CI-users

ci-users@lists.centos.org

1 participants
418 discussions

Unplanned Outage: Inbound connections to ci.centos.org Mon 11-Apr-2016 (14:10 UTC)
by Brian Stinson 11 Apr '16

11 Apr '16

Hi Folks, This morning we had an unplanned outage of the NGINX proxy inbound to ci.centos.org. We were able to reboot the machine, and add some resources to help us better handle the inbound traffic. Thanks to Fabian for doing that! We've seen this behavior once or twice before, where inbound http connections fail intermittently. During those periods and during our emergency maintenance, connections internally between the slaves and the jenkins master remained up and no jobs were interrupted. If you have any questions, please let us know here on-list or in #centos-devel on freenode. Cheers! -- Brian Stinson CentOS CI Infrastructure Team

1 0

Disk space consumption on artifacts.ci.centos.org
by Fabian Arrotin 07 Apr '16

07 Apr '16

So, while working on the LV for the exported dir for rdo-store, I noticed (and Zabbix complained a lot about this in the last weeks) that the artifacts LV is getting bigger and bigger. Actually, the rdo project is more and more using disk space : 760G rdo It would be good to do some clean-up there, and only keep what's need to be publicly available (behind http://artifacts.ci.centos.org/rdo/ | https://ci.centos.org/artifacts/rdo/ ) Happy to implement that at the server itself (through a cron job) as long as we agree on the criteria (so for example delete everything older than 30 days) OTOH, keep in mind that I see multiple .qcow2 images (for undercloud ?) so those would be also deleted. Or would you want to delete yourself the unneeded files on your side ? -- Fabian Arrotin The CentOS Project | http://www.centos.org gpg key: 56BEC54E | twitter: @arrfab

4 3

new NFS export for RDO-store
by Fabian Arrotin 04 Apr '16

04 Apr '16

Just to let the RDO people be aware that a specific nfs export has been made available for them on the artifacts node : artifacts.ci.centos.org:/srv/rdo-store (and will be accessible from n30.pufty.ci.centos.org) The underlying LV is 250GB (as discussed) but can be grown later online if needed. -- Fabian Arrotin The CentOS Project | http://www.centos.org gpg key: 56BEC54E | twitter: @arrfab

2 1

Unplanned Maintenance: Jenkins Safe Restart 30-Mar-2016 11h25 Eastern
by Brian Stinson 30 Mar '16

30 Mar '16

Hi Folks, We're working on an issue with our Gerrit trigger plugins, and we need to perform a saferestart of Jenkins to restore connection to gerrithub. You may notice the banner on ci.centos.org coming up soon (We're expecting this to happen around 11:25 AM Eastern time, 15:25 UTC). We attempted this last night, but there were a few long-running jobs that we didn't want to interrupt. During a saferestart, Jenkins will queue up any incoming jobs and resubmits them after the restart is finished. If you have any questions, please let us know here or in #centos-devel Cheers! -- Brian Stinson CentOS CI Infrastructure Team

1 1

CI+GitHub integration?
by James 29 Mar '16

29 Mar '16

I heard GitHub plugins now exist. Has anyone prepared some docs or a quick terminal session of the correct way to get this running? Once this works, my git tpush [1] script should also automatically work! Thanks, James [1] https://ttboj.wordpress.com/2016/02/16/introducing-git-tpush/

2 2

Update re: our Cloud plans
by Karanbir Singh 19 Mar '16

19 Mar '16

hi, We've been looking at bringing up a cloud infra inside of ci.centos.org so that projects are able to test in and against cloud workloads, without having to first deploy their own! As some people on this list might know, the RDO effort is looking at bringing up an RDOCloud in the coming months, this will have the capacity to run a large number of instances, and also offer a very comprehensive openstack feature set. Our intention is to collaborate with them, and essentially use their cloud instance as the cloud workload host for ci.centos.org - in return we would offer them baremetal instances on our side, for when and how they want to test openstack-ironic driven workloads. This is made possible largely due to the fact that the RDO Cloud is likely going to come up, hosted in the same Data Center as the main ci.centos.org infra is, and we would have 10gb ( or better ) interconnect with their setup. However, this effort is still in PoC stages. In the mean time, we have a portion of our ci.centos.org infra, that we are going to onramp as an RDO Cloud in the coming days. I hope to have this setup for testing by the 28th March. Also, the aim here is going to be to offer as-little-features as needed to deliver a minimal viable IaaS service level. ie, Nova + Keystone + Glance + horizon ( and any hard ascendancies from there ), backed with a ceph storage cluster. We will aim to setup the network in a way that it maps 1:1 with the baremetal machines network ( ie, there will be no tenant isolation etc ). Ofcourse, with the RDOCloud coming up down the road, the lack of extended features wont be a thing. I am hoping to deploy enough capacity to run ~ 500 VMs. Regards -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

1 0

Is there an easy way to run a scheduled Jenkins job with different parameters?
by Niels de Vos 17 Mar '16

17 Mar '16

Hi, I would like to run a Jenkins job on a regular schedule, and this job should run with different CentOS and Gluster versions. It is easy to write the job and pass the different environment. This works fine for manual execution with the Jenkins web interface Now, is there a trick or job-type in Jenkins that I can use to setup a schedule *and* pass different environment parameters for different times? The workaround would be to create multiple Jenkins jobs with different default parameters... But I do not think that is a very elegant solution. Thanks, Niels

3 7

Broken email notifications
by Pino Toscano 15 Mar '16

15 Mar '16

Hi, apparently I have failures in the email notifications of some of my jobs; for example see the bottom of the log: https://ci.centos.org/view/libguestfs/job/libguestfs/label=libguestfs-ci-de… Obviously the email is valid... Thanks, -- Pino Toscano

3 3

Pufty Chassis Outage This Morning (14-Mar-2016)
by Brian Stinson 14 Mar '16

14 Mar '16

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Folks, One of our Seamicro chassis had an outage this morning, and all but 5 of the compute nodes were powered off. We were able to reset the chassis and restore service to those nodes, but we'll only release them back into the active pools in small batches. There did not appear to be any active reservations in Duffy at the time, but the following projects/users were potentially affected: - - kbsingh - - bstinson - - libvirt - - libguestfs I've sent a set of logs to our vendor to see if they can reason about the root cause of this outage. If you have any questions, please let us know Cheers! - -- Brian Stinson CentOS CI Infrastructure Team -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJW5vQHAAoJEIMGvNKzCweM4EwQAIAxENqgIPZPa7PzNJUQKPvy l4ikTUDHgJKYtp47nfE3WOfYg8Qh7RCQKOV3nXwLk2HL6K2B9EJYzQVwrtOeXbRZ 6kkIqpLtVdGyBSIyMF/le8vGsWID2BbKGuCYygej5INc5tjGym+MkgXCzgJseXuF TeC64BILDZMOndHyV+vD9pjZCwuQSMK855xQ4Ia5elqMiQpkqxaBHZSxY4eFsFQP qAaaHNoxvb4KLlTHQgs7hmdGNX29rPxO2Pf220uEKpGOeAApnjKc6rTHkr9ctD++ wBHoJQsNa+mrAJsmDWLi3i2G1LQ2iWoVoFOpAK2o5jVrgoL9jVhnMTfOypKvEsvJ sWflzOohtKq7H/Jw8JlP+c3/vQFumYXVDwzZ/wpnk06u9pEMP6WM/4SL9Em/usBx Exz1eKh1/CYiHq09n7k/GX6wbH1QV2qbzhmGMznU56BjF1XZL1eGiRV13XC+Rtc4 F3qTegKtnigHyJ7/0bNsTHlIWHKFp/Neyq4usnd5zQRuSKU/ok8+DtpTecWVr3v4 v8jkc2Gc+/277Ydr0KH2rUGZV+tUP9CDLKnnV4lkeIcCN5kHc/J9jXPA29MtpzJY C0E/2bYha3OhL4U9RYso2gwUK582Xyq0WOECSX7yEcdt/369Kle5W46zDKh/c4NS hkOGJIO5GDQpjLD8JXwU =CnMG -----END PGP SIGNATURE-----

1 0

short SSID's
by Karanbir Singh 08 Mar '16

08 Mar '16

hi we've deployed a new version of duffy - the only change users should see from this is a shorter 8 char ssid, instead of the longer one. Existing sessions will still work, but all new ssid'd allocated from here will be the shorter format ones. regards -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

3 2

← Newer
1
...
34
35
36
37
38
39
40
41
42
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CI-users