init: fix S3 ARN parsing #617 #619
Conversation
init/fluent_bit_init_process.go
Outdated
| bucketAndFile := strings.SplitN(arnBucketFile, "/", 2) | ||
| s3ARN, err := arn.Parse(userInput) | ||
| if err != nil { | ||
| logrus.Fatalf("[FluentBit Init Process] Unrecognizable arn: %s\n", userInput) |
There was a problem hiding this comment.
Should it indicate that this is the s3 arn, similar to the task arn failure?
There was a problem hiding this comment.
It'll print the arn tho, so the user will see its the S3 arn they passed in?
There was a problem hiding this comment.
I think I'll change "unrecognizable" to "failure to parse" or something like that
|
I need to test this with the init integ test before we can merge and release it. |
Signed-off-by: Wesley Pettit <[email protected]>
PettitWesley
force-pushed
the
init-arn-parse-fix
branch
from
7a0873a to
87093f2
Compare
Have tested it on my previously broken deployment in GovCloud and can confirm this is working as expected now 👍 |
|
So I am struggling to test this myself due issues that I don't fully understand with working in gov cloud. I ran the integ test we have with an S3 bucket in the main partition, it worked, so no regression. I then tried running the init with gov cloud credentials and I'm getting: It got my bucket name and key correct so the parsing definitely does work. |
I noticed that a lot of the tests have the region hardcoded to I also noticed that the tests are using legacy Other than that, I was having trouble with it expecting various values which I haven't seen documented, like S3_BUCKET_NAME, KINESIS_STREAM, and probably something around S3 actions (causing errors of: That said; i'm actively using this in GovCloud right now, so i'm happy that the changes are working anyway :D |
Issue #, if available:
Fixes #617
Description of changes:
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.