Since it's not in cfstats anymore, is there another way to monitor this?
I'm working with a dev cluster and I've got Opscenter set up, so I tried
taking a look through that, but it just shows "NO DATA." Does that mean the
key cache isn't enabled? I haven't changed the defaults there, so the key
ca
You still add one row. The column name is the remaining part of the composite
key (repeat for each column) plus each of the column which is not in the
composite key. I found it is much clearer to look at the data through Cassandra
-cli which shows you how data is stored.
Thanks.
-Wei
Sent f
Jmx has cache hit rate.
Thanks.
-Wei
Sent from my Samsung smartphone on AT&Tnull
Agreed. I actually flip between cli and cqlsh these days.
cqlsh shows the logical view.
cli shows the physical view.
This is useful, especially when developing using a thrift-based client.
Here are the slides and video if you want to have a look.
-brian
On Dec 22, 2012, at 3:36 AM, Wz1975
What I saw in all cases was
a) set JAVA_HOME to java7, run program fail
b) set JAVA_HOME to java6, run program success
I should have better notes but I'm at a 6 person startup so working tools
gets used and failing tools get deleted.
Brian
On Fri, Dec 21, 2012 at 3:54 PM, Bryan Talbot wrote:
>
Hey everyone,
I'm seeing some conflicting advice out there about whether you need to run
nodetool repair within GCGraceSeconds with 1.x. Can someone clarify two
things:
(1) Do I need to run repair if I'm running 1.x?
(2) Should I bother running repair if I don't have any deletes? Anything
drawbac
Thanks, I'll take a look at that too.
I also found that "nodetool info" gives some information as well. For
instance, here's what one node reads:
Key Cache: size 104857584 (bytes), capacity 104857584 (bytes), 15085408
hits, 17336031 requests, 0.870 recent hit rate, 14400 save period in
seconds.
The key cache and row cache graphs in OpsCenter broke when the caches moved
from per column family to global. The next release of OpsCenter should fix
that.
-Nick
On Sat, Dec 22, 2012 at 10:42 AM, Andrew Bialecki wrote:
> Thanks, I'll take a look at that too.
>
> I also found that "nodetool in
Is there still a way to have composite row keys ?
There are times when you want to partition wide rows by a tuple instead of
pushing the composites into column names.
Lists could do the trick but would not allow multiple types and aren't
allowed as primary keys anyhow.
At some point I remember se
