Page MenuHomePhabricator
Create Task
Maniphest T211684

Toolforge: Port sge.py stats to Prometheus
Closed, ResolvedPublic
Actions

Description

Convert sge.py from a Diamond collector to a Python script that will make metrics available to node-exporter.

Find another submit host to run it on instead of tools-services (so they can be non-submit hosts to simplify infrastructure).

Details

SubjectRepoBranchLines +/-
toolforge: Prometheus replacement for sge.py diamond collectoroperations/puppetproduction+167 -0
Customize query in gerrit

Related Objects
Search...

Event Timeline

• GTirloni triaged this task as High priority.Dec 11 2018, 1:16 PM
• GTirloni created this task.
Restricted Application added a subscriber: Aklapper. · View Herald TranscriptDec 11 2018, 1:16 PM
• GTirloni moved this task from Inbox to Doing on the cloud-services-team (Kanban) board.Dec 11 2018, 7:36 PM
aborrero added a parent task: T207591: tools-services: Migrate to Stretch.Dec 14 2018, 1:54 PM
bd808 mentioned this in T214182: Find tools.tools-services-01.sge.hosts.* metrics replacement.Jan 18 2019, 6:05 PM
• GTirloni removed • GTirloni as the assignee of this task.Jan 18 2019, 6:12 PM

https://github.com/prometheus/node_exporter#textfile-collector

Dzahn unsubscribed.Jan 18 2019, 6:39 PM
bd808 claimed this task.Jan 19 2019, 7:33 PM
bd808 edited projects, added Toolforge; removed Goal, SRE.
gerritbot subscribed.Jan 19 2019, 7:39 PM

Change 485372 had a related patch set uploaded (by BryanDavis; owner: Bryan Davis):
[operations/puppet@production] toolforge: Prometheus replacement for sge.py diamond collector

https://gerrit.wikimedia.org/r/485372

gerritbot added a project: Patch-For-Review.Jan 19 2019, 7:39 PM
gerritbot added a comment.Jan 23 2019, 5:05 PM

Change 485372 merged by GTirloni:
[operations/puppet@production] toolforge: Prometheus replacement for sge.py diamond collector

https://gerrit.wikimedia.org/r/485372

• GTirloni added a comment.Jan 23 2019, 5:25 PM
# curl http://tools-sgegrid-master.tools.eqiad.wmflabs:9100/metrics | grep ^sge
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0sge_hostjobs{host="tools-sgeexec-0901.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0902.tools.eqiad.wmflabs"} 2
sge_hostjobs{host="tools-sgeexec-0903.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0904.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0905.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0906.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0907.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0908.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0909.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0910.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0911.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0912.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0913.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0914.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0915.tools.eqiad.wmflabs"} 2
sge_hostjobs{host="tools-sgeexec-0916.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0917.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0918.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0919.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0920.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0921.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0922.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0923.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0924.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0925.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0926.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0927.tools.eqiad.wmflabs"} 2
sge_hostjobs{host="tools-sgeexec-0928.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0929.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0930.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0931.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0932.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0933.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0934.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0935.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0936.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0937.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgeexec-0938.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0939.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0940.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0941.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgeexec-0942.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-generic-0901.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-generic-0902.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-generic-0903.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-generic-0904.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0901.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0902.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0903.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0904.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0905.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0906.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0907.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0908.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0909.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0910.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0911.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0912.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0913.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0914.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0915.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0916.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0917.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0918.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0919.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0920.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0921.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0922.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0923.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0924.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0925.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0926.tools.eqiad.wmflabs"} 1
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0927.tools.eqiad.wmflabs"} 0
sge_hostjobs{host="tools-sgewebgrid-lighttpd-0928.tools.eqiad.wmflabs"} 0
sge_jobseqnum 22173
sge_queuejobs{queue="continuous",state="r"} 20
sge_queuejobs{queue="task",state="r"} 6
100  100k  100  100k    0     0  5119k      0 --:--:-- --:--:-- --:--:-- 5268k
sge_queuejobs{queue="webgrid-generic",state="r"} 1
sge_queuejobs{queue="webgrid-lighttpd",state="r"} 9
bd808 added a comment.Jan 23 2019, 5:33 PM

Screen Shot 2019-01-23 at 10.32.07.png (106×1 px, 40 KB)

Looks like we might be having cross-region network issues between the eqiad1-r instances and https://tools-prometheus.wmflabs.org/tools/targets

Stashbot subscribed.Jan 23 2019, 5:44 PM

Mentioned in SAL (#wikimedia-cloud) [2019-01-23T17:44:46Z] <bd808> Added rules to default security group for prometheus monitoring on port 9100 (T211684)

bd808 closed this task as Resolved.Jan 24 2019, 4:00 PM
bd808 added a subscriber: Andrew.

The security group stuff needed a bit of help from @Andrew who figured out that we were out of quota in eqiad1-r for security groups/rules. After raising the limit things started working.

bd808 closed subtask T215845: Add monitoring for disabled grid nodes to the prometheus collector as Resolved.Aug 8 2019, 4:22 AM
bd808 mentioned this in T109730: Add monitoring for SGE queue status .Jan 4 2020, 10:22 PM
Content licensed under Creative Commons Attribution-ShareAlike (CC BY-SA) 4.0 unless otherwise noted; code licensed under GNU General Public License (GPL) 2.0 or later and other open source licenses. By using this site, you agree to the Terms of Use, Privacy Policy, and Code of Conduct. · Wikimedia Foundation · Privacy Policy · Code of Conduct · Terms of Use · Disclaimer · CC-BY-SA · GPL