Tobias Schmidt
e76096d4f6
Provide option to control Prometheus labels
...
This change generalizes the existing ContainerNameToLabelsFunc to allow the user to fully control all labels attached to exported Prometheus metrics. The existing behavior is available as DefaultContainerLabelsFunc and is used if no custom function is provided.
This will allow Kubernetes to filter out its internal Docker labels.
2016-08-22 17:26:36 -04:00
Tim St. Clair
2ed7198f77
Merge pull request #1423 from chenchun/memory_reservation
...
GetSpec missing memory reservation
2016-08-16 17:37:08 -07:00
Chun Chen
844c98e073
GetSpec missing memory reservation
2016-08-16 23:26:46 +08:00
Tim St. Clair
6213bfab8d
Merge pull request #1309 from f0/master
...
add cgropu swap usage and export as prometheus metric
2016-08-09 10:08:15 -07:00
Florian Koch
3ce98a46c4
add cgropu swap usage and export as prometheus metric
2016-08-09 07:33:37 +02:00
Vish Kannan
b1d8f6f900
Merge pull request #1386 from grobie/grobie/cpu-cfs-metrics
...
Expose cpu cgroup CFS prometheus metrics
2016-08-08 14:08:56 -07:00
Tobias Schmidt
1653733ea7
Expose cpu cgroup CFS prometheus metrics
...
If CPU quota is configured (cpu.cfs_quota != -1) the CFS will provide
stats about elapsed periods and throtting in cpu.stats. This change
makes these information available as container_cpu_cfs_* metrics.
2016-08-06 18:08:26 -04:00
Vish Kannan
f7a30e0b43
Merge pull request #1407 from timstclair/dockerfile
...
Update Dockerfile and dependencies
2016-08-05 17:22:36 -07:00
Tim St. Clair
8fa31bc627
Merge pull request #1411 from ncdc/thin-ls-kernel-check
...
Ensure minimum kernel version for thin_ls
2016-08-04 13:20:09 -07:00
Andy Goldstein
2b525ff87e
Ensure minimum kernel version for thin_ls
...
Ensure that kernel >= 4.4.0 or RHEL/Centos 7 kernel >= 3.10.0-366 exists before starting the thin
pool watcher. Prior versions have a bug in which reserving and releasing the metadata snapshot can
cause thin pool corruption.
2016-08-04 16:14:40 -04:00
Tim St. Clair
4530a1c7c6
Update Dockerfile and dependencies
...
- Update base image to alpine:3.4
- Fix zfs-utils reference (moved to zfs in main)
- Update glibc version to 2.23-r3
2016-08-02 15:30:02 -07:00
Tim St. Clair
c6c06d440a
Merge pull request #1365 from derekwaynecarr/expose_total_inodes
...
Expose total inodes
2016-08-02 10:56:24 -07:00
derekwaynecarr
6c114be580
Expose total inodes
2016-08-02 10:47:51 -04:00
Brent Hughes
e9f8dd31bb
Added CPU Usage details to statsd ( #1393 )
...
* Added CPU Usage details to statsd
* Fixed bad path to load average
* Fixed some issues with formatting and types. This is now tested
* Updated to make load average follow the same schema as other cpu metrics
2016-08-01 17:01:10 -07:00
Tim St. Clair
a40e0923a0
Merge pull request #1402 from jimmidyson/prometheus-docker-labels
...
Prefix Docker labels & env vars in Prometheus metrics to prevent conflicts
2016-08-01 13:38:59 -07:00
Jimmi Dyson
f1d8b2da70
Prefix Docker labels & env vars in Prometheus metrics to prevent conflicts
2016-08-01 21:33:31 +01:00
Tim St. Clair
bbe54c0a10
Merge pull request #1368 from derekwaynecarr/inodes_pointer
...
Allow clients to know if inodes data is available
2016-07-29 10:10:30 -07:00
Vish Kannan
8fc0cdbb85
Merge pull request #1359 from ncdc/thin-pool-watcher-logging
...
Make ThinPoolWatcher loglevel consistent
2016-07-28 09:56:08 -07:00
derekwaynecarr
cccf9d5fec
Allow clients to know if inodes are supported on a filesystem
2016-07-26 11:15:07 -04:00
Vish Kannan
04f73fefa5
Merge pull request #1395 from derekwaynecarr/cadvisor_wrong_values
...
filesystem usage stats are wrong in v2 api
2016-07-25 14:41:56 -07:00
derekwaynecarr
8551b0ebf7
filesystem usage stats are wrong in v2 api
2016-07-25 17:33:26 -04:00
Andy Goldstein
d05098b314
Lower ThinPoolWatcher cache miss log level
2016-07-23 07:07:42 -04:00
Andy Goldstein
b4cd05b879
Make ThinPoolWatcher loglevel consistent
2016-07-23 07:07:42 -04:00
Tim St. Clair
fe765981c2
Merge pull request #1391 from atombender/fix_prometheus_scrape_error_gauge
...
Fix for Prometheus metric "container_scrape_errors" getting stuck
2016-07-21 18:21:24 -07:00
Alexander Staubo
f8f474e791
This fixes a bug where any errors encountered during Prometheus metric
...
collection would stay counted in the gauge "container_scrape_errors",
making that particular metric useless. Instead, it must be reset on
every scrape.
2016-07-21 20:04:21 -04:00
Tim St. Clair
552d2578ec
Merge pull request #1379 from mwringe/collector_httpclient
...
Update collectors to use a customized httpClient.
2016-07-21 13:15:49 -07:00
mwringe
b8b541d86a
Update collectors to use a customized httpClient.
2016-07-21 16:00:21 -04:00
Vish Kannan
e47efa0e8a
Merge pull request #1390 from Random-Liu/bump-up-go-systemd
...
Bump up go-systemd to 4484981625c1a6a2ecb40a390fcb6a9bcfee76e3.
2016-07-21 11:04:02 -07:00
Lantao Liu
b351411510
Bump up go-systemd to 4484981625c1a6a2ecb40a390fcb6a9bcfee76e3.
2016-07-20 20:41:21 -07:00
Vish Kannan
533e6704d8
Merge pull request #1388 from vishh/docker-bump
...
Update docker to v1.11.2
2016-07-19 16:59:18 -07:00
Vishnu kannan
d4ce3dccdd
update docker to v1.11.2
...
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-07-19 16:55:05 -07:00
Tim St. Clair
734db4c437
Merge pull request #1387 from vishh/runc-bump
...
Update libcontainer deps
2016-07-19 12:45:32 -07:00
Vishnu kannan
4c0d60d9de
update libcontainer deps
...
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-07-19 12:35:14 -07:00
Tim St. Clair
510e6e8bfd
Merge pull request #1381 from Random-Liu/fix-watch-error
...
Continue watching other sub directories when there is watch error.
2016-07-18 18:24:06 -07:00
Lantao Liu
41e74494b3
Continue watching other directories when there is watch error.
2016-07-18 16:49:47 -07:00
Vish Kannan
0f253fe687
Merge pull request #1380 from mtaufen/working-set-fix
...
Modify working set memory stats calculation
2016-07-15 11:10:07 -07:00
Michael Taufen
307d1b1cb3
Modify working set memory stats calculation
...
Change working set calculation to usage - total_inactive_file, rather than
usage - total_inactive_anon - total_inactive_file. Since writes to tmpfs
get tracked as total_inactive_anon when swap is disabled, the old
calculation would under-report memory pressure.
See this Kubernetes issue for context:
https://github.com/kubernetes/kubernetes/issues/28619
2016-07-15 10:58:25 -07:00
Tim St. Clair
b37f9970ef
Merge pull request #1376 from mwringe/collector_endpoints
...
Update collectors to be able to directly use the ip address of containers
2016-07-14 18:03:59 -07:00
Matt Wringe
6ef612f21e
Update collectors to be able to directly access containers by their ip address.
2016-07-14 10:36:53 -04:00
Tim St. Clair
000cedb480
Merge pull request #1349 from mwringe/docker_ip_address_files
...
Add support to the handler to support fetching the ip address and files
2016-07-13 13:55:27 -07:00
Matt Wringe
213c254bb7
Add in the ability to expose the ip address of a container.
2016-07-13 13:58:43 -04:00
Vish Kannan
2290707a3b
Merge pull request #1369 from vishh/move-to-vendor
...
Switch to using vendor directory for deps.
2016-07-11 17:32:30 -07:00
Vishnu kannan
2eecea1353
Move presubmit to a make rule. Update docs to not use make all unless necessary.
...
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-07-11 17:27:24 -07:00
Vishnu kannan
c1f3340830
update jenkins script to use make
...
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-07-11 13:22:18 -07:00
Vishnu kannan
d2dba37eb8
ignore vendor pkgs from tests
...
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-07-11 12:40:17 -07:00
Vishnu kannan
38ea12a533
update scripts and docs to not mention godep for builds
...
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-07-11 12:40:05 -07:00
Vishnu kannan
de4af1288b
moving deps to vendor subdirectory
...
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-07-11 12:19:43 -07:00
Tim St. Clair
40fd72ef26
Merge pull request #1358 from ncdc/ignore-devicemapper-container-mounts
...
Exclude Docker devicemapper mounts from fs info
2016-07-07 13:17:52 -07:00
Andy Goldstein
a8c592e601
Exclude Docker devicemapper mounts from fs info
...
Exclude Docker devicemapper mounts from the list of filesystem partitions (e.g.
/var/lib/docker/devicemapper/mnt/*) because these are handled by the ThinPoolWatcher for
per-container filesystem information.
2016-07-07 14:35:22 -04:00
Andy Goldstein
dcf5d575eb
Always set partition fsType
2016-07-07 14:35:22 -04:00