Hi Toby -
Are you using the stock bitcask configuration for merging?
--
Luke Bakken
Engineer
lbak...@basho.com
On Tue, Jan 20, 2015 at 5:07 PM, Toby Corkindale wrote:
> Hi Kota,
> I had a bit of an off-list chat about this a while ago, plus continued
> to investigate locally, and eventually ach
Hi Kota,
I had a bit of an off-list chat about this a while ago, plus continued
to investigate locally, and eventually achieved some faster speeds,
around 15MByte/sec writes.
Things that were changed:
* Adjusted Riak CS GC to be spread out over the cluster much more.
* Tweaked up the put buffers
Toby and David,
Thank you for working on Riak CS and I apologize for being late responder.
I believe the reason of being slow down is different between Toby's
and David's cluster.
Toby's reason looks like that's just because of the data increasing.
How much data per vnode do you have in your clu
Hi Claudio,
The "indexes_not_supported" error means you attempted to write
secondary index information to a bucket that is stored in a backend
that does not support indexes.
Could you please confirm that the bucket you're using for this data is
set to use the "eleveldb_mult" backend? You will hav
Hello,
I am trying to build a small demo app involving the Java Client and the Data
types by following the official documentation.
However, map operations are not working. A fetch with the Java client always
returns an empty map, writes create new maps, but they remain empty.
In the source file
Got it Brain, Thanks.
Any suggestions on the 2nd point.
On Tue, Jan 20, 2015 at 10:43 PM, Brian Roach wrote:
> Santi -
>
> The core of the 2.0 Java client uses Netty. There's a fixed number of
> worker threads that process sockets using non-blocking IO ( w/
> polling/select).
>
> Due to the sy
Cosmin -
Link walking is deprecated as of Riak 2.0 and will be removed in a
future version.
As to whether something similar will replace it, as of now I do not
believe we have anything on the roadmap, no.
Thanks,
- Roach
On Tue, Jan 20, 2015 at 10:24 AM, Cosmin Marginean wrote:
> (Apologies
(Apologies if this is a recurring topic, but I haven’t read a clear statement
yet in relation to this)
Using Riak, I sometimes feel that link walking might be a corner stone for
certain data modelling techniques. The Riak documentation though states clearly
that this is not feasible while also
Santi -
The core of the 2.0 Java client uses Netty. There's a fixed number of
worker threads that process sockets using non-blocking IO ( w/
polling/select).
Due to the synchronous nature of the Riak API, each fetch operation
requires its own socket connection and we can't pipeline.
As noted in
Thanks.
On Tue, Jan 20, 2015 at 5:07 PM, Russell Brown wrote:
> FYI https://github.com/basho/riak/issues/667
>
> I’ll get to work on it at once, we might be able to hit the 2.0.5 build
> date at the end of the week.
>
> On 20 Jan 2015, at 14:04, Russell Brown wrote:
>
> > I’ll open a ticket for
FYI https://github.com/basho/riak/issues/667
I’ll get to work on it at once, we might be able to hit the 2.0.5 build date at
the end of the week.
On 20 Jan 2015, at 14:04, Russell Brown wrote:
> I’ll open a ticket for it. We changed the internal structure of both Maps and
> Sets, and tested t
I’ll open a ticket for it. We changed the internal structure of both Maps and
Sets, and tested that they were backwards compatible, but clearly missed
something.
On 20 Jan 2015, at 13:54, Alexander Popov wrote:
> Upgraded riak 2.0.2-> 2.0.4
> trying to add nodes to cluster( was single node )
Upgraded riak 2.0.2-> 2.0.4
trying to add nodes to cluster( was single node )
all new nodes located on another host, but multi-nodes installation from
source
each new node have config settings tuned like
nodename = riak1@10.0.0.28
platform_data_dir = /var/lib/riak1
listener.http.internal = 0.0.0.0
13 matches
Mail list logo
