Andrew Thompson
224f8996c3
Actually make the intercepts work...
2013-02-25 13:52:47 -05:00
Andrew Thompson
afe2714f33
Add missing function
2013-02-25 09:57:13 -05:00
Andrew Thompson
431620975c
Change PR/PW test to use intercepts instead of kill -STOP
...
Thanks to @rzezeski for the help with intercepts
2013-02-23 00:25:08 -05:00
Andrew Thompson
58f017c7c9
Half finished PR/PW test
2013-02-22 22:10:25 -05:00
Andrew Thompson
aa5f97cc92
Merge pull request #216 from basho/adt-wait-for-repl
...
Wait for repl to actually start before shutting the node down again
2013-02-21 09:55:23 -08:00
Andrew Thompson
bcc5a117d8
Wait for repl to actually start before shutting the node down again
2013-02-21 12:53:25 -05:00
Andrew Thompson
14136b21a3
Use an alphanumeric md5 hash for bucket to avoid angering HTTP
2013-02-21 12:35:57 -05:00
Andrew Thompson
2df124494d
Merge pull request #212 from basho/adt-ssl-fixes
...
Adt ssl fixes
2013-02-20 09:20:09 -08:00
Andrew Thompson
0c89cb92ba
Handle too_busy response
2013-02-20 11:11:18 -05:00
Andrew Thompson
a959e70945
Fix some path and ring convergance issues in repl tests
2013-02-19 22:50:12 -05:00
Bryan Fink
3c22aae327
Merge pull request #143 from branch 'bwf-pipe'
2013-02-19 15:05:53 -05:00
Bryan Fink
397c800171
Merge pull request #209 from branch 'bwf-move-mapred-test'
2013-02-19 13:21:31 -05:00
Bryan Fink
25fe714c89
build 3-node clusters for mapred tests (instead of 1 node)
...
may as well take advantage of riak_test's ease of building clusters
2013-02-19 12:25:25 -05:00
Bryan Fink
efd98c06dd
remove unnecessary unlink from riakc_pb_socket
...
the process stops 'normal' so there's no problem with it being linked
2013-02-19 11:39:49 -05:00
Bryan Fink
739641623c
remove redundant assert
...
this is already handled by rt:build_cluster; it made it here through
copy/paste from old tests
2013-02-19 11:35:20 -05:00
Bryan Fink
0d5819df48
don't import rt_pipe.hrl until it's merged from the other branch
2013-02-19 10:54:36 -05:00
Bryan Fink
cd4cf58a08
migrate riak_kv's mapred_test suite
...
The mapred_test eunit suite had a fragile setup stage that frequently
caused the automated builders to fail. It should be more stable here.
The suite has also been broken into several independent tests, to make
it possible to run each major piece individually.
2013-02-19 08:58:51 -05:00
Dave Parfitt
90ba560cc6
Merge pull request #204 from basho/dip_shutdown_dw
...
change w -> dw for shutdown tests
2013-02-16 12:07:42 -08:00
Andrew Thompson
d7dca09df9
Make sure we exercise the code path when # of keys > diff_batch_size
...
We do this by lowering diff_batch_size to 10, and then writing 10,000
keys to a 64 vnode ring, thus ensuring that each vnode will have more
than 100 keys.
2013-02-15 14:03:31 -05:00
Dave Parfitt
1a67cb6a89
change w -> dw for shutdown tests
...
this is @vagabond's patch
2013-02-14 15:53:19 -05:00
Andrew Thompson
5af919b812
Make sure we wait for elections to complete when starting repl tests
2013-02-14 13:40:47 -05:00
Joe DeVivo
7e8cd7b668
Added {error, notfound} to kv check, more robust connection recycling
2013-02-12 17:56:29 -07:00
Joe DeVivo
d9a2acdcf2
added no-op for 2i on legacy
2013-02-12 10:38:56 -07:00
Joe DeVivo
d521af1dfb
Added version awareness to test functions
2013-02-12 09:33:39 -07:00
Joe DeVivo
f46f82f6f0
Added process message on search failure and easy reading asserts
2013-02-08 09:24:35 -07:00
Sean Cribbs
59845fef1d
Make the number of load workers configurable.
2013-02-07 22:20:43 -06:00
Joe DeVivo
02591fdb9d
Added supervisor pid to node monitor
2013-02-07 12:23:55 -07:00
Joe DeVivo
309a530c0c
Add Node to return tuple for workers
2013-02-07 11:44:53 -07:00
Joe DeVivo
4663b4f9dc
added _sup to loaded_upgrade_worker
2013-02-07 11:25:58 -07:00
Joe DeVivo
f331bd196a
loaded_upgrade_worker gen_server to supervisor
2013-02-07 11:10:46 -07:00
Joe DeVivo
f3ef738649
some iolist stuffs
2013-02-07 09:48:33 -07:00
Joe DeVivo
e27b0fc60f
Some copyright headers
2013-02-07 09:45:55 -07:00
Joe DeVivo
9681438334
Everyting from PR #196 except the big ones
2013-02-07 09:38:06 -07:00
Joe DeVivo
f41a5f591d
Added node monitor
2013-02-06 12:44:15 -07:00
Joe DeVivo
6d3d1e675f
Added catch for search badfun
2013-02-06 10:20:38 -07:00
Joe DeVivo
0b8d0f7889
Moved new_loaded_upgrade to loaded_upgrade
2013-02-05 13:25:38 -07:00
Joe DeVivo
60f8b60367
More message passing, less logging
2013-02-05 13:01:01 -07:00
Joe DeVivo
14a2d376c5
A passing full losed upgrade run
2013-02-05 13:01:01 -07:00
Joe DeVivo
d9563559e4
10 concurrent workers per node, breaks at first upgrade
2013-02-05 13:01:00 -07:00
Joe DeVivo
a453284201
Load applied
2013-02-05 13:01:00 -07:00
Joe DeVivo
b1b7730292
POC new_loaded_upgrade
2013-02-05 13:01:00 -07:00
Bryan Fink
b6a6066d65
use rt:systest_write instead of listcomp of riakc_pb_socket:put
...
Yay for using builtin utilities instead of hand-rolled hacks.
2013-02-02 08:54:57 -05:00
Bryan Fink
c8efe95dac
reproduction test for basho/riak_kv#475 (MR reduce with node down)
...
This test is based off one submitted by Alexander Gunin to the
riak-users mailing list as an issue reproducer.
[http://lists.basho.com/pipermail/riak-users_lists.basho.com/2013-January/010896.html ]
2013-02-01 15:47:42 -05:00
Andrew Thompson
77547c9d88
Don't test for leader election on 1.2.1 or earlier, fix version checking
...
Also, make some RPC calls more resiliant
2013-01-31 14:49:14 -05:00
Andrew Thompson
96916e97d4
Wait for connection to re-establish after leader election
2013-01-30 14:54:17 -05:00
Chris Tilt
30dd520065
Fix broken calls to make_cluster
2013-01-30 11:20:43 -08:00
Chris Tilt
24d49cd7e8
Add some debugging for replication test
2013-01-30 11:20:43 -08:00
Chris Tilt
109d464c76
Fix my refactor error and add some debug
2013-01-30 11:20:43 -08:00
Andrew Thompson
c705247f29
Wait for del_site to disconnect the clusters before continuing
2013-01-30 13:44:33 -05:00
Engel A. Sanchez
eb6c3eba15
Fix insufficient vnodes in verify_listkeys
...
Fixes issue #149 , related to basho/riak_core#228
If queried right after it has joined the cluster, a node may not have
the other nodes registered by the node watcher yet. Switching to waiting
until all nodes in the cluster have been registered by the node watcher
before performing query.
Also, querying for all nodes after stopping/starting just in case,
although I haven't seen problems there, but sound reasonable.
Removing assert from the call to wait_until_no_pending_changes since it
is already asserting internally.
2013-01-30 10:48:13 -05:00
