Commit Graph

254 Commits

Author SHA1 Message Date
Ben Kochie
f8af350ae2 Merge pull request #346 from mcdan/people/mcdan/issues/219
Fix additional mdadm parsing cases
2016-11-17 21:13:38 +01:00
dan mcweeney
13aa37025f Feedback on PR, thanks @tcolgate for the review 2016-11-17 10:23:01 -05:00
Ben Kochie
4fd03c31e4 Merge pull request #323 from stuartnelson3/dfly-devstat
Dragonfly devstat
2016-11-17 13:33:50 +01:00
Ben Kochie
7a9aad01b4 Merge pull request #310 from stuartnelson3/dfly-cpu
export DragonFlyBSD CPU time
2016-11-17 13:33:11 +01:00
stuart nelson
e589a2b8af Remove gauges and convert to NewConstMetric format 2016-11-17 13:23:54 +01:00
stuart nelson
2b74cf7498 Export devstat for dragonfly 2016-11-17 13:23:54 +01:00
dan mcweeney
1f6b5aee39 #219 - add fixes for @samzhang111 super token 2016-11-16 14:49:57 -05:00
dan mcweeney
8d756cab50 Fixes end to end test 2016-11-16 14:47:03 -05:00
dan mcweeney
00c9a88a55 Fixes #219 - use the default to catch personalities that are unknown
Assumes all raid configurations start with raid and that anything
else is unknown.
2016-11-16 14:47:03 -05:00
Ed Schouten
9749c2c0b3 mdstat: Fix parsing of RAID0 lines that contain additional attributes.
We seem to have a small number of Linux servers here that have lines in
/proc/mdstat that cannot be parsed by the node exporter, due to them
containing attributes that are not matched by the regular expression
("super 1.2").

Extend the regular expression to skip this data, just like we do for all
of the other status lines.
2016-11-16 17:21:25 +01:00
Rene Treffer
abe8e297a6 Prefer device path based names over exported names (#334)
* Prefer device path based names over exported names

For some sensors (like coretemp) it is possible that multiple
instances exist, thus base the name on the device path and not on
the exported name.

* Update end-to-end test for dual socket machines

Explicitly have 2 coretemp instances with a symlink for the device
such that the hwmon collector must pick that name (or fail)
2016-10-28 20:25:44 +01:00
Ben Kochie
c6162312f2 Add Linux NUMA "numastat" metrics (#249)
* Add Linux NUMA "numastat" metrics
  Read the `numastat` metrics from /sys/devices/system/node/node* when reading NUMA meminfo metrics.
* Update end-to-end test output.
* Add `numastat` metrics as counters.
* Add tests for error conditions.
* Refactor meminfo numa metrics struct
* Refactor meminfoKey into a simple struct of metric data.
  This makes it easier to pass slices of metrics around.
* Refactor tests.
* Fixup: Add suggested fixes.
* Fixup:  More fixes
* Add another scanner.Err() return
* Add "_total" to counter metrics.
2016-10-12 13:07:49 +02:00
Rene Treffer
081ecc5db0 Add hwmon /sensors support (#278)
* Add hwmon support (mainly known from lm-sensors)

This commit adds initial support for linux hardware sensors, exported
through sysfs.

Details of the interface can be found at
https://www.kernel.org/doc/Documentation/hwmon/sysfs-interface

* Add end-to-end test with some real life data

* Cleanup comments on hwmon collector

* Drop raw sensor name from hwmon output

* Let the sensor label be "sensor"

* Add hwmon short description to README.
2016-10-06 16:33:24 +01:00
stuart nelson
450fe0f3ba Add test 2016-09-28 09:10:05 +02:00
stuart nelson
cf3710191a Compile meminfo for dfly (#315)
* Compile meminfo for dfly

* Update README.me
2016-09-28 08:08:19 +01:00
stuart nelson
ef1925db7d Compile netdev on dragonfly (#314)
* Compile netdev on dragonfly

* Only run netdev bsd test on bsd

* Update README.md
2016-09-27 21:44:13 +01:00
stuart nelson
ee37a27d91 Export values as uint64_t 2016-09-20 23:27:56 +02:00
stuart nelson
e942d7e234 Maintain granularity in cpu data
Export cpu mode times as original uint64_t data,
and update frequency, and do the conversion to
float64 and subsequent division in go.
2016-09-20 09:10:53 +02:00
Ben Kochie
afac1f7433 Update mdstat fixture based on linux source.
Update `Contains` matching for `resync=`
2016-09-19 16:11:16 +02:00
stuart nelson
57f88ac4f6 Update comment 2016-09-19 09:48:53 +02:00
stuart nelson
78c84b1a47 Remove old freq finding code
This is the code that was lifted from the freebsd
implementation, but was not correct.
2016-09-19 09:48:34 +02:00
stuart nelson
45ac033d9e Use correct frequency for calculating cpu time
The correct frequency is the systimer frequency,
not the stathz.

From one of the DragonFly developers:

The bump upon each statclock is:
((cur_systimer - prev_systimer) * systimer_freq) >> 32

systimer_freq can be extracted from following
sysctl in userspace:
sysctl kern.cputimer.freq
2016-09-19 09:35:41 +02:00
stuart nelson
8cc06aab04 Remove unneeded ncpu variable 2016-09-18 17:36:39 +02:00
stuart nelson
9f7822ccdc Remember to bzero string
Duplication was caused by malloc returning a
region of memory that already had data in it.
2016-09-18 16:17:49 +02:00
stuart nelson
c02dcdeb35 Remove unused comment. 2016-09-18 14:21:54 +02:00
stuart nelson
3e4a154656 Correctly exporting values
Moved to exporting via a string, which is then
split and parsed.

The string is sometimes duplicated, however.
2016-09-18 14:16:26 +02:00
Ben Kochie
64b82596ef Fix mdadm collector for resync=PENDING.
Add fix for mdadm devices in state `resync=PENDING`.
* Update test and fixture.
2016-09-18 08:30:20 +02:00
stuart nelson
4b4385bd44 Remove free
Don't need it since we aren't malloc'ing
2016-09-17 19:14:31 +02:00
stuart nelson
cc5142a007 Remove some unnecessary includes 2016-09-17 19:13:05 +02:00
stuart nelson
a9d27ea722 Appears to be working? Still need to cleanup after self 2016-09-17 18:50:08 +02:00
stuart nelson
1d75b376ae checkpoint 2016-09-17 14:43:29 +02:00
stuart nelson
1b7a18c271 Creating slice from C-array
Might not be lined up correctly? Weird output data
in the second CPU.
2016-09-17 14:05:36 +02:00
stuart nelson
61f36ac1ab Activate filesystem collector on DragonFly (#302) 2016-09-11 12:08:00 -04:00
stuart nelson
03da1ea516 successfully exporting one of 2 cpus 2016-09-10 21:13:06 +02:00
stuart nelson
cda2f493ee wip 2016-09-10 19:22:34 +02:00
Takuya Nakaike
0f2a87ab22 add s390x support 2016-08-24 08:06:04 -04:00
Calle Pettersson
e890743f79 Add white/blacklisting of units for systemd collector 2016-08-20 13:08:03 +02:00
Julius Volz
428746c48e Merge pull request #285 from prometheus/meminfo-bsd-constmetrics
Convert BSD meminfo collector to use ConstMetrics
2016-08-14 20:14:13 +02:00
Julius Volz
f0796683fe Convert BSD meminfo collector to use ConstMetrics
This suffers from the same concurrency bug as the netstat one:
https://github.com/prometheus/node_exporter/issues/280
2016-08-13 00:40:29 +02:00
Julius Volz
c0b46391e2 Merge pull request #283 from prometheus/meminfo-constmetrics
Convert meminfo collector to use ConstMetrics
2016-08-12 19:02:30 +02:00
Brian Brazil
25289c5024 Merge pull request #284 from prometheus/sockstat-constmetrics
Convert sockstat collector to use ConstMetrics
2016-08-12 00:42:23 +01:00
Julius Volz
cef3d98256 Convert sockstat collector to use ConstMetrics
This suffers from the same concurrency bug as the netstat one:
https://github.com/prometheus/node_exporter/issues/280
2016-08-12 01:36:42 +02:00
Julius Volz
eac396c637 Convert meminfo collector to use ConstMetrics
This suffers from the same concurrency bug as the netstat one:
https://github.com/prometheus/node_exporter/issues/280
2016-08-12 01:30:15 +02:00
Julius Volz
f91bca427b Convert filefd collector to use ConstMetrics
This suffers from the same concurrency bug as the netstat one:
https://github.com/prometheus/node_exporter/issues/280
2016-08-12 01:26:25 +02:00
Julius Volz
9128952454 Fix end-to-end tests after netstat conversion 2016-08-12 01:09:20 +02:00
Julius Volz
8c5c880705 Convert netstat collector to use ConstMetrics
Fixes https://github.com/prometheus/node_exporter/issues/280
2016-08-12 00:34:00 +02:00
Thomas Frössman
32e3445d72 Fix mdstat tabs parsing 2016-08-06 14:08:11 +02:00
Daniel
c3eba340e3 Update e2e-output.txt 2016-07-09 15:36:52 +02:00
Daniel
87b3df90e6 Update diskstats 2016-07-09 15:23:31 +02:00
Daniel
a7a8d088da Ignore partitions on NVME devices by default
The convention of the linux driver is nvme($device)n($namespace)p($partition). On *bsd it seems to be different, using "ns" instead of "n" as the namespace separator.
2016-07-09 13:23:01 +02:00