CI-users

ci-users@lists.centos.org

1 participants
418 discussions

ci.centos.org Maintenance Thursday, 10-December-2015 01:00 UTC (20:00 EST)
by Brian Stinson 11 Dec '15

11 Dec '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Folks, There is a Jenkins update (including security fixes) scheduled for next week. We will be rebooting Jenkins to apply this fix, and to clear the backlog of config updates (adding IRC channels, plugins, etc.). We do not expect a long outage, but we'll schedule 1hr to be safe. As usual we will pause jobs 1 hour beforehand to let things clear out, and will post updates here and to #centos-devel on freenode. Cheers! - -- Brian Stinson CentOS CI Infrastructure Team -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJWYJBpAAoJEIMGvNKzCweM7/cP/1H5MQYoXFkDbPX/Wiar6+PH lxXFVWeKk/DMV2m5FGHbpe4JFJP5rLNs1wRc/5eAJWa6iVnxaGJ/34jGqjBBhJDN p9CisRvPvFQP+0/0jUhc460MuAbtTCgO7bmRyuCWMxzO+UH5+62mD1M+tblcR7F7 S4gFOnkSwjOWRIFgCDsfAS6fh6KFt/9nX10BjdDeCxad5rbhwy2iYcAT71kzjtD2 yYZ4/GdXHqWaO12dCvB4q8ndb9J6+s+vORb0UlzrrRezbkWFY7s+v2Sah4ypTcbN FAkJe7rwDKzK6WDXmBAMYNsmRyi2DJ1PR4DlHkBS9ootMUlNYye1dio/5C0+681B Xldy4NHR+Gs7m80XSpXGn3LsUyGdk43Ks8XZGl8vwj0v532qBdL7oM96mg2gbxfe OPpDKsSNtdznjysjeYmK6emkQ/+cKb4OZ7NBF1+Y9kviqCRHyn7r/e+kQdn+/P9v AIk7VmNKp3VosPoNWUCUiNobilaW4hOajcKFpkjFBN4fu+99wqmNrdJk9KwgBy4s 9CEtEtivmhLT0v49D1aiNEqIHLv4qmn+W/SjSIo4oqJ6SWAT2tLe+nPZj+Wdwdmv SMxjCSbQoAGEDoHEKXgTteLl2xa6h+PJV75LvocwdKoLHhuRDxmCy8V4Gy7P8muS 7tY25FBBs1Tw57CSX9GP =o79f -----END PGP SIGNATURE-----

1 2

openstack-puppet orphaning machines
by Karanbir Singh 07 Dec '15

07 Dec '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi Guys, We've had a few sessions get orphaned in the openstack-puppet setup these last few days ( we reclaimed 36 hosts day before, and 7 more today ). These are machines that were blocked for 24hrs+ and not released back to the deployment pools. Its not a problem, we expect to be reaping machines; just wanted to make sure the relevant people were aware of it. regards - -- Karanbir Singh, Project Lead, The CentOS Project +44-207-0999389 | http://www.centos.org/ | twitter.com/CentOS GnuPG Key : http://www.karan.org/publickey.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJWXXQRAAoJEI3Oi2Mx7xbtvRoH/2H87GSZYyYvuWtX8tXKWh+K vaFtmPvvnTbGvny1f2BlTM8Jc4wwN+JZ8iNyi7A/Dy2/UUdnyK5zvgX9TAceBkFm gE/ytYgy6qez+SDSsRP+Oh2n01qNDZAJoOg5nU7A1oLDqA+qxotjTMjgPEtIwYW0 O6q266Tx6wyAq+Pjlnh+b3eFrQgN2DR9A4SM6vDSzowoSRKT40e7G3vrqbDRopls 43nsSWsSLBsMPlX7GGeb1mHVxfossfTCleGcONQN/yAu4kANkoV4WVGxhnsVP4du DP8wPJTD5pnyTcO13tdIFXy7s3tEC0hz75SvObqKPJy3MrfVRNaRnxxSgqCXt2M= =aDzi -----END PGP SIGNATURE-----

3 2

upcoming CentOS Linux 7 release
by Karanbir Singh 02 Dec '15

02 Dec '15

hi guys, RHEL 7.2 was released a few days back and we've been busy sorting through the code and working on the downstream CentOS Linux release ( its going to likely be released in November and will be called 7.1511 ). I'd like to make that content available on ci.centos.org ahead of public release. And there are a couple of ways to do this : 1) either we update the provisioning service, so all requests for a centos linux 7 machine will give you a 7.1511 build ( its 7.1503 at the moment ), with all the updates etc posted in sync that way. This also means that mirror.centos.org inside the ci infra will look DIFFERENT from mirror.centos.org the real users see, since it will be setup to look like the next release. 2) we make the content available in a new set of repos, called CentOS.next or something; and the ci users are able to optionally tweak their build scripts etc to include or not include these new changes. 3) we dont make this content available before its publicly released. of all those bits, I like (1) best; (2) only delays the content from hitting the ci infra by a week or 10 days. We dont usually take longer than that from when the builds are done to when the distro update is released. (3) just means that if there are changes, failures or breakages in the ci tested code, we only find out at the same time as the users - which isnt nice. thoughts ? -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

3 4

Jenkins trigger/notification issues
by James 01 Dec '15

01 Dec '15

I've finally got jenkins building working when clicking "Build Now". w00t. Thanks to bstinson for finding one of the issues. I'll write this up, but before I do that, I've got three open issues: 1) IRC notification doesn't work on post-build -- It's not even clear where you set the server, only the channel name field is visible 2) Email notification didn't work. 3) Most importantly: triggering the builds: best I can come up with is putting a git post-update hook in my local repo, and `wget` the special: "Trigger builds remotely (e.g., from scripts)" url when I want to run it. This actually works, but it requires you to actually already be logged in to jenkins, and I wasn't sure how to dump in my username and password into a wget command. No I don't want to screenscrape, phantomjs, or wget with my firefox session cookies! I also don't want to "poll scm" because that's mean/ugly. Thanks, James @purpleidea

2 5

Request to add an IRC channel to ci.centos.org jenkins
by Ben Kero 25 Nov '15

25 Nov '15

Hi there, I am doing some work building some CI for the RDO OpenStack Puppet Modules, and would like to enable IRC notifications for my builds. The channel I would like to add is #rdo-opm-ci on Freenode. -- Ben Kero RedHat, Senior Software Engineer, OPM-CI

2 1

RFE Ability to select specific chassis
by John Trowbridge 25 Nov '15

25 Nov '15

Howdy folks. How feasible do we think $subject is? For RDO-Manager jobs it is pretty much near zero chance of success on a machine with 16G of RAM. This is especially true for our HA jobs. I made a bit of a hack to our centosci provisioner in khaleesi[1] to throw out all hufty chassis nodes since they only have 16G RAM. This is slightly wasteful though, and would be better to be something that could be selected via the API. Thanks, John Trowbridge [1] https://github.com/redhat-openstack/khaleesi/commit/14d19c317c0a3fb6b551f4b…

2 3

Upcoming Feature: extendable sessions
by Karanbir Singh 24 Nov '15

24 Nov '15

hi, I'm going to be working on extending the admin.ci API. The intention is to add the following calls: /Session/get?key=<api key>&alias=<some text> : this will create a new session for the api key specified, but will not allocate any nodes to this session. The alias param is optional, if specified the session can be referenced with that alias ( in addition to / rather than the ssid ). /Session/done?key=<api key>&ssid=<ssid | alias for session > : this will tear down the session, including all Nodes allocated to the session. The ssid can be specified as either the UUID based ssid returned from the /Session/get call or the alias requested. In addition to these two new calls, the /Node/get call will get an optional ssid= param; calling /Node/get with an ssid will add the requested nodes to the existing session, rather than create a new session. The ssid can be either the uuid based ssid allocated at /Session/get or an alias to an existing session. The /Node/done call will also get an optional additional param in node= where you can specify the hostname of the machine to be dropped from an existing session, without terminating the entire session. With these four changes going in, it would be possible to setup a session, add nodes to that session as they are needed, drop hosts as required, and then tear down the session when done. The alias option could be used for various things, including accounting by job name, job type etc. Existing scripts that use the API will not need any changes, there will be no change to the existing calls. -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

3 3

Network stability/reliability accessing external resources
by David Moreau Simard 16 Nov '15

16 Nov '15

Hi, I've heard that network connectivity for accessing resources outside of the ci.centos.org infrastructure can sometimes be unreliable. I haven't yet experienced this first hand as the jobs I'm constantly monitoring and am responsible for have not yet been migrated there. A lot of things can be put within the ci.centos.org infrastructure if need be but it's unrealistic to have *every* resource there. Is anyone aware of ongoing issues ? Should we report jobs that fail due to network errors ? A recent example is this job that failed due to DNS resolution errors [1] "Could not resolve host: trunk.rdoproject.org; Name or service not known". Thanks, [1]: https://ci.centos.org/view/rdo/job/rdo_manager-promote-build-delorean-rdo-l… David Moreau Simard Senior Software Engineer | Openstack RDO dmsimard = [irc, github, twitter]

3 3

Emergency ci.centos.org Maintenance 12-Nov-2015 01:00 UTC (20:00 EST)
by Brian Stinson 13 Nov '15

13 Nov '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi All, We will be rebooting Jenkins to apply an update that contains patches for multiple security vulnerabilities: https://wiki.jenkins-ci.org/display/SECURITY/Jenkins+Security+Advisory+2015… We will pause accepting new jobs 1 hour beforehand to let things settle. We do not expect a long outage for this update. As usual, we will post updates to this list and be available in #centos-devel on freenode for any questions. Cheers! - -- Brian Stinson CentOS CI Infrastructure Team -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJWQ2UKAAoJEIMGvNKzCweMMxcP/i9X0f6ETaLqlZ+/3sJ4vJ9f l4u7vwqFAj+FkrxLyvWNp/NLOh2EiYtXIM7oenQaI3Iy3xGKPgmxy7hf6ptpFeqZ ndh+vYCFr3fcGopzCsnk1qi4ENkAE+X7QRCZVve3FKHcfDmfz5HHMhkf7xMdcK62 //t91f5bcS49hBCKHF1v1hdrd8e5QzeGWYycKGK0N3NEubH4ruml5IgSF51ylYJM derKn45E+bn5iINBGxiT/OEIdl22ksvXSKOc2bzp+kkrzj/kqW67kGjlPIVSYQMF A1UJQNe8QkTtAvPtRalffJtmJU1ERfkAW4Nvm90hB0CUvDytEmp66Ks08FiSgAJa BrnhRe4rId3xPl95+yLzaD0IXR/lYOCAa/0FOAE1ynwm5YYvdak1yuSUL69MUGX8 VhDE9JhpUFKq0n8wiE5rAxAB51Q0Jk2c42mfPhMm/NISIpq4DHUNAH/TlNvB4ffg t/AhSEo49D858Eqe3/oGj3c/DOx+e1G073IOqyj7FXNFj/DQ/bhVk304j5wevTC7 j1bMca9l16M/FJlqBDejki5a3JVWvI9+j/F0Z3BJhYaaoNqaYqRoL1PnrHPIlB7e WEP9fhy8VpPQ/UKpBgiXxBkzSHr5wOAlz4XCuOczU2JllufLcDUKFZfIqIOvbY3z +d4iIjPD9DqqTLmh9gbx =2h3g -----END PGP SIGNATURE-----

1 2

Maintenance Period: Pausing jobs in 15 Minutes
by Brian Stinson 02 Nov '15

02 Nov '15

Hi Folks, We'll be pausing jobs starting at 2300UTC in preparation for our Maintenance period. Stay tuned here, and to #centos-devel for updates. Cheers! Brian

2 2

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CI-users