Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Do not upload files outside task sandbox #1777

Merged
merged 1 commit into from Apr 12, 2018
Merged

Do not upload files outside task sandbox #1777

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
6 changes: 6 additions & 0 deletions ...rc/main/java/com/hubspot/singularity/executor/task/SingularityExecutorTaskLogManager.java
View file
Open in desktop
Expand Up @@ -69,6 +69,12 @@ private boolean writeS3MetadataFileForRotatedFiles(boolean finished) {

for (SingularityS3UploaderFile additionalFile : taskDefinition.getExecutorData().getS3UploaderAdditionalFiles()) {
Path directory = additionalFile.getDirectory().isPresent() ? taskDefinition.getTaskDirectoryPath().resolve(additionalFile.getDirectory().get()) : taskDefinition.getTaskDirectoryPath();

if (!directory.toAbsolutePath().startsWith(taskDefinition.getTaskDirectoryPath())) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we want to && this with isCheckSubdirectories(). The main thing we're trying to protect against is the uploader going crazy checking all subdirectories of / or something like that. Debating if we should allow a way for folks to use this for things outside of the sandbox, within reason. @baconmania thoughts?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As far as I can tell the startsWith method of checking this should not start recursively checking directories so I don't think we have to worry about going crazy checking all subdirs of /.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's not the startsWith, that's just a string check, I'm worried about. It's the isCheckSubdirectories() value of the additionalFile. If it isn't checking subdirectories/not recursive, the most files the uploader will eventually scan is whatever is in the specified folder. If it is recursive, that's the problem we are trying to prevent. (e.g. isCheckSubdirectories() -> true with a folder of / essentially causes it to check every folder on the host

That said I could go either way on limiting this. It's basically a question of whether or not we expect a task to write to and want to upload something outside it's sandbox, and if it does we certainly need to impose a limit that the folder isn't checked recursively

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I could see potentially only doing this limitation based on isCheckSubdirectories() == true, but my vote here right now would be to default to limiting everything to the sandbox unless we see an ask where people want to be able to upload things outside the sandbox.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍 fair enough, let's leave the check as-is for now then

log.warn("Received request to upload files in directory outside task sandbox; these will not be uploaded ({})", directory);
continue;
}

String fileGlob = additionalFile.getFilename() != null && additionalFile.getFilename().contains("*") ? additionalFile.getFilename() : String.format("%s*.[gb]z*", additionalFile.getFilename());
result = result && writeS3MetadataFile(additionalFile.getS3UploaderFilenameHint().or(String.format("file%d", index)), directory, fileGlob, additionalFile.getS3UploaderBucket(), additionalFile.getS3UploaderKeyPattern(), finished,
additionalFile.getS3StorageClass().or(taskDefinition.getExecutorData().getS3StorageClass()), additionalFile.getApplyS3StorageClassAfterBytes().or(taskDefinition.getExecutorData().getApplyS3StorageClassAfterBytes()),
Expand Down