May 2016 - lago-devel - oVirt List Archives

[2016-05-09] Lago weekly meeting minutes
by David Caro 09 May '16

09 May '16

Agenda ======== * oVirt system tests status: - 3.6 working - Master seems to fail on add-hosts still (investigating) * Merged workaround for reposync issues * Lago tutorial on it’s way - Using the unstable/latest lago (with the new patches) - Latest version: http://jenkins.ovirt.org/job/lago_master_check-patch-el7-x86_64/51/artifact… * Gal released 0.16.0 last week, kudos! * Merged patch to copy the deploy scripts inside the prefix on init, that takes us a step closer to relocatable prefixes (and exportable prefixes too) Let's have a great week! -- David Caro Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D Tel.: +420 532 294 605 Email: dcaro(a)redhat.com IRC: dcaro|dcaroest@{freenode|oftc|redhat} Web: www.redhat.com RHT Global #: 82-62605

1 0

[05-05-2016] New lago release - 0.16.0
by Gal Ben Haim 05 May '16

05 May '16

Hi everyone! There's a new lago release with version 0.16.0 ready for you to upgrade! One important change in version 0.16.0 is the support for the following environment variables: $LAGO_PREFIX_PATH = .lago/current $LAGO_WORKDIR_PATH = .lago it's now highly recommended to use those variables when configuring the path to the deployment scripts. Here are the other changes: * Wed May 04 2016 David Caro <dcaroest(a)redhat.com> - 0.16.0 45764a95: Fixed small typo in a string * Wed May 04 2016 Lago CI bot <dcaroest+cibot(a)redhat.com> - 0.15.5 66dec63b: Merge pull request #231 from lago-project/fix_el7_dirlock ccf6dca4: Added lock class for el7 locking 9a6e370c: Use lago.utils.LockFile instead of lockfile's one 98322f01: Removed thread pool on reposync 167c8b7d: Moved some code out of a with block db836eda: Yapf fixes * Tue May 03 2016 Lago CI bot <dcaroest+cibot(a)redhat.com> - 0.15.4 0e535676: Merge pull request #230 from lago-project/add_prefix_env_vars 17e9be31: tests:Modified deploy funct tests to use vars 1440adee: Add some lago extra vars to the env 577c442a: Expand vars and user contraptions on deploy scripts * Tue May 03 2016 Lago CI bot <dcaroest+cibot(a)redhat.com> - 0.15.3 a1a57978: Merge pull request #229 from lago-project/extra_gitignore 2a5a3359: Added extra ignores to .gitignore * Fri Apr 29 2016 Lago CI bot <dcaroest+cibot(a)redhat.com> - 0.15.2 25c91601: Merge pull request #227 from lago-project/fix_unhandled_exception 8eeb43cc: Fixed wrong error log on reposetup * Fri Apr 29 2016 Lago CI bot <dcaroest+cibot(a)redhat.com> - 0.15.1 5c5bc197: Merge pull request #226 from lago-project/fix_repoverify_check c4380a9a: Fixed repoverify not behaving like reposync d8c9ac55: Added new deps for the tests * Sat Apr 23 2016 Lago CI bot <dcaroest+cibot(a)redhat.com> - 0.15.0 4b98d690: Merge pull request #223 from lago-project/try_cleaning_reposync_caches_on_first_failure f87101ee: Reposync first retry with clean caches Enjoy! Gal Ben Haim gbenhaim(a)redhat.com RHEV-CI Team

2 1

Weekly bug status for ovirt-system-tests (using Lago) [ 01/05/16 ]
by Eyal Edri 01 May '16

01 May '16

FYI, Some updates from recent activity on ovirt-system-tests running on Lago. *Open bugs:* https://goo.gl/eUaPUY *Open issues:* Improvements on the host install issue, some env are not reproducing anymore but not sure we're 100% clear yet. *Open patches:* (reviews are welcome) https://gerrit.ovirt.org/#/q/project:ovirt-system-tests+status:open * There are quite a few open, some are just temp for helping with the debug of 'stuck in installing state' *Jenkins jobs: (failing on known issues)* http://jenkins.ovirt.org/view/oVirt%20system%20tests/ *Recently merged patches:* commit 211e475c004a2635896e528cd3aed6cd7d0dc5fd Author: Sandro Bonazzola <sbonazzo(a)redhat.com> Date: Thu Apr 28 16:44:50 2016 +0200 common: yum-repos: update glusterfs repos - dropped Gluster community repo for Fedora23: the repo is empty and Gluster community is maintaining Fedora packages within Fedora repositories. For reference: http://download.gluster.org/pub/gluster/glusterfs/LATEST/Fedora/What_About_… - on EL7 moved from Gluster community repo to CentOS Storage SIG repositories: we've been advised by Gluster community that their packaging effort is focused on CentOS Storage SIG and CentOS infrastructure is more reliable. Change-Id: I4d2c752be825fbbb9035222845728d10c5dd288c Signed-off-by: Sandro Bonazzola <sbonazzo(a)redhat.com> ---------------------------------------------------------------------------------------------------------- commit 19c102110279975d274c32477cc28c7c29e75fbc Author: David Caro <dcaroest(a)redhat.com> Date: Wed Apr 20 21:10:37 2016 +0200 Fixed answerfile for master A new question was added to the setup Change-Id: I6be958a70d67126d8067ff1855018125623b10c1 ---------------------------------------------------------------------------------------------------------- commit 5829862d17e92f91cfd04e7b9d5549cd775c800c Author: Yaniv Kaul <ykaul(a)redhat.com> Date: Mon Apr 25 14:06:27 2016 +0300 Add 'device=eth0' to EL7 deploy scripts. Apparently, /etc/sysconfig/network-scripts/ifcfg-eth0 lacks it on EL7 images. This works with NetworkManager, but not without it. When something goes bad, VDSM fails to rollback properly as its missing. For now, add it. Change-Id: I1c4ca07499a3947022e6f943cf05842ee1def408 Signed-off-by: Yaniv Kaul <ykaul(a)redhat.com> ---------------------------------------------------------------------------------------------------------- commit ddb42782968e5c6d411cdc747503fd90bed249d7 Author: David Caro <dcaroest(a)redhat.com> Date: Tue Mar 29 18:53:42 2016 +0200 Fix the master suite runs This is just copying over and adapting a bit (just so it works) the 3.6 suite into the master suite. Only for the 3.6 api, the newapi tests will be added later Still affected by the storage deactivation issues Change-Id: I55781a5d5a746aaf81e403eeeee40880beacb8a0 Signed-off-by: David Caro <dcaroest(a)redhat.com> ---------------------------------------------------------------------------------------------------------- commit c7d50cf8aebd6b566ae5c6ee8d6e3b8ec24ffc2b Author: Yaniv Kaul <ykaul(a)redhat.com> Date: Tue Apr 26 13:57:43 2016 +0300 Remove snapshot and reverting as part of the ovirt-system-tests I currently did not find a use case for it. It slows down the tests and causes (real? false?) issues with SDs. Mainly their activation and deactivation. It dramatically reduced the time to run the whole suite for me. 3-4 minutes on my desktop, and >10 minutes on my laptop. Change-Id: I9f7e509220c83805df78abb94b0ddc390c4a7f54 Signed-off-by: Yaniv Kaul <ykaul(a)redhat.com> ---------------------------------------------------------------------------------------------------------- -- Eyal Edri Associate Manager RHEV DevOps EMEA ENG Virtualization R&D Red Hat Israel phone: +972-9-7692018 irc: eedri (on #tlv #rhev-dev #rhev-integ)

1 0