Alan Bernstein
8c9db373d0
Fix some metrics names
2020-04-10 20:59:18 -05:00
Alan Bernstein
5137f56f9c
Use const from package
2020-04-10 20:59:18 -05:00
Alan Bernstein
34c6d42063
Fix broken metrics label and log when others are encountered
2020-04-10 20:59:17 -05:00
Alan Bernstein
3c275681d2
Profile -> Column
2020-04-10 20:59:17 -05:00
Alan Bernstein
b1adcd91fc
Use consistent metric name convention
2020-04-10 20:59:17 -05:00
Alan Bernstein
70111b5604
Define metrics names as constants
2020-04-10 20:59:17 -05:00
Alan Bernstein
84e6a25bad
Update help message
2020-04-10 20:59:17 -05:00
Alan Bernstein
857ddf73c2
Reduce snapshot verbosity
2020-04-10 20:59:17 -05:00
Alan Bernstein
eceef6b42b
Use 'query_' prefix to identify query metrics
2020-04-10 20:59:16 -05:00
tgruben
550fcec9ee
Merge pull request #279 from tgruben/fix-closers
...
Closed all post request bodies and optimized available shard with new view capabilities
2020-04-10 20:29:04 -05:00
Todd Gruben
ccc7ca3aa5
close not needed
2020-04-10 20:12:32 -05:00
Todd Gruben
5590f1d954
lint fix
2020-04-10 19:07:01 -05:00
Todd Gruben
ef5a8cefef
Added request Close and optimized availble shard with new view capabilities
2020-04-10 17:59:44 -05:00
seebs
4325d62fe7
Merge pull request #277 from seebs/between
...
Return empty rows for impossible ranges
2020-04-10 16:21:09 -05:00
Seebs
017e65cd99
Return empty rows for impossible ranges
...
If the high end of a range is below the low end of the range, there's
no values in it, so we can short-circuit that. If we don't, if the
low end is zero or higher, and the high end is below zero, we can
get very surprising behaviors, such as accepting values up to the
inverse of the high end. Add a test case for this and treat it the
same as a low range end above the field's maximum or a high end
below the field's minimum, returning an empty row immediately.
2020-04-10 13:24:08 -05:00
seebs
9b002bcc24
Merge pull request #271 from seebs/decimalfault
...
Handle nonexistent shards in min/max decimal queries.
2020-04-09 22:01:07 -05:00
Seebs
8e662d33a5
Handle nonexistent shards in min/max decimal queries.
...
If a shard has never had any decimal values in it at all for a
field, the ValCount object returned has no DecimalVal, which could
cause a segfault if we don't check for it. Add a test case which
sporadically triggers that behavior (it's timing/luck related,
unfortunately), and then also fix it.
2020-04-09 21:30:34 -05:00
Matthew Jaffee
179cab91e7
Merge pull request #275 from travisturner/serialize-null
...
serialize null operation (!= null)
2020-04-09 21:25:05 -05:00
Travis
88b2d79812
serialize null operation (!= null)
2020-04-09 17:56:27 -05:00
alanbernstein
0e6ed80bd4
Merge pull request #269 from alanbernstein/long-query-full
...
WIP Show full query string when logging slow queries
2020-04-09 16:12:11 -05:00
Alan Bernstein
92416018d5
Use more general log message
2020-04-09 15:53:40 -05:00
Alan Bernstein
5aebb242f8
Fix off-by-one error
2020-04-09 15:53:40 -05:00
Alan Bernstein
e3ace544ca
Show full query string when logging slow queries
2020-04-09 15:53:39 -05:00
tgruben
57b2f32fa1
Merge pull request #260 from tgruben/sync-start
...
delay start for non-coordinator
2020-04-09 14:42:41 -05:00
Todd Gruben
e7bbc3a0a8
Added delay to allow cooridinator a head start in launch on multi node clusters.
...
considered using Cluster.Disabled to identify if pilosa was stand alone but settled
on using an empty gossip seeds list
2020-04-09 14:20:01 -05:00
alanbernstein
bdfad719df
Merge pull request #263 from alanbernstein/grpc-multi
...
Support multiple dialTargets and cycle through on connection reset
2020-04-08 13:39:58 -05:00
alanbernstein
1846d21765
Merge pull request #1 from travisturner/dialtargets
...
cycle through dial targets regardless of error
2020-04-08 12:55:17 -05:00
Travis
8f0ae1b6f5
cycle through dial targets regardless of error
2020-04-08 12:29:15 -05:00
Alan Bernstein
adb21589f0
Support multiple dialTargets and cycle through on connection reset
2020-04-08 10:09:49 -05:00
seebs
1b5d86c8f0
Merge pull request #240 from seebs/q2perf
...
improve performance of difference/not
2020-04-07 20:24:51 -05:00
Seebs
be379e7806
cache AvailableShards
...
The computation of available shards is cheap, because realistically, virtually
no one has enough shards that the resulting bitmap is more than one container.
We don't try to fix this at the field/index levels because it's significantly
harder to do there, but I think the creation of these bitmaps is probably
the most expensive part, and switching the unions to union-in-place probably
reduces cost significantly.
Note that the bitmaps being unioned almost certainly have exactly one small
container in them.
2020-04-07 19:33:21 -05:00
Seebs
6c797e0c4e
TODO => TODONE: use masks for runToBitmap
...
Had the code lying around from mad science elsewhere, backported.
2020-04-07 19:33:21 -05:00
Seebs
d4e496887b
make differenceRunBitmap smarter
...
We avoid using bitmapContains so often because that turns out to be expensive.
Also, if we produce more than runMaxSize runs, we're going to convert to
a bitmap container (or possibly an array container if there were over
2048 items, but they're all singletons), and we can streamline that by just
converting the source to bitmap and returning differenceBitmapBitmap, which
is faster in this case.
This appears to overall take about half as long in the workload I was
looking at.
2020-04-07 19:33:20 -05:00
Kuba Podgórski
7868188670
Merge pull request #200 from kuba--/groupby-int/124
...
Group by on ints
2020-04-08 01:47:17 +02:00
Kuba Podgórski
1217deee1c
Rebase
2020-04-08 01:06:18 +02:00
tgruben
64caffebe8
Merge pull request #196 from tgruben/async-available-shards
...
limit frequency of writes for available shards broadcast
2020-04-06 09:08:04 -05:00
Todd Gruben
6712f8cf06
removed debug log
2020-04-06 08:43:30 -05:00
Todd Gruben
5470753cb5
fix conflict
2020-04-05 18:37:11 -05:00
Travis
0e2bb550db
add deleted (rebalanced) shards to remoteAvailableShards
2020-04-05 18:37:11 -05:00
Todd Gruben
6bd81b87eb
unexport availableShardFileFlushDuration
2020-04-05 18:37:11 -05:00
Todd Gruben
000ea90877
unbuffer channel
2020-04-05 18:34:43 -05:00
Todd Gruben
af068e08e7
cleanup and comments
2020-04-05 18:34:43 -05:00
Todd Gruben
40a3dce93c
Co-authored-by: Travis Turner <github@calfrope.com>
2020-04-05 18:34:43 -05:00
Todd Gruben
bb04f7f6ac
limit frequency of writes for available shards
2020-04-05 18:31:28 -05:00
Cody Soyland
0a86f6a97b
Increase gRPC maximum message length
2020-04-03 16:35:57 -05:00
Travis
b7fcfb41d9
add proto OldMin/OldMax for backward compatibility
2020-04-03 12:25:07 -05:00
Travis Turner
48879107b0
Merge pull request #244 from travisturner/fix-race
...
make sure frag.maxRow() call is lock-protected
2020-04-02 23:52:38 -05:00
Travis
b4f1781d30
make sure frag.maxRow() call is lock-protected
2020-04-02 23:25:57 -05:00
seebs
632f6856ab
Merge pull request #229 from seebs/checkptr
...
disable checkptr with go 1.14
2020-04-02 22:19:45 -05:00
seebs
a20532073b
Merge branch 'enterprise' into checkptr
2020-04-02 20:48:49 -05:00
