fix (S3): truncate byte buffer when putting object#4214
fix (S3): truncate byte buffer when putting object#4214mikeprimm merged 1 commit intowebbukkit:v3.0from
Conversation
|
Thank you for providing this fix to the S3 output issue! I’ve tested it with my own setup, and it resolves the problem with the extra null characters breaking JSON files. For anyone interested, here's a live demo using this patch. If you’re unable to build the patched version yourself, I’ve also provided a compiled JAR from my Jenkins CI instance. Disclaimer: Always be cautious when downloading JAR files from third parties. I’ve tested this build on my own production server, but please verify and use at your own discretion. The build is provided solely for convenience. |
|
@MoNoLidThZ I was dogfooding this and encountered #4217 , which I resolved in #4218 |
|
Looks good - thanks |
possible duplicate/necrobump of #4129, resolves #4128
as mentioned in #4129, #4030 introduced a bug that uploads the entire buffer, resulting in null bytes being appended to all files.
this reverts that change, as well as changing to call the overloaded
RequestBody.fromString().before the fix:

after the fix:

access volume looks normal, tested on a fresh world against an initially empty minio bucket:
