prometheus/cadvisor
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
2,469 Commits 1 Branch 39 Tags 50 MiB
f988ade02d
Commit Graph

2469 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
David Ashpole
436d75ce8d
fix #1902 bug with retryDockerStatus (#1903) 2018-03-14 16:58:57 -07:00
nielsole
08f0c2397c Adding /proc/<pid>/schedstat (#1872) 
Add /proc/<pid>/schedstat metrics for scheduler metrics
 2018-03-08 09:27:06 -08:00
David Ashpole
b817801307
Merge pull request #1894 from Random-Liu/avoid-containerd-race 
Use backoff to tolerant race condition with containerd.
 2018-02-23 14:23:09 -08:00
Lantao Liu
d5ee05fc25 Use backoff to tolerant race condition. 
Signed-off-by: Lantao Liu <lantaol@google.com>
 2018-02-23 22:04:35 +00:00
David Ashpole
8ec51bb848
Merge pull request #1887 from dashpole/refactor 
Refactor handlers to move common code to libcontainer handler
 2018-02-21 09:01:06 -08:00
David Ashpole
e1d602d7af create libcontainer handler for common code 2018-02-21 08:53:42 -08:00
David Ashpole
c094ef0d2a
Merge pull request #1859 from andyxning/reduce_labels_for_container_info 
reduce labels for container info
 2018-02-21 08:33:17 -08:00
David Ashpole
aaaa65dba0
Merge pull request #1893 from dashpole/changelog_v0.29 
changelog for v0.29.0
 2018-02-20 14:57:56 -08:00
David Ashpole
b9761a74ff changelog for v0.29.0 2018-02-20 14:32:30 -08:00
David Ashpole
3a4fae2b3f
Merge pull request #1888 from dashpole/test_doc 
Add node-e2e based testing instructinos
 2018-02-20 08:47:48 -08:00
David Ashpole
6ab10fe433
Merge pull request #1852 from bboreham/prom-total-cpu 
Add Prometheus total CPU metric and allow per-cpu stats to be disabled
 2018-02-20 08:47:26 -08:00
Bryan Boreham
ec6da3acae Prometheus metrics: optionally export total CPU instead of per-CPU 
Per-CPU stats are more expensive to transport and store, and that
level of detail is not required in many cases.

We export overall total cpu in the same metric as per-cpu, so that
dashboards which previously summed over cpu will work identically.
 2018-02-20 13:58:44 +00:00
David Ashpole
6116f26530
Merge pull request #1882 from dashpole/overlay_upper 
monitor only the upper directory for overlay
 2018-02-08 11:18:43 -08:00
David Ashpole
eb15723cb0
Merge pull request #1886 from miaoyq/singleton-client 
All handlers of containerd containers share a single client instance
 2018-02-08 10:46:41 -08:00
David Ashpole
c13b7bb544 add node-e2e based testing instructinos 2018-02-08 10:45:55 -08:00
Yanqiang Miao
1dd9469469 All handlers of containerd containers share a single client instance 
Signed-off-by: Yanqiang Miao <miao.yanqiang@zte.com.cn>
 2018-02-08 12:21:51 +08:00
David Ashpole
a31881621b
Merge pull request #1885 from Random-Liu/no-uuid-error-out 
Do not error out when /dev/disk/by-uuid is not found.
 2018-02-07 11:33:28 -08:00
Lantao Liu
ef7f0696fe Do not error out when /dev/disk/by-uuid is not found. 
Signed-off-by: Lantao Liu <lantaol@google.com>
 2018-02-07 08:34:19 +00:00
David Ashpole
aa6d6f1195
Merge pull request #1871 from jsravn/connect-to-docker-forever-on-startup 
Connect to docker with no timeout on startup
 2018-02-06 15:29:26 -08:00
James Ravn
e660d8b8b7 Retry docker status on startup 
For https://github.com/google/cadvisor/issues/1866.
 2018-02-06 10:31:11 +00:00
David Ashpole
812cc819ac monitor only the upper directory for overlay 2018-02-05 15:49:27 -08:00
Bhuvanchandra DV
cbf13cdfa2 cadvisor: build: fix broken build with Makefile 
When using `make build` to compile the source,
build fails with error/s. The reason is because
of GO_CMD variable is set to 'install' as default,
which is not valid when running `make build` or
`make`.

Depending on the valid git tag, release.sh will
set the GO_CMD variable. If valid git tag is
available e.g.: vx.y.z.beta1, then GO_CMD is
set to `build`.

If valid git tag is not available, release.sh
will any way bails out serving the purpose.
So drop using GO_CMD, to fix the default build.
 2018-02-05 19:17:10 +05:30
David Ashpole
828ac5eb11
Merge pull request #1878 from dims/fix-warning-to-warningf 
Fix Warning->Warningf for better logging
 2018-02-02 16:42:02 -08:00
Davanum Srinivas
b1656b253f Fix Warning->Warningf for better logging 2018-02-02 19:19:07 -05:00
David Ashpole
05f9f74aa4
Merge pull request #1868 from dashpole/migrate_prow 
Migrate to prow
 2018-02-02 14:47:00 -08:00
David Ashpole
66e12ca652 migrate to prow, which uses node-e2e to run tests 2018-02-01 15:20:53 -08:00
Derek Carr
b111f65653
Merge pull request #1861 from vikaschoudhary16/bump_runc 
Bump runc to d5b4a3eddbe4c890843da971b64f45a0f023f4db
 2018-01-12 15:18:01 -05:00
vikaschoudhary16
13d955d6a9 Bump runc to d5b4a3eddbe4c890843da971b64f45a0f023f4db 
Signed-off-by: vikaschoudhary16 <vichoudh@redhat.com>
 2018-01-12 13:56:11 -05:00
Andy Xie
1ccbe6fdd0 reduce labels for container info 2018-01-12 00:14:01 +08:00
David Ashpole
5a2e3d06d6
Merge pull request #1845 from dashpole/fix_ui 
Fix small UI issues
 2018-01-02 15:20:15 -08:00
David Ashpole
aacb48c7d0
Merge pull request #1847 from dims/skip-get-vfs-stats-when-mountpoint-does-not-exist 
Skip getVfsStats when file does not exist
 2018-01-02 15:08:53 -08:00
Davanum Srinivas
5f8eea99dd Skip getVfsStats when file does not exist 
There are a lot of spurious exceptions in the kubernetes kubelet logs
like:
E1018 21:03:09.616581   22780 fs.go:332] Stat fs failed. Error: no such
file or directory

Since we know that calling syscall.Statfs will just fail when the path
does not exist, we should just skip making the call.

NOTE: fixing 2017->2018 problems in build by running `./build/jenkins_e2e.sh`
 2018-01-02 17:52:38 -05:00
David Ashpole
9ffa37396f
Merge pull request #1806 from sentinelt/master 
fix #1607; use container creation time provided by Docker handler
 2017-12-20 11:25:35 -08:00
David Ashpole
e917087ad5 fix small logo and slash in container 2017-12-19 13:28:26 -08:00
David Ashpole
1e567c2ac3
Merge pull request #1835 from dashpole/changelog_0.28.3 
changelog for v0.28.3
 2017-12-07 09:52:08 -08:00
David Ashpole
e6b4e4c38a changelog for v0.28.3 2017-12-07 09:46:35 -08:00
David Ashpole
fc6d4b920c
Merge pull request #1830 from jsravn/add-docker-timeouts 
Add timeouts for docker queries
 2017-12-07 09:37:53 -08:00
James Ravn
57e17d8be2 Add timeouts for docker queries 
As these can otherwise block indefinitely due to docker issues.

This is to fix https://github.com/kubernetes/kubernetes/issues/53207,
where kubelet relies on cadvisor for gathering docker information as
part of its periodic node status update.
 2017-12-05 13:50:48 +00:00
David Ashpole
0bde1c615c
Merge pull request #1831 from brian-brazil/prometheus-labels 
Ensure all Prometheus metrics have the same labelnames.
 2017-11-30 09:54:44 -08:00
Brian Brazil
27f103b266 Ensure all Prometheus metrics have the same labelnames. 
Fixes #1704
 2017-11-30 16:33:37 +00:00
David Ashpole
7d11f4243f
Merge pull request #1827 from tallclair/logging 
Clean up cAdvisor logging
 2017-11-29 10:16:58 -08:00
David Ashpole
b26bf6ebb2
Merge pull request #1826 from mindprince/gpu-docs 
Add docs for using nvidia gpu monitoring.
 2017-11-28 17:49:29 -08:00
Tim Allclair
1eb1355ae6
Default logging to V(2) 2017-11-27 19:49:49 -08:00
Tim Allclair
5b435b4b70
Clean up cAdvisor logging 2017-11-27 19:48:05 -08:00
Tim Allclair
3a40bbfc5c
Raise verbosity on runtime registration failure 2017-11-27 19:48:04 -08:00
Rohit Agarwal
6ba3fa4e8c Add docs for using nvidia gpu monitoring. 2017-11-27 17:43:14 -08:00
David Ashpole
49440c7e0a
Merge pull request #1818 from dashpole/changelog 
changelog for v0.28.2
 2017-11-21 16:32:31 -08:00
David Ashpole
9689d84e7f changelog for v0.28.2 2017-11-21 16:27:22 -08:00
David Ashpole
e420065e7d
Merge pull request #1817 from dashpole/util_clock 
Switch from apimachinery clock to k8s.io/utils/clock
 2017-11-21 16:24:51 -08:00
David Ashpole
3166cdae87 add utils/clock dependency 2017-11-21 16:19:57 -08:00