While loading some data into a test cluster I using riak 1.0.0pre2 I came
across some sort of search related bug.
The cluster is configured with 3 nodes, N is 3, W is 1, WD is 0, and I am
connecting to protocol buffers. The program to load data can be configured
to use one or more connections usi
Is there some way to drop a search index, other that forcibly removing the
merge_index data directory on each node?
Change a bucket's search property to false or executing search-cmd uninstall
do not appear to have this effect. I can continue to execute queries
against the index. They appear to
On Wed, Sep 14, 2011 at 9:21 PM, Jeremiah Peschka
wrote:
> I should also note that Riak-1.0.0pre2 builds beautifully.
So you're building from the provided source archive or a tag?
D.
>
> Jeremiah Peschka
> Founder, Brent Ozar PLF
>
> On Sep 14, 2011 10:19 PM, "Jeremiah Peschka"
> wrote:
>> Bas
I should also note that Riak-1.0.0pre2 builds beautifully.
Jeremiah Peschka
Founder, Brent Ozar PLF
On Sep 14, 2011 10:19 PM, "Jeremiah Peschka"
wrote:
> Bash
>
> Jeremiah Peschka
> Founder, Brent Ozar PLF
> On Sep 14, 2011 10:17 PM, "David Smith" wrote:
>> Out of curiosity, what shell are you u
Bash
Jeremiah Peschka
Founder, Brent Ozar PLF
On Sep 14, 2011 10:17 PM, "David Smith" wrote:
> Out of curiosity, what shell are you using? That looks like a shell
> problem to me...
>
> D.
>
> On Wed, Sep 14, 2011 at 7:25 PM, Jeremiah Peschka
> wrote:
>> I get errors trying to build eleveldb on
Out of curiosity, what shell are you using? That looks like a shell
problem to me...
D.
On Wed, Sep 14, 2011 at 7:25 PM, Jeremiah Peschka
wrote:
> I get errors trying to build eleveldb on OS X 10.7.1
>
> ==> eleveldb (compile)
> : command not foundform: line 2:
> : command not foundform: line 5:
whoops, uname -s
macpro1:~ siculars$ uname -s
Darwin
-Alexander Sicular
@siculars
http://siculars.posterous.com
On Sep 14, 2011, at 11:09 PM, Jared Morrow wrote:
> I don't have a Lion system to test it out, but I suspect it is due to the
> "uname -s" command in eleveldb/c_src/leveldb/build_d
uname -s returns Darwin on Lion
---
Jeremiah Peschka - Founder, Brent Ozar PLF, LLC
Microsoft SQL Server MVP
On Sep 14, 2011, at 10:09 PM, Jared Morrow wrote:
> I don't have a Lion system to test it out, but I suspect it is due to the
> "uname -s" command in eleveldb/c_src/leveldb/build_detect_p
uname -a on 10.7.1
macpro1:~ siculars$ uname -a
Darwin macpro1.local 11.1.0 Darwin Kernel Version 11.1.0: Tue Jul 26 16:09:02
PDT 2011; root:xnu-1699.22.81~1/RELEASE_I386 i386
-Alexander Sicular
@siculars
http://siculars.posterous.com
On Sep 14, 2011, at 11:09 PM, Jared Morrow wrote:
> I don'
I don't have a Lion system to test it out, but I suspect it is due to the
"uname -s" command in eleveldb/c_src/leveldb/build_detect_platform returning
something different for Lion than it does for Snow Leopard. If you take a
look in that file, it is pretty simple to see what is going on and it mi
Just some things I've been playing around with...
http://www.control-alt-del.org/2011/09/14/fun-with-bloom-filters-using-riak-mapreduce/
Mark Steele
Bering Media Inc.
___
riak-users mailing list
riak-users@lists.basho.com
http://lists.basho.com/mailman
I get errors trying to build eleveldb on OS X 10.7.1
==> eleveldb (compile)
: command not foundform: line 2:
: command not foundform: line 5:
: command not foundform: line 12:
: command not foundform: line 15:
'/build_detect_platform: line 17: syntax error near unexpected token `in
'/build_det
Evening, Morning, Afternoon to All -
For today's Recap: blog posts, new code, and more.
Enjoy.
Mark
Community Manager
Basho Technologies
wiki.basho.com
twitter.com/pharkmillups
Riak Recap for September 12 - 13
=
1) Greg Stein and Re
Does the leveldb backend support riak search, or will it support it soon?
http://wiki.basho.com/Riak-Search.html
--Craig
___
riak-users mailing list
riak-users@lists.basho.com
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
Enjoy:http://downloads.basho.com/riak/riak-1.0.0pre3/
The newly restyled release-notes are here:
https://github.com/basho/riak/blob/1.0/RELEASE-NOTES.org
As always, please let us know if you find any issues and remember these are
still pre-releases so we do not recommend you use them in a pro
FWIW, we also ran in to intermittent segfaullts in erlang_js_drv.so on
Ubuntu 10.04on EC2 and ended up rewriting our M/R jobs in erlang. We saw
log lines like this:
Aug 11 16:29:08 ip-10-66-190-207 kernel: [3281113.735644] beam.smp[22173]:
> segfault at 8 ip 7fe7842a6412 sp 7fe7877f5d40 e
Ahh, OK I get it now... if _any_ of these thresholds are met _and_ the files
are not active (i.e. they have grown larger than max_file_size) they'll be
merged. Thanks!
- Jeremy
On Wed, Sep 14, 2011 at 3:16 PM, Dan Reverri wrote:
> At any point in time Bitcask may have data spread across a numb
At any point in time Bitcask may have data spread across a number of data
files. Bitcask occasionally runs a merge process which reads the data from
those files and writes a merged set of data to a new file. Once completed
the old files can be removed and the new file is used for future read
operat
Excited for an overhaul. Something brought up in discussions with other PHP
devs, is the need to separate each class into it's own file.
Also, separating out the actual transport logic would be a good move.
And yes, 5.3+ minimum would be a good step forward. Anyone on 5.2 could just
stick with the
That assumes a consistent conflict resolution strategy for all objects in
the bucket (which makes sense under most circumstances I suppose).
I personally like the statebox approach, but you can't please everyone and
it doesn't apply to all possible situations. On the down side that kind of
impleme
Yep, that's partition 0. Partitions are spaced evenly around the hash
range, which is [0, 2^160) and cover hashes starting at their name. If
you have two partitions, they'll be {0, N/2}. three partitions: {0, N/3,
2N/3}.
--Kyle
On 09/14/2011 10:56 AM, Jeremy Raymond wrote:
In /var/lib/riak/b
Yes, vnodes/partitions are named by their integer offset into the SHA1 hash
space (2^160). The first one is '0'.
On Wed, Sep 14, 2011 at 1:56 PM, Jeremy Raymond wrote:
> In /var/lib/riak/bitcask there are a bunch of data directories whose name
> are a string of numbers. On once node I noticed I
In /var/lib/riak/bitcask there are a bunch of data directories whose name
are a string of numbers. On once node I noticed I have a directory whose
name is 0. This seemed out of place as the directory names are typically a
large string of numbers. Is this 0 directory normal?
- Jeremy
__
agree on optional read-before-write.
good note on optional returnbody.
I recall a talk by Yammer about their use of Riak in Scala and the gist was
that they achieve retry using closures.
I has envisioned resolvers as being bucket specific and being automatically
called at read time if present befo
Ok, thanks I'll give that a try.
What does small_file_threshold do then?
- Jeremy
On Wed, Sep 14, 2011 at 12:48 PM, Dan Reverri wrote:
> Hi Jeremy,
>
> The max_file_size parameter controls when Bitcask will close the currently
> active data file and start a new data file. The active data file
Hi Jeremy,
The max_file_size parameter controls when Bitcask will close the currently
active data file and start a new data file. The active data file will not be
considered when determining if a merge should occur. The default
max_file_size is 2GBs. This means that each partition in the system ca
If I'm reading the docs correctly, only files smaller
than small_file_threshold will be included in a merge. So
if small_file_threshold must be bigger than max_file_size for a merge to
happen?
- Jeremy
On Wed, Sep 14, 2011 at 10:23 AM, Jeremy Raymond wrote:
> Maybe I just need to tweak the Bitc
Looks great!
We are currently still using the 5.2 series, but a better client that only
supports 5.3 might motivate us to upgrade.
I would just add to make any read-before-write stuff optional (for
performance reasons), same with returnbody behaviors.
Sibling resolution is very application speci
On 14 Sep 2011, at 12:37, Mathias Meyer wrote:
> The short answer: yes, we can and we should. I had that on my radar for a
> while too, because it felt un-Pythonic.
>
> As for deprecation, there's no specific rule for the Python client yet. I'm
> happy to accept a patch for it for e.g. a vers
Maybe I just need to tweak the Bitcask parameters to merge more often?
I have approx 17000 keys which get overwritten once an hour. After each
updated the /var/lib/riak/bitcask folder grows by 20 MB (so about 1200 bytes
per key). With the default frag_merge_trigger at 60 I should get a merge
every
Riak 1.0 is coming soon.
Now would be a good time to overhaul the PHP client.
Goals
- Add support for Protocol Buffers
- Add support for new Riak 1.0 features (2i, pr, pw, client_id, etc)
- Add performance benchmarks
- Improve network performance
- Improve tests
- Add support fo
Riak 1.0 is coming soon.
Perhaps it makes sense for client point releases to coincide with server
point releases?
- Kev
c: +001 (650) 521-7791
On Wed, Sep 14, 2011 at 4:37 AM, Mathias Meyer wrote:
> The short answer: yes, we can and we should. I had that on my radar for a
> while too, because
Is it possible at all you indexed the documents using the Solr interface, and
you're now trying to use them in a MapReduce query? If so, that won't work,
because the get() call expects the objects with their respective bucket and key
to exist in Riak KV, which means you'd have to index via a pre
I would think that the InnoDB backend would be a better backend for the use
case you're describing.
---
Jeremiah Peschka - Founder, Brent Ozar PLF, LLC
Microsoft SQL Server MVP
On Sep 14, 2011, at 8:09 AM, Jeremy Raymond wrote:
> Hi,
>
> I store data in Riak whose keys constantly get overwritte
Hi,
I store data in Riak whose keys constantly get overwritten with new data.
I'm currently using Bitcask as the back-end and recently noticed the Bitcask
data folder grow to 24GB. After restarting the nodes, which I think
triggered Bitcask merge, the data went down to 96MB. Today the data dirs ar
The short answer: yes, we can and we should. I had that on my radar for a while
too, because it felt un-Pythonic.
As for deprecation, there's no specific rule for the Python client yet. I'm
happy to accept a patch for it for e.g. a version of the client 1.4.0 with an
announcement that support
Hi all,
There are some non-Pythonic patterns in riak-python-client that should
be pretty easy to switch. Things like client.get_r() and
client.set_r() are kinda silly. Python long ago moved past the
getter/setter paradigm, with the notion of directly exposing instance
attributes. As Guido has said
--- and this time, to the list too ---
Hi Harshal,
On 14 Sep 2011, at 00:41, Harshal Dhir wrote:
> Folks,
>
>
> I am using Riak Search v0.14.2 5-Node-cluster. We are trying to write a
> distributed application using RabbitMQ. But, for some reason riak is
> bombing out on me and giving me "Brok
38 matches
Mail list logo
