-
Notifications
You must be signed in to change notification settings - Fork 2.7k
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
[3.14] 3.14.0 backports 1 #42664
[3.14] 3.14.0 backports 1 #42664
Conversation
(cherry picked from commit cc7bc0b)
Some manifests are quite large, for instance the one for commons-codec is 21 KB. Given most of the information in the Manifest are of no interest to us, let's trim the Manifest to only the information we need. Also we still have the problem of us keeping the jars open multiple times in test and dev mode so it amplifies this issue. In some cases, I ended up with 1700+ jars opened so the Manifests add up quite quickly. (cherry picked from commit a1b3fa9)
A Path can actually contain references to the filesystem: this is the case of the ZipPath. We want to make sure we don't keep ZipFileSystem around when a path tree is closed so we also need to nullify the path. Finally, we make sure only DirectoryPathTree is marked as Serializable as we only serialize the workspace directories. (cherry picked from commit 6780c2c)
(cherry picked from commit a17b30a)
(cherry picked from commit 43a3ccb)
We need to catch the NoSuchFileException and ignore it. Per gripe from Marco Bungart in the Quarkus Artemis project. (cherry picked from commit 2c87bb5)
(cherry picked from commit 429e15f)
When the Redis cache is invoked from an ordered Vert.x blocking execution, which happens for example with SmallRye GraphQL or with chained caching (one blocking `@CacheResult` method invoking other blocking `@CacheResult` method), the Redis cache ends up hanging. This is because the next execution cannot start until the previous execution finishes, but the previous execution waits for the next execution. This commit fixes that by making the Redis cache blocking executions unordered. (cherry picked from commit 1fd2a9a)
When dealing with test archives, such as in the Platform, we shouldn't try to create a temp directory in the archive. Related to the failure described here: quarkusio/quarkus-platform#1255 (comment) (cherry picked from commit ff11475)
Fixes quarkusio#42591 (cherry picked from commit 51834c5)
The header should be: Authorization: Bearer <token> and not Authorization: Bearer: <token> (cherry picked from commit cd12e4c)
The container path is the path of the container (could be a directory or an archive and shouldn't be used for browsing. The root path is the path of the root. In the case of an archive, it's actually a ZipPath, which can be used to browse the archive. Fixes quarkusio#42571 (cherry picked from commit 6dd6150)
…revention.adoc (cherry picked from commit 0bb5206)
Details: quarkusio#41962 (comment) (cherry picked from commit 6ce7dee)
(cherry picked from commit 0c565d4)
Related to quarkusio/quarkus-quickstarts#1442 (cherry picked from commit d502e8c)
Fixes quarkusio#42614 (cherry picked from commit f399671)
(cherry picked from commit 35389fd)
(cherry picked from commit ca2c06e)
They came from the string -> qute conversion. (cherry picked from commit 93f02be)
We used to cache a repo once a week but only after building the Java 17 build. The new approach is the following: - We use dependency:go-offline to populate the cache, hoping the quality will be better - We drop the cache entirely every month - We have one cache entry per branch and we keep updating it during the month by creating new caches - We use restore-keys to find the best cache possible for a given run - We share this cache with all the workflows (cherry picked from commit b3756d8)
This workflow wasn't populating the cache with the right key as we haven't changed it when we switched to a weekly-refreshed cache. Let's drop it. (cherry picked from commit 4e99832)
For forks, let's use a 'fork' suffix common to all branches as the push events will target feature branches. (cherry picked from commit 5b4188b)
A daily cache is enough. Otherwise we will have a lot of entries. (cherry picked from commit 7762ac5)
… doc (cherry picked from commit 57e0a1f)
(cherry picked from commit 778834e)
See quarkusio#42367 (cherry picked from commit 8b1f641)
When creating projects or extensions, the codestarts from the tooling version are always used even for older versions so we need to be careful about it. Fixes quarkusio#42649 (cherry picked from commit d6123f2)
Status for workflow
|
🙈 The PR is closed and the preview is expired. |
Status for workflow
|
Please don't merge, I will merge it myself.