Greetings,
I'm interested in having cassandra client installed separately. I thought that
java classes in cassandra-all.jar are those needed by cassandra server. Also
that the cassandra-clientutil.jar would hold all the necessary classes to run
nodetool and all other sstabletools. My asumptions
Hi,
I have a second index on a static column and I don't
understand the answer I get from my select.
Maybe someone who understands the inner working of the second index can
give me a hint on this (cassandra 3.9)
A cut down version of the table is:
create table demo (id text, id2 bigint static,
Hi Michael,
It sound like a bug but I could not reproduced it on 3.9 or on the current
3.11 branch (which will become 3.10).
Now, that does not mean that there are not problem. Something might be
different in your environement.
Do you still see that problem when you start from a clean environment
Hi,
my cluster is quite new, with three (jessie) nodes and only some test
tables with a few rows of data in it.
I just started a fresh one-node-cluster on another machine, created the
table then the second index on the static column and inserted one row of
data.
create table demo (id text, adde
Hi,
forget my the last mail. On the single node cluster it works.
I can try it on the three-node cluster with a keyspace with replication
factor of 1 and see what happens.
I left most of the default config of cassandra untouched, except storage
directories and ip addresses.
Cheers,
Michael
Sorry,
I am not sure that I understand you correctly. Do you see the problem on a
one Node cluster or not?
On Mon, Jan 30, 2017 at 12:05 PM, Micha wrote:
> Hi,
>
> forget my the last mail. On the single node cluster it works.
>
> I can try it on the three-node cluster with a keyspace with repl
I have restarted the three node cluster with new directories for data
and commitlog and made the test again.
This time the resultset size is 62 rows for the select.
If I execute the select often it jumps between 62 and 65 rows.
After inserting a second row I get 129 rows back I select using the i
I will try to reproduce the problem.
Thanks for the information
On Mon, Jan 30, 2017 at 2:34 PM, Micha wrote:
> I have restarted the three node cluster with new directories for data
> and commitlog and made the test again.
>
> This time the resultset size is 62 rows for the select.
> If I execut
So, far I do not have any sucess in trying to reproduce the problem.
I created a 3 node clusters using 3.9 with ccm and used CQLSH to reproduce
the problem but I only got back one row:
cqlsh> create KEYSPACE test WITH replication = {'class': 'SimpleStrategy',
'replication_factor': 3};
cqlsh> use
OK, thanks, that was good!
You have allocated the keyspace with replication factor 3. If I do this
it works on my cluster too!
If I try this in a new keyspace with replication factor 2 I get the same
result as before, nearly at least, this time 58 rows.
I can reproduce this: 3-node cluster and r
I do not see why you should get that problem with a replication factor of 2.
I will look into it.
On Mon, Jan 30, 2017 at 5:06 PM, Micha wrote:
> OK, thanks, that was good!
>
> You have allocated the keyspace with replication factor 3. If I do this
> it works on my cluster too!
>
> If I try this
Hello list,
why is datastax java-driver so tightly fixed to the cassandra base?
Is there any reason to have it as a shaded dependency?
And what even a shaded dependency is?
I'd be glad for any response.
Thanks Tomas
Some of the tools use the driver, and there is some reliance on classes
for user defined types that come from the driver. I suppose ideally
these would get split out into a cassandra-common.jar, or such.
"Shaded" in this case means that the jar contains code from netty, but
the package struct
13 matches
Mail list logo
