HADOOP-16233. S3AFileStatus to declare that isEncrypted() is always true #685
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
Contributor
Author
|
tested: S3A ireland. All well except
|
Contributor
|
Seems to me that this means of determining whether something is public or private is brittle since it is derived by assumptions about certain filesystems. I'd like to see a follow up JIRA for addressing this in a more explicit way to consider something public. Here is my +1 for the change to align S3A filesystem with those assumptions. |
|
🎊 +1 overall
This message was automatically generated. |
steveloughran
added
the
fs/s3
asfgit
pushed a commit
that referenced
this pull request
Apr 3, 2019
…rue (#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
asfgit
pushed a commit
that referenced
this pull request
Apr 3, 2019
…rue (#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
asfgit
pushed a commit
that referenced
this pull request
Apr 3, 2019
…rue (#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3 (cherry picked from commit 366186d)
smengcl
pushed a commit
to smengcl/hadoop
that referenced
this pull request
Oct 8, 2019
…rue (apache#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
smengcl
pushed a commit
to smengcl/hadoop
that referenced
this pull request
Oct 8, 2019
…rue (apache#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
smengcl
pushed a commit
to smengcl/hadoop
that referenced
this pull request
Oct 8, 2019
…rue (apache#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
shanthoosh
pushed a commit
to shanthoosh/hadoop
that referenced
this pull request
Oct 15, 2019
Author: Hai Lu <[email protected]> Reviewers: Jagadish<[email protected]> Closes apache#685 from lhaiesp/master
bentito
pushed a commit
to bentito/hadoop
that referenced
this pull request
Dec 2, 2020
…rue (apache#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
bentito
pushed a commit
to bentito/hadoop
that referenced
this pull request
Dec 3, 2020
…rue (apache#685) This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start. Production code changes: * S3AFileStatus Adds "true" to the superclass's encrypted flag during construction. Tests * Base AbstractContractOpenTest can control whether zero byte files created in tests are encrypted. Not done via an XML attribute, just a subclass point. Thoughts? * Verify that the filecache considers paths to not have the permissions which trigger reduce-privilege downloads * And extend ITestDelegatedMRJob to test a completely different bucket (open street map), to verify that cached resources do get their tokens picked up Docs: * Advise FS developers to say all files are encrypted. It's otherwise harmless and it'll stop other people seeing impossible to debug error messages on app launch. Contributed by Steve Loughran. Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is needed to fix up some confusion about caching of job.addCache() handling of S3A paths; all parent dirs -the files are downloaded by the NM without using the DTs of the user submitting the job. This means that when you submit jobs to an EC2 cluster with lower IAM permissions than the user, cached resources don't get downloaded and the job doesn't start.
Production code changes:
Tests
Docs:
Contributed by Steve Loughran.
Change-Id: Ifaae4c9d735ccc5eafeebd2584b65daf2d4e5da3