062de961c9
Fix #524 . Arithmetic operator and where conditions don't play nice
2014-05-20 16:20:21 -04:00
bd03a579bd
update the year in the LICENSE file
2014-05-20 15:34:20 -04:00
890f291f28
Fix #550 . Fix tests on 32-bit ARM
2014-05-20 15:20:21 -04:00
9a06ec4031
Close #547 . Add difference function
2014-05-20 14:43:22 -04:00
86a80f7a09
fix the style of the changelog
2014-05-20 14:35:41 -04:00
af98ecaf27
Fix #529 . Don't run influxdb as root
2014-05-20 14:22:26 -04:00
9c8204b64d
Fix #460 . Don't start influxdb automatically after installing from pkg
2014-05-20 14:22:25 -04:00
b6a5a10912
Fix #557 . Using group by time(1y) doesn't work while time(365d) is fine
2014-05-20 13:15:24 -04:00
a64ee2ae1a
update the changelog
2014-05-19 20:41:45 -04:00
44688fc914
remove newlines from log statements
2014-05-19 20:20:39 -04:00
1369ca75fb
Fix #555 . Fix a regression in d593348 where the snapshot format changed
2014-05-19 19:44:35 -04:00
379215f9cc
add some logging
2014-05-19 19:28:32 -04:00
5b318548dc
fix a log statement
2014-05-19 15:51:43 -04:00
39e659cfc2
update the changelog
2014-05-19 15:31:44 -04:00
d1b0597044
Merge branch 'pr-551'
2014-05-19 15:30:30 -04:00
a08e63ea9c
more idiomatic go
2014-05-19 15:28:51 -04:00
3c8aafd46e
modify some tests
2014-05-19 15:22:49 -04:00
d382ea0b3e
Close #551 . Fix #409 . Add aggregate funciton top and low
...
this patch provides aggregate function top and low. these function are
useful when executing analytics or summarize queries.
Added functions:
low(columns..., limit)
top(columns..., limit)
For instance:
`select low(value, 10) from series` will output 10 `value` values with ascending order.
`select top(value, 10) from series` will output 10 `value` values with descending order.
How to build:
go get github.com/ryszard/goskiplist/skiplist
go build src/daemon.go
Currently, these function support float64 and int64 types.
Limitations:
* can't handle same value correctly (goskiplist limitation)
Misc:
* each column should support alias.
e.g) low(score as low_score, 10)
And, I'm not good at English and Golang. so if you find some wrong
points, plz feel free to fix it.
2014-05-19 15:22:49 -04:00
a4ece2a897
Update CHANGELOG.md
2014-05-16 17:58:27 -04:00
0744a292e2
Update CHANGELOG.md
2014-05-16 17:47:47 -04:00
6e81a63ccb
Merge pull request #545 from freeformz/typo
...
Fix typos
2014-05-16 15:30:55 -04:00
6a74cfdee1
Merge pull request #546 from peekeri/graphite_udp_api
...
graphite api: add support for sending metrics over UDP
2014-05-16 14:44:18 -04:00
50d8050e10
graphite api: add support for sending metrics over UDP
...
There already exists an API for sending metrics to InfluxDB over TCP using
Graphite plaintext protocol. Sometimes it is more convenient to send metrics
over UDP, so adding that support for the Graphite as well. For now at least,
using the same port for UDP traffic than for TCP to make configuration really
simple.
UDP listener is disabled by default, but can be turned on by setting udp_enabled
to true in graphite plugin configuration.
2014-05-16 10:52:07 +03:00
29d6330cfb
Fix typos
2014-05-15 16:40:05 -07:00
bd33adf4c2
Fix #540 . Add a test to make sure regex matching with literal dot work
2014-05-15 15:02:34 -04:00
d593348283
Fix #489 . Remove replication factor from CreateDatabase command
2014-05-15 14:41:45 -04:00
702bf3d7b5
update the changelog
2014-05-15 14:33:48 -04:00
964f5b23b6
don't append to the original slice
2014-05-15 14:23:08 -04:00
b3fac38501
Optimize deletes
2014-05-15 14:23:07 -04:00
8286cede10
Optimize writing to multiple series
2014-05-15 14:22:50 -04:00
d6050c9ca1
remove some unused constants
2014-05-15 14:22:50 -04:00
998cf1e55c
Make the write batch size configurable
...
It turns out that smaller batch sizes are causing too much
overhead. It's not clear why, but empirical testing shows a 10x increase
in write speed when the batch size increases from the previous hard
coded 64K to 10M.
2014-05-15 14:22:49 -04:00
c7c79bc441
Fix #531 . Compare s1's to s2's name
2014-05-14 11:24:37 -04:00
73f9c4439b
Merge pull request #220 from xiangli-cmu/fix_join
...
fix(join safty) only allow commit to the most recent join command if there is one
2014-05-14 10:57:53 -04:00
645aac7006
update the changelog
2014-05-13 18:26:56 -04:00
2e0f6c1700
Fix #534 . Create a new series when interpolating
2014-05-13 18:21:00 -04:00
2d11e2271c
Use the abbreviated commit sha
2014-05-13 17:00:20 -04:00
3479a7a0b5
Fix #520 . Print influxdb's version in the log
2014-05-13 17:00:20 -04:00
b2f229ef09
update the changelog
2014-05-13 16:49:37 -04:00
9f95a44e3d
Fix #539 . count(distinct()) with fill shouldn't panic on empty groups
2014-05-13 16:48:41 -04:00
ad7bfa7aa0
rename the test
2014-05-13 16:38:22 -04:00
4813ddb7ca
fix the test
2014-05-13 16:36:13 -04:00
c3ab1bc0a5
add a shout out in the changelog
2014-05-13 16:15:06 -04:00
51306db3c9
Close #505 . Expose the version in the response headers
2014-05-13 16:14:41 -04:00
1ea96952db
Fix #536 . New nodes should add old shards without panicing
2014-05-13 15:58:49 -04:00
04256a8426
Revert "Fix #526 . New nodes should add old shards without panicing"
...
This reverts commit ac98817a55
2014-05-13 15:58:32 -04:00
ac98817a55
Fix #526 . New nodes should add old shards without panicing
2014-05-13 15:56:30 -04:00
2b61feea26
Fix #538 . Don't panic when the same series is written with different columns
2014-05-13 15:45:12 -04:00
bc61f562d5
fix the order in the log
2014-05-13 15:41:00 -04:00
4d6d7ff23d
Update CHANGELOG.md
2014-05-13 15:24:41 -04:00
John Shahid
Matthijs Boelstra
Shuhei Tanuma
Paul Dix
Jari Sukanen
Edward Muller
Xiang Li
Stefan Majer