Commit Graph

1976 Commits

Author SHA1 Message Date
Jon Anderson
0671b59b4c add a intercept/server accounting of get_fsm processes. 2014-07-22 16:48:25 -04:00
Christopher Meiklejohn
e38c5423aa Only run CRDT bench by default. 2014-07-17 14:43:45 -04:00
Christopher Meiklejohn
216a8a273e Add benchmark helper script. 2014-07-17 11:30:00 -04:00
Christopher Meiklejohn
238826d9da Change dimensions of graphic. 2014-07-17 11:30:00 -04:00
Christopher Meiklejohn
9040155139 Add CRDT team-style benchmark. 2014-07-17 11:30:00 -04:00
Christopher Meiklejohn
b6b278fe83 Add additional operations to digest. 2014-07-17 11:30:00 -04:00
Kelly McLaughlin
98681cd658 Increase the test code coverage and make the coverage checking more
robust to failures.
2014-07-15 16:48:43 -06:00
Kelly McLaughlin
6cec9f82e0 Add pbc/2 function to the rt module to allow custom options to be specified 2014-07-15 16:48:36 -06:00
Kelly McLaughlin
ceb24fc3e2 Merge pull request #661 from basho/bugfix/replication-ssl-site-ip-verification
Fix failure of replication_ssl test introduced by 297090d
2014-07-15 15:08:50 -06:00
Kelly McLaughlin
1ba3fed5e3 Merge pull request #662 from basho/bugfix/avoid-replication-cert-expiry
Change replication SSL ACL tests to avoid certificate expiration
2014-07-15 15:08:17 -06:00
Christopher Meiklejohn
62feeeaa14 Remove perf_restart for now. 2014-07-15 14:15:23 -04:00
Christopher Meiklejohn
433b320be9 Cleanup, add debugging. 2014-07-15 14:15:23 -04:00
Christopher Meiklejohn
a4f05d1aa4 Default strong_consistency on. 2014-07-15 14:15:23 -04:00
Christopher Meiklejohn
0f919c3e54 Update example to use 2.0RC1. 2014-07-15 14:15:23 -04:00
Christopher Meiklejohn
70e029083a Remove extraneous logging. 2014-07-15 14:15:23 -04:00
Christopher Meiklejohn
bd6c55dc2c Update consistent_get_put to the new API. 2014-07-15 14:15:22 -04:00
Kelly McLaughlin
0ab2393559 Change replication SSL ACL tests to avoid certificate expiration
Change the ACL test case in the replication_ssl and replication2_ssl
tests to use certificates generated within the tests instead of
relying on certificates created outside the test that are prone to
expire and cause spurious test failure.

Also change the replication_ssl and replication2_ssl tests to avoid a
cycle of standing up the test clusters and then immediately restarting
them before any tests cases execute. This should make the test
execution slightly faster for both test modules.

This commit also changes the tests to be a bit more robust in checking
for cluster state when restarting nodes and removes an unnecessary
five second sleep call in the replication_ssl test.
2014-07-15 12:06:06 -06:00
Kelly McLaughlin
9c5daf0f31 Fix failure of replication_ssl test introduced by 297090d
Change replication_ssl to use the wait_for_site_ips function from the
replication module introduced in
297090ded6 instead of the defunct
verify_site_ips function.
2014-07-14 12:37:32 -06:00
Christopher Meiklejohn
dd1581ceaf Merge pull request #660 from basho/bug/rdb/pb-api-change
Update expected return to match change in riak_pb API
2014-07-14 12:33:51 -04:00
Russell Brown
a212b99a75 Update expected return to match change in riak_pb API
See 2b68a97710
for details.
2014-07-14 17:14:51 +01:00
Christopher Meiklejohn
70f8cc85e3 Resolve function_clause with atom. 2014-07-14 11:54:38 -04:00
Christopher Meiklejohn
0359c309f6 Merge pull request #586 from basho/perf-harness-merge
At last, merge the perf harness.
2014-07-12 14:20:33 -04:00
Christopher Meiklejohn
c4ee9aa96f Merge pull request #659 from basho/csm-perf-harness
Csm perf harness
2014-07-12 14:20:06 -04:00
Christopher Meiklejohn
cf08f652c3 Checkpoint. 2014-07-12 14:18:06 -04:00
Christopher Meiklejohn
c8399daaed Checkpoint. 2014-07-12 13:45:53 -04:00
Christopher Meiklejohn
04e4fa6f76 Checkpoint. 2014-07-12 13:02:15 -04:00
Christopher Meiklejohn
2315efa046 Checkpoint. 2014-07-12 12:49:18 -04:00
Joseph Blomstedt
695853cc94 Merge pull request #657 from basho/bugfix/ensemble-interleave-error-failed
Fix ensemble_interleave error condition
2014-07-11 21:42:59 -07:00
Joseph Blomstedt
6f34608e00 Merge pull request #658 from basho/bugfix/ensemble_sync_failed
Fix ensemble_sync by allowing {error, <<"failed">>
2014-07-11 21:42:46 -07:00
Christopher Meiklejohn
c1ff7e4d84 Futher checkpoint. 2014-07-11 22:56:15 -04:00
Christopher Meiklejohn
b906a31be9 Ignore rebar. 2014-07-11 20:47:56 -04:00
Christopher Meiklejohn
a013227fa7 Checkpoint. 2014-07-11 20:39:20 -04:00
Andrew J. Stone
f0643db473 Fix ensemble_sync by allowing {error, <<"failed">>
Allow {error, <<"failed">>} as an error response in ensemble_sync. Fixes
the test with basho/riak_ensemble#37 and basho/riak_kv#1002
2014-07-11 18:11:30 -04:00
Andrew J. Stone
38bd8399d1 Fix ensemble_interleave error condition
Include {error, <<"failed">>} as allowed failure so that test passes
with changes for basho/riak_ensemble#37 and basho/riak_kv#1002
2014-07-11 17:41:48 -04:00
Kelly McLaughlin
b59fb48611 Expand the overload test to include strong consistency
Change the overload test to exercise the strongly consistent code
paths in addition to the eventually consistent paths during overload
conditions.
2014-07-11 13:23:48 -06:00
Christopher Meiklejohn
9db9f51744 Merge branch 'master' into perf-harness-merge
Conflicts:
	src/riak_test_escript.erl
	src/riak_test_runner.erl
	src/rtssh.erl
2014-07-11 12:19:42 -04:00
Kelly McLaughlin
e3df2fbf2f Merge pull request #655 from basho/bugfix/replication-verify-site-ip-race
Avoid a race condition in the replication test module
2014-07-02 10:02:08 -06:00
Kelly McLaughlin
297090ded6 Avoid a race condition in the replication test module
Avoid a race condition in the replication test module when checking
for site IP addresses in the replication status output.  The test
waits for a connection on the leader, but it only queries the
replication status to check for the expected site IP addresses a
single time. Change the test to wait and re-check the status output to
give greater assurance that if the expected site IP addresses are not
present it is due to legitimate failure and not a race condition in
checking the replication status. This change affects the replication
and replication_upgrade tests as well as any other tests that call the
replication:replication function.
2014-07-01 16:36:45 -06:00
Jon Anderson
054c015d10 Merge pull request #651 from basho/feature/jra/verify_listkeys_eqcfsm
Expand verify_listkeys_eqcfsm to track varying buckets and n_vals.
2014-07-01 13:19:41 -04:00
Sean Cribbs
17b94da468 Fix nitpick about new/old API from #653. 2014-07-01 08:53:00 -05:00
Christopher Meiklejohn
f085f70169 Merge pull request #653 from basho/features/csm/crdt-capability
Prevent autoreconnect problem.
2014-06-30 18:29:05 -04:00
Christopher Meiklejohn
da34719fe3 Prevent autoreconnect problem.
Prevent a situtation where the auto-reconnect hasn't triggered yet
causing the result to be an error, instead of ok, on the next operation
after reconnecting.  Force a disconnect and reconnect to make sure the
test is deterministic.
2014-06-30 17:22:20 -04:00
Christopher Meiklejohn
92314b308b Merge pull request #652 from basho/features/csm/riak-control
Resolve riak-control version conditionals.
2014-06-30 13:06:33 -04:00
Christopher Meiklejohn
8e14cc1fc1 Resolve riak-control version conditionals.
Simply the logic and make it a bit more resilient to failures when doing
a rolling upgrade.
2014-06-30 13:04:33 -04:00
Jon Anderson
7c2d7cc827 Expand verify_listkeys_eqcfsm to track varying buckets and n_vals. 2014-06-26 16:13:58 -04:00
Eric Redmond
0eb2d1c443 Merge pull request #650 from basho/er/yz-ensemble-test
Test that ensemble delete functions in yokozuna
2014-06-23 17:38:09 -07:00
Eric Redmond
b2b4d33dc1 Expand pbc_write to accept content type 2014-06-23 13:01:25 -07:00
Engel A. Sanchez
3662965705 Merge pull request #649 from basho/feature/ensembles-wait-for-riak-kv
Feature/ensembles wait for riak kv
2014-06-23 14:24:12 -04:00
Eric Redmond
266f9858eb Test the ensemble delete function 2014-06-20 14:44:30 -07:00
Engel A. Sanchez
d32d007f4d Fix service/peer check race
Changing to fetching the list of peers first, then check if the riak_kv
service is up. If the service is up, then check the peers. Otherwise it
is possible to see the service down, then peers up because it went up in
the interim.

Also, making KV vnode delay configurable.
2014-06-20 14:26:24 -04:00