mount local openshift epel mirror only during image builds, no final …

[gabemontero]

…image output

@stevekuznetsov FYI - per our chat at your cube, here is the vagrant-openshift prototype. Let me know how you want this reflected in the new origin-ci-tool component.

Also note, after this change, we need to change the Dockerfile's in the origin repo to install yum-plugin-priorities in a separate RUN step prior to installing epel-release, etc.

[gabemontero]

Yep .... I was experimenting trying to get it to work (prior to the install yum priority plugin before epel-release revelation) and then did not either switch it back or update the comment.

At this point I think it is better to leave at priority 1 to avoid unintended interference from other repo files and update the comment.

If you want to merge this change in, let me know and I'll update the comment.

On Mon, Feb 13, 2017 at 2:57 PM, Steve Kuznetsov notifications@github.com wrote:

@stevekuznetsov commented on this pull request.

In lib/vagrant-openshift/resources/local_epel.repo https://github.com/openshift/vagrant-openshift/pull/533#pullrequestreview-21599540 :

@@ -0,0 +1,24 @@ +[local-epel] +name=Extra Packages for Enterprise Linux 7 - $basearch +baseurl=https://mirror.openshift.com/mirror/epel/7/x86_64 +#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7&arch=$basearch +failovermethod=priority +enabled=1 +gpgcheck=1 +gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7 https://mirror.openshift.com/mirror/epel/7/x86_64+#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7&arch=$basearch+failovermethod=priority+enabled=1+gpgcheck=1+gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7 +# according to note from sdodson in https://github.com/openshift/origin/issues/8571, default priority is 99, +# so 98 should be higher priority, and override the default epel repo +priority=1

doesn't seem to match the comment above

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/openshift/vagrant-openshift/pull/533#pullrequestreview-21599540, or mute the thread https://github.com/notifications/unsubscribe-auth/ADbadLFDg-BuQE8uKD4mzFi2HE8nYQAiks5rcLWhgaJpZM4L_pSe .

[gabemontero]

@stevekuznetsov bump

how about we actually merge this and update ami's ? ... the base/dev ami jobs still seem to be using vagrant

[stevekuznetsov]

Those AMIs will remain in circulation for a while, so that makes sense. I don't have the bandwidth to do a full review of this right now though. If you're confident let's merge it but you will also need to own updating the plugin on our servers and making sure that change doesn't break anything.

[gabemontero]

On Mon, Feb 27, 2017 at 4:08 PM Steve Kuznetsov notifications@github.com wrote:

Those AMIs will remain in circulation for a while, so that makes sense. I don't have the bandwidth to do a full review of this right now though. If you're confident let's merge it but you will also need to own updating the plugin on our servers https://ci.openshift.redhat.com/jenkins/job/vagrant-openshift/ and making sure that change doesn't break anything.

sounds like a plan; I'll review that link you posted on Tuesday and assuming I get the gist of it we'll hit the merge button here and go from there

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/openshift/vagrant-openshift/pull/533#issuecomment-282853787, or mute the thread https://github.com/notifications/unsubscribe-auth/ADbadE9sQpsSYLEZQKAaRO9I_XMoUxoQks5rgztMgaJpZM4L_pSe .

[gabemontero]

@stevekuznetsov @danmcp:

taking @stevekuznetsov 's prior comment in this PR, in looking at https://ci.openshift.redhat.com/jenkins/job/vagrant-openshift/, and the message: "NOTE: Make sure, there is no other build in progress as reinstalling this plugin will make some vagrant commands unavailable for a moment!"

I'm assuming that includes PR testing .... if so, is the implication that you have to find a window "off hours" when no PRs are cooking to update the vagrant-openshift plugin?

please advise/confirm .. thanks

[stevekuznetsov]

@danmcp advice was given. we'll wait to put this in until the merge queue is not 20 deep

[gabemontero]

The merge queues appeared idle this AM. I just ran the job to update the plugin: https://ci.openshift.redhat.com/jenkins/job/vagrant-openshift/446/console

[stevekuznetsov]

Cool. We'll want to run a base_ami and make sure the correct repo is used for the image builds.

[gabemontero]

https://github.com/openshift/origin/pull/12979 needs to merge first.

On Sat, Mar 4, 2017 at 10:24 AM, Steve Kuznetsov notifications@github.com wrote:

Cool. We'll want to run a base_ami https://ci.openshift.redhat.com/jenkins/job/base_ami and make sure the correct repo is used for the image builds.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/openshift/vagrant-openshift/pull/533#issuecomment-284158080, or mute the thread https://github.com/notifications/unsubscribe-auth/ADbadDmb7dBe8tRNmaUQb1K0_SjYB6bgks5riYIfgaJpZM4L_pSe .

[stevekuznetsov]

Ah, right. I'll take a look.