Bugzilla – Full Text Bug Listing
| Summary: | use fetchCacheLocation to reuse git clones across builds | ||
|---|---|---|---|
| Product: | [Eclipse Project] Platform | Reporter: | Kim Moir <kim.moir> |
| Component: | Releng | Assignee: | Platform-Releng-Inbox <platform-releng-inbox> |
| Status: | CLOSED WONTFIX | QA Contact: | |
| Severity: | normal | ||
| Priority: | P3 | CC: | aniefer, daniel_megert, david_williams, gunnar, pwebster |
| Version: | 3.8 | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Windows XP | ||
| Whiteboard: | stalebug | ||
|
Description
Kim Moir
I talked to Andrew and you should set a fetchCacheLocation for every type of build. (I,M,N) Apparently there is a bug in the EGit fetch factory that it doesn't switch to the correct branch sometimes if you are pulling multiple branches from the same repo. Okay I should have implemented this a long time ago. Some numbers Without fetchCacheLocation (times from start of build, no signing) build source zips: 34 minutes build tests: 1h 4minutes Finish build: 2h 30minutes Some numbers With fetchCacheLocation and git repos content primed(times from start of build, no signing) build source zips: 18 minutes build tests: 44 minutes Finish build: 1h 37minutes Total savings: 53 minutes Wow. Thanks Andrew for implementing this feature in the fetch factory! I've implemented this for the integration and nightly builds. I can't do this for 3.7.2 yet since we are in the endgame and changes require PMC approval etc. Not sure if bug 351384 is relevant. Was talking to Ian and he mentioned he had issues with the caching but this may just have been related to how he specified his branch. bug 347151 is the egit bug, it does a fetch but no merge (In reply to comment #4) > bug 347151 is the egit bug, it does a fetch but no merge Did the fetch factory already moved back to PDE? Gunnar, no the fetch factory hasn't moved to PDE. I don't think we have any committers to work on it. Andrew, from reading bug 347151, it's not clear to me that this causes an issue for our builds. I've configured our build so there are directories for the Git cache for every stream (M, N, I) but these directories are reused when that build runs. You seem to suggest in that bug that a new directory be used for the cache for every build which seems to negate the advantage of having a cache in the first place. This seems to be working. Closing. This changed caused problems in bug 369409. Thus I'm reverting it. upsetting 3.8 M6 target since this wasn't fixed in 3.8. This bug hasn't had any activity in quite some time. Maybe the problem got resolved, was a duplicate of something else, or became less pressing for some reason - or maybe it's still relevant but just hasn't been looked at yet. If you have further information on the current state of the bug, please add it. The information can be, for example, that the problem still occurs, that you still want the feature, that more information is needed, or that the bug is (for whatever reason) no longer relevant. If the bug is still relevant, please remove the stalebug whiteboard tag. Infra is moving to containerized builds so the idea of reusing git clones is a no-go. |