-
Notifications
You must be signed in to change notification settings - Fork 803
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
Jenkins test suites are disabled because fails #859
Comments
I don't think we should disable Jenkins, for (1) we are currently relying on appveyor for our CI efforts, and (2) we are incrementally bringing jenkins to par. We can look into each of these issues until it is on the same level/predictability as appveyor. For GAC failures, @mmitche our build scripts need admin permissions. Is this possible? |
There is one machine that has elevated permissions right now (windows-elevated is the label I think). It is possible to get around using the gac and altering global state? Matt From: Omar Tawfik [mailto:notifications@github.com] I don't think we should disable Jenkins, for (1) we are currently relying on appveyor for our CI efforts, and (2) we are incrementally bringing jenkins to par. We can look into each of these issues until it is on the same level/predictability as appveyor. For GAC failures, @mmitchehttps://github.com/mmitche our build scripts need admin permissions. Is this possible? — |
I'd like to have jenkins. You can have a single build job "WindowsRelease" , in master does compile only The gac is also a problem. It's not possible to run concurrent jobs with shared gac, so need to exists a limit |
We don’t run concurrent on Windows machines for other reasons (msbuild process reuse for one), so we’re at least good there. From: Enrico Sada [mailto:notifications@github.com] I'd like to have jenkins. You can have a single build job "WindowsRelease" , in master does compile only all,build_only The gac is also a problem. It's not possible to run concurrent jobs with shared gac, so need to exists a limit — |
@enricosada I agree it is annoying, but we cannot keep Jenkins running tests in a PR, as the job definitions are regenerated on the CI server, not pulled from every commit. I'm planning on doing work on the GAC items in the next period, so this should be resolved. @mmitche any pointers on using the elevated-permissions machine for now? |
Change the label in your netci.groovy file from “windows” to “windows-elevated” Matt From: Omar Tawfik [mailto:notifications@github.com] @enricosadahttps://github.com/enricosada I agree it is annoying, but we cannot keep Jenkins running tests in a PR, as the job definitions are regenerated on the CI server, not pulled from every commit. I'm planning on doing work on the GAC items in the next period, so this should be resolved. @mmitchehttps://github.com/mmitche any pointers on using the elevated-permissions machine for now? — |
Job definition can be the same. Just change cigroovy to pass different arguments to appveyorbuild script Inviato da Mobile On Fri, Jan 8, 2016 at 12:36 PM -0800, "Omar Tawfik" notifications@github.com wrote: @enricosada I agree it is annoying, but we cannot keep Jenkins running tests in a PR, as the job definitions are regenerated on the CI server, not pulled from every commit. I'm planning on doing work on the GAC items in the next period, so this should be resolved. @mmitche any pointers on using the elevated-permissions machine for now? Reply to this email directly or view it on GitHub: |
FYI We can bring up more machines eventually with elevated permissions eventually, but you’ll see reduced TP for now. The other projects generally do not use elevated permissions (mostly because there is a tendency for bad practices to creep in, especially xplat) From: Omar Tawfik [mailto:notifications@github.com] @enricosadahttps://github.com/enricosada I agree it is annoying, but we cannot keep Jenkins running tests in a PR, as the job definitions are regenerated on the CI server, not pulled from every commit. I'm planning on doing work on the GAC items in the next period, so this should be resolved. @mmitchehttps://github.com/mmitche any pointers on using the elevated-permissions machine for now? — |
@mmitche I agree. I've planned work in this area to remove this requirement in the future.
|
@otawfik-ms ah my bad, the jenkins doesnt read the netci.groovy each run? like appveyor.ylm? |
@enricosada No, since it's generating much more than just the PR jobs. |
well, it's the same, we can use a the also one job maybe is enogh, we create multiple jobs because there is a 1h timeout in appveyout, but that's not the case in jenkins, so maybe we can have only one |
ok like #860 now, a |
@enricosada I like the idea. We can go ahead with that. |
ok, updated #860 making the same as https://github.com/dotnet/cli/blob/master/netci.groovy only using |
@oamsath but i cannot regen build defs, but the pr is ok i think |
current Jenkins build fails. /cc @otawfik-ms @KevinRansom
it's ok, but because is merged, all new pull request fails, and that a bit annoying.
It's possible to:
appveyor-build.cmd all,build_only
)QASuite
example log: http://dotnet-ci.cloudapp.net/job/Microsoft_visualfsharp/job/QASuite_prtest/8/console
update.cmd release -ngen
fails:Failed to open registry key -- Administrator permissions are needed
Administrator permissions are needed to use the selected options
Failure adding assembly to the cache: Administrator permissions are needed
test fails because cannot find FSharp.Core 4.4.1 (
System.IO.FileNotFoundException: Could not load file or assembly 'FSharp.Core, Version=4.4.1.9055, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a'
)I think that's expected because it failed to install in gac
UnitTests
example log: http://dotnet-ci.cloudapp.net/job/Microsoft_visualfsharp/job/UnitTests_prtest/8/
the flacky test
Async.AwaitWaitHandle does not leak memory
( ref #811 )CambridgeTests
example log: http://dotnet-ci.cloudapp.net/job/Microsoft_visualfsharp/job/CambridgeTests_prtest/8/console
same problems of QASuite, the gac install fails, and build timeout ( maybe because some test wait because gac'ed versions doesnt exists )
The text was updated successfully, but these errors were encountered: