Alternative to Post-Commit in EDS

2012-04-02 Thread Anthony Molinaro
Hi, I'm evaluating Riak EDS for a system where I'd like to have data replicated between several colo's. The replication would be bidirectional, but the flow would only reverse if a colo goes down (ie, normally all writes to a bucket go to one colo and are replicated to others, but if that colo

Nodes not leaving in Riak 0.14.2

2012-04-02 Thread Anthony Molinaro
Hi, So I found one of my riak clusters misbehaving and found out it was running on some inferior hardware. After getting some new machines from my ops group. I went ahead and added several nodes. Everything went relatively smoothly until I tried to have the old nodes leave the ring. I need t

Re: Lager Question: how to get the pid:module:function format

2012-04-02 Thread Bryan Hughes
Thanks Andrew! That did the trick.   You know a sign of good technology and sweet implementation is when it slips quietly into the background and you begin to take it for granted.   Lager continues to rock and roll. Cheers, B On 3/31/12 10:45 AM

preflist_exhausted with map-reduce and riak-js

2012-04-02 Thread Matthew Tovbin
Unfortunately, yes, Jon. Here are my log files after setting 128/128 for 'map_js_vm_count' and 'reduce_js_vm_count' accordingly and restarting all the nodes (see attached app.config). If you need any other information provided or tests performed please let me know. *-Console.log:* 2012-04-02 15:

Re: slow mapred_search key lookups for single terms

2012-04-02 Thread Ryan Zezeski
Michael, These are some interesting observations. Thank you for digging into this more. At first glance it appears that it has something to do with the PB client/server. If you are feeling adventurous might I suggest using fprof to analyze where the time is being spent. I wrote up a gist on ho

Re: help with riak search

2012-04-02 Thread Wes James
Ah, OK - thanks! Yes - that worked -wes On Mon, Apr 2, 2012 at 2:35 PM, Daniel Reverri wrote: > From the Erlang console you have escape twice (not sure why): > search:search_doc(<<"searchable-dbooks">>, <<"notes:\\&">>). > > Thanks, > Dan > > -- > Daniel Reverri > Developer Advocate > Bash

Re: help with riak search

2012-04-02 Thread Daniel Reverri
>From the Erlang console you have escape twice (not sure why): search:search_doc(<<"searchable-dbooks">>, <<"notes:\\&">>). Thanks, Dan -- Daniel Reverri Developer Advocate Basho Technologies, Inc. d...@basho.com On Monday, April 2, 2012 at 1:29 PM, Wes James wrote: > On Mon, Apr 2, 2012 at

Re: help with riak search

2012-04-02 Thread Wes James
On Mon, Apr 2, 2012 at 1:45 PM, Daniel Reverri wrote: > > I was able to search for "&" by escaping the character: > search-cmd search wildcards "value:\&" > with the erlang command, I get an error: (riak@127.0.0.1)25> search:search_doc(<<"searchable-dbooks">>,<<"notes:\&">>). ** exception er

Riak Ruby Client v1.0.2

2012-04-02 Thread Sean Cribbs
Hey riak-users, We've just released the official Ruby client for Riak version 1.0.2, which is now available on Github and Rubygems. (Note: v1.0.1 was yanked but its fixes are included in this release.) This release fixes a few bugs, including HTTPS/SSL configuration and compatibility with Excon an

Re: help with riak search

2012-04-02 Thread Daniel Reverri
Hi Wes On Monday, April 2, 2012 at 10:05 AM, Wes James wrote: > Is there documentation to indicate what riak search supports in search > operations. The best available documentation is here: http://wiki.basho.com/Riak-Search.html > > I have been testing some searches and even looked at: >

Re: preflist_exhausted with map-reduce and riak-js

2012-04-02 Thread Jon Meredith
Very strange. Are you seeing the 16:40:50.037 [notice] JS call failed: All VMs are busy. message in your console.log files on any of the nodes in your cluster? Jon On Mon, Apr 2, 2012 at 11:48 AM, Matthew Tovbin wrote: > Jon, > > Thanks for the reply. I tried increasing the 'map_js_vm_count'

Re: slow mapred_search key lookups for single terms

2012-04-02 Thread Michael Radford
I modified the pb client code to print the elapsed time when it receives the first map-reduce result (in the function wait_for_mapred), and tried to run a more apples-to-apples comparison of the same map-reduce job via the local client and the pb client connecting to 127.0.0.1. In a nutshell, the

Riak Recap for April 1

2012-04-02 Thread Mark Phillips
Afternoon, Evening, Morning to All - Today's Recap (live from DevOpsDays in Austin): meetups in Dallas and Atlanta, new code, and more. Happy Monday. Mark twitter.com/pharkmillups --- Riak Recap for April 1 1) Tryn Mirell is giving a Riak talk

Re: preflist_exhausted with map-reduce and riak-js

2012-04-02 Thread Matthew Tovbin
Jon, Thanks for the reply. I tried increasing the 'map_js_vm_count' and 'reduce_js_vm_count' on the cluster to the values you suggested 24/18 and even larger ones 32/24 and 128/128, but with no success - I still get the same error in the console log of the node. What are we missing?! -Matthew

help with riak search

2012-04-02 Thread Wes James
Is there documentation to indicate what riak search supports in search operations. I have been testing some searches and even looked at: http://lucene.apache.org/core/old_versioned_docs/versions/3_5_0/queryparsersyntax.html for some ideas. These work: search:search_doc(<<"searchable-dbooks">>,

Re: slow mapred_search key lookups for single terms

2012-04-02 Thread Michael Radford
Hi Ryan, This is getting interesting: the same queries when executed using local clients from 'riak attach' are taking only 100-250 ms. However, I just tried the same test I was running remotely on Saturday on one of the machines in the Riak cluster, using the protobufs client to connect to 127.0

Re: slow mapred_search key lookups for single terms

2012-04-02 Thread Ryan Zezeski
Hi Michael, you'll find my responses inline... On Sat, Mar 31, 2012 at 5:04 PM, Michael Radford wrote: > I'm seeing very slow performance from Riak search even when querying > single terms, and I'd appreciate any advice on how to get insight into > where the time is going. > > Right now, I'm usi