kilt-node
kilt-node copied to clipboard
ci: caching
fixes KILTProtocol/ticket#3345
Breaks up the on gitlab-ci.yml & adds one more stage release.
- [x] Gitlab
needsto speed up pipeline - [x] Reuses artifacts from previous jobs
- [x] Parallel command execution
Metadata Diff to Develop Branch
Peregrine Diff
Spiritnet Diff
Checklist:
- [ ] I have verified that the code works
- [ ] No panics! (checked arithmetic ops, no indexing
array[3]useget(3), ...)
- [ ] No panics! (checked arithmetic ops, no indexing
- [ ] I have verified that the code is easy to understand
- [ ] If not, I have left a well-balanced amount of inline comments
- [ ] I have left the code in a better state
- [ ] I have documented the changes (where applicable)
- Either PR or Ticket to update the Docs
- Link the PR/Ticket here
I'd like to get a bit more context into the different pieces of this PR, especially regarding the new
try-runtimefile.
I'd like to get a bit more context into the different pieces of this PR, especially regarding the new
try-runtimefile.
Tested self hosted github action yet It takes way longer than gitlab. Check here
Hey @ggera I'll unsubscribe from this issue to reduce notifications. Please re-request a review when all the changes are in!
A few more questions that would help me understand the setup and the assumptions you took even better.
Also, in general, how will the Gitlab and GitHub jobs interact with each other? I did not see any logic that would trigger one based on the result of the other?
There is no direct dependency between GitLab and GitHub jobs. We have a pull mirror setup, which usually experiences a delay of at least 5 minutes due to its regular interval pull synchronization.If needed we can introduce instant webhook synch in another ticket
There is no direct dependency between GitLab and GitHub jobs. We have a pull mirror setup, which usually experiences a delay of at least 5 minutes due to its regular interval pull synchronization.If needed we can introduce instant webhook synch in another ticket
So how does one platform know when to run the tests? What if they fail on one platform? Is the other notified? E.g. we don't want to build a Docker image if we know the tests do not go through, or don't want to run try-runtime tests if we fail to run the unit tests.