Remove Quarkus artifacts before caching the repository on Azure #5520
Conversation
|
@gsmet If you want me to also check anything out on this, let me know. |
|
Sure, I can help out but dunno how far @gsmet got |
|
I need to revisit this, Azure doesn't support something I wanted to try. |
|
Cool, let me know if I can be of help :-) |
gsmet
force-pushed
the
azure-avoid-caching
branch
from
1e74426 to
a771f67
Compare
|
Should we close this one or is there another attempt planned? |
|
I did another experiment already 2 days ago but apparently without much success. Let me check the log. |
gsmet
force-pushed
the
azure-avoid-caching
branch
2 times, most recently
from
02cc911 to
52fe3b9
Compare
ci-templates/prepare-cache.yaml
Outdated
| @@ -0,0 +1,4 @@ | |||
| steps: | |||
| - script: find $(MAVEN_CACHE_FOLDER)io/quarkus/ -mindepth 1 -maxdepth 1 \( ! -name http -a ! -name gizmo -a ! -name security \) -exec rm -rf {} \; | |||
There was a problem hiding this comment.
Can you comment on this one a little?
There was a problem hiding this comment.
I don't see what is unclear ;). Yeah right, I will add a comment.
| @@ -32,7 +32,8 @@ stages: | |||
|
|
|||
| - task: CacheBeta@0 | |||
| inputs: | |||
| key: maven | bom/runtime/pom.xml #if we attempt to use all poms then when they get copied to target everything breaks. This should be good enough, it does not need to be perfect | |||
| # the number below is a cache version when we want to force a cache refresh | |||
| key: maven | "2" | bom/runtime/pom.xml #if we attempt to use all poms then when they get copied to target everything breaks. This should be good enough, it does not need to be perfect | |||
There was a problem hiding this comment.
Does this mean that we will need to increment the version number somehow?
There was a problem hiding this comment.
No, it's only because I really wanted to go from a clean cache and I didn't find another way to do it.
So basically, you just have to increment when you have buggy artifacts in your cache and you want to start from scratch.
There was a problem hiding this comment.
That's what I thought, but just wanted to be sure
I haven't found a proper way to reset the caches from the UI.
gsmet
force-pushed
the
azure-avoid-caching
branch
from
d459b1a to
da6b9af
Compare
|
I don't have any Azure CI experience, but these changes do make sense to me |
|
@geoand I think we should try to merge that one and see if it improves things. |
|
Perhaps we should wait for after |
|
Sure, we can wait until tomorrow afternoon. |
|
@gsmet time to merge? |
gsmet
changed the title
No description provided.