89e7eae8c6
fix the sleep since the process takes about 5 seconds to startup
2014-03-11 18:52:37 -04:00
3c664f4f08
fix #302 . Should be able to set negative times
2014-03-11 18:52:30 -04:00
a33bb3733d
fixup! we shouldn't be returning error if the function takes a response channel
2014-03-11 18:52:30 -04:00
64571fc972
Clean up test spacing.
2014-03-11 18:10:04 -04:00
c54588dd43
A massive group by test with real data.
2014-03-11 18:10:04 -04:00
fe10d5b1b0
Output the shard's start and end timestamps on String().
2014-03-11 18:10:04 -04:00
ef92908983
Add StartMicro and EndMicro methods to the shard.
2014-03-11 18:10:04 -04:00
e698a56552
sleep for a few seconds to give a chance for raft to replay the log
2014-03-11 16:32:52 -04:00
e1b996ea3a
more logging
2014-03-11 16:32:52 -04:00
5db312db88
fix a compilation error
2014-03-11 16:32:52 -04:00
5c56bfab91
change the name of the series in the test
2014-03-11 16:32:52 -04:00
2fe6fdc325
fix a bug, use value comparison instead of pointer comparison
2014-03-11 16:32:52 -04:00
ad99747e39
we shouldn't be returning error if the function takes a response channel
...
This is causing a bunch of deadlocks since those functions gets called
in a goroutine and their return values are never checked. Instead we
should return an endStreamResponse with the error message.
2014-03-11 16:32:52 -04:00
c33f94c14b
remove some logging statements
2014-03-11 16:32:52 -04:00
359584df3d
Drop the max delete size from 1M to 64K.
2014-03-11 14:31:50 -04:00
6fab943cbe
Revert "panic if we ever receive commits out of order"
...
This reverts commit 20a921eb32
2014-03-11 14:26:09 -04:00
bdb6cf43ff
fix a compilation error
2014-03-11 12:39:29 -04:00
4ede6f847e
fix #233 . Put back the code that limited the delete batch size
2014-03-11 12:38:55 -04:00
e0e7144c93
cleanup default start and end time code
2014-03-10 20:32:31 -04:00
710980c30e
fix a typo
2014-03-10 20:32:31 -04:00
b3c35ff2c6
Fix #315 . Make delete queries not use WAL. They should instead use Raft to replicate.
2014-03-10 18:10:02 -04:00
6dc127bde5
add a test to make sure invalid time group by interval don't crash the server
2014-03-10 17:33:35 -04:00
39cec27b48
update the changelog
2014-03-10 17:30:21 -04:00
e5b1486a3d
Merge pull request #324 from influxdb/fix-316-access-denied-goroutine-starvation
...
Make shard hold access denied responses until all response channels are ...
2014-03-10 17:18:33 -04:00
2a780d3edb
Fix #316 . Make shard hold access denied responses until all response channels are emptied out.
2014-03-10 16:52:14 -04:00
7ba33479ae
remove some unused constants
2014-03-10 15:45:55 -04:00
d70972cdd0
update the changelog
2014-03-10 14:01:27 -04:00
7b74e5de29
Propagate the error back and assert on the body of the response
2014-03-10 14:00:00 -04:00
be7137a246
Merge branch 'plumb-nqe-fail' of github.com:zorkian/influxdb
...
Conflicts:
src/cluster/shard.go
2014-03-10 12:19:11 -04:00
74218be735
uncomment some of the engine tests that tested invalid queries
2014-03-10 12:16:55 -04:00
db5ba7155d
update the changelog
2014-03-10 12:02:44 -04:00
dd55e2cafc
Merge pull request #301 from influxdb/ref-count-on-open-shards
...
Add ref counting to which leveldb shards are open so they can be closed ...
2014-03-10 12:00:36 -04:00
afd2d8c9be
Updated ref counting to remove accessed shards from the list of ones to close. Updated the close to close as many as it takes to get below the leve
2014-03-10 11:47:57 -04:00
e39508f834
Add test for max shards and fix bug in closing out old one
2014-03-10 11:47:57 -04:00
2bce2837fb
Fix bug where I was deleting the shard reference before pulling out the shard to close it
2014-03-10 11:47:57 -04:00
0d6e2d66ff
Add ref counting to which leveldb shards are open so they can be closed after references released
2014-03-10 11:47:57 -04:00
386d3d6aaf
Allow NewQueryEngine to return an error
...
Some errors only bubble up from pretty deep. A "SELECT COUNT(*)" was
ending up as a panic, when instead it should have been returned as a
query error to the user.
This adds plumbing for returning an error from NewQueryEngine.
2014-03-09 16:41:02 -07:00
0e98711613
accept any request number initially
2014-03-07 18:32:12 -05:00
20a921eb32
panic if we ever receive commits out of order
2014-03-07 18:29:17 -05:00
378d185632
fix a bug introduced by my refactoring
2014-03-07 18:29:13 -05:00
a2f0bf82ac
flush on every write in the tests
2014-03-07 18:19:44 -05:00
41b776cb88
don't do this twice, 5 seconds is good enough
2014-03-07 18:19:44 -05:00
73648b5a6f
move the sleep outside the loop in the test
2014-03-07 18:19:44 -05:00
d029273f89
more logging
2014-03-07 18:19:39 -05:00
10b2820d9e
fix another bug in the wal, not all requests have series
2014-03-07 15:21:26 -05:00
564c0f1d3d
fix the broken tests, now recovery happens after server id is set
2014-03-07 15:05:53 -05:00
0c7542d6e4
fix #312 . WAL should wait for server id to be set before recovering
2014-03-07 14:28:31 -05:00
6a3b16e854
more changes to the changelog
2014-03-07 12:59:46 -05:00
ab430f8d6a
update changelog
2014-03-07 12:59:21 -05:00
8f95ce4b76
update changelog
2014-03-07 12:58:36 -05:00
John Shahid
Todd Persen
Paul Dix
Mark Smith