one node, but fails with the same exception
on a two node cluster. CASSANDRA-3044 mentioned that a temporary work around
is to disable node auto discovery. Can anyone tell me how to do that in
