New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Minor fixes to make SLES work again #1390
Conversation
trevor-vaughan
commented
May 14, 2017
- Pull the updated RPM GPG key
- Remove rubygem-puppet prior to trying to install puppet-agent
- Tested with opensuse/openSUSE-42.2-x86_64
Refer to this link for build results (access rights to CI server needed): |
@puppetlabs/beaker Can someone clue me in as to why Jenkins is mad? All of the tests ran fine on my system. |
@trevor-vaughan here is the relevant error message from Jenkins, from the test check_for_package:
I am a bit surprised that the tests passed with your opensuse system; I would have thought that the you would have run into the same error the jenkins testing encountered above ^^. Looping in @shrug : the resource at yum.puppetlabs.com, could that be updated to mirror the nightlies resource? |
@tvpartytonight Actually, it makes complete sense. You're signing things with two different keys and you have to pick one. Jenkins wants your special sauce, not on the Internet, key but I changed it to import the Internet key so that it would work with the publicly published packages. https://github.com/puppetlabs/beaker/pull/1390/files#diff-bcaa88e697be9464be7e94c1c8a44fa7L35 |
* Pull the updated RPM GPG key * Remove rubygem-puppet prior to trying to install puppet-agent * Tested with opensuse/openSUSE-42.2-x86_64
Rebased against the current HEAD |
Refer to this link for build results (access rights to CI server needed): |
@shrug and @tvpartytonight any update on this? I'm not sure what the "correct" answer is for the keys. |
Refer to this link for build results (access rights to CI server needed): |
@shrug and @tvpartytonight Need insight into the new Jenkins failure. If you could get Jenkins to output the errors to GitHub, that would be most helpful. |
|
Refer to this link for build results (access rights to CI server needed): |
This has failed on a transient in our infrastructure. I've kicked the job, and opened QENG-5088 for tracking. |
PS: it seems like this has passed the jenkins now. cc @puppetlabs/beaker |
The failure above is due to a Windows testing failure, not related to the changes in this PR. @shrug , do you have any input here for the changes in this PR? |
@DavidS @tvpartytonight and @shrug I'm not sure what has been updated to conflict with foss_utils. I tried a quick rebase but it's a bit more complicated than I have time for right now. |
@DavidS @tvpartytonight and @shrug Do you have any ideas on what the issues are with the foss_utils files and how to fix them? |
@@ -367,6 +367,12 @@ def install_puppet_agent_on(hosts, opts = {}) | |||
end | |||
|
|||
if package_name | |||
# Some SLES images already have rubygem-puppet installed and this |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What SLES images are you dealing with that have rubygem-puppet installed? That sounds like that should be dealt with at an image level or within the pre-suite of your testing. I don't think that adding that here is the appropriate place for this logic.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@tvpartytonight I've been using the images directly from SuSE and OpenSuSE. If you take that out...you're gonna have a bad time....sometimes....
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@tvpartytonight What I'm getting at is that the Beaker code itself calls that before I even get to my pre-suite anything from what I remember and that's a bit beyond the level of Beaker mangling that I'd like to do.
@trevor-vaughan we have done work to remove some of beaker's libraries into independent gems, and so the files conflicting now exist in this separate repo: https://github.com/puppetlabs/beaker-puppet. You'll need update your PR to resolve those conflicts or reopen the PR with only the changes necessary for your sles fixes. Apologies about letting this PR get a bit stale. |
last comment is still current |
Last comment still current. |
Last comment still current. If no motion is made on this PR in another week, we’ll close this PR due to inactivity. |
@ferglor Last response is still current. Since Beaker significantly changed, I have not had time to figure out the new structure. The option for maintainers to commit is open so feel free to update as necessary if you want SLES support in Beaker. |
@@ -33,6 +33,9 @@ def check_for_package(name, opts = {}) | |||
# time. It's just to make sure that we only do the key import once, & | |||
# isn't for setting or use outside of beaker. | |||
execute('rpmkeys --import http://nightlies.puppetlabs.com/07BB6C57', opts) | |||
|
|||
# We also need the real key in case users are pulling from the stable packages | |||
execute('rpmkeys --import https://yum.puppetlabs.com/RPM-GPG-KEY-puppet', opts) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've recently started work to try to fix the sles repos
I've added #1449, which alone isn't going to fix this. But I'm also working on getting our repo automation updated so it signs the yum repos, and provides the gpg key in the repo. With that, zypper will automatically import the key in the repo if directed to do so. This opens us up to transients for network blips, but it's certainly better than what we have. However, this is also a perfectly reasonable solution to fix these failures. I'm wondering if this key is available under /etc/pki/rpm-gpg/RPM-GPG-KEY-puppet-PC1
. It would only be dropped there by the puppetlabs-release package. I'm not sure if that's getting installed or not
It looks like there have been a number of moves from code around this section in beaker, as well as work for keys & issues handled here in other PRs as well. Since that's the case & no movement has been made on this PR recently, I'll close this due to inactivity. If you'd like to re-open this PR, please update it so that it's in a merge- & test-able state for review. |