Cover #3
Closed
Cover #3
Conversation
This file contains 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
…_examples Better examples
Minor tweaks to README.release.md
…_docs Better docs
Python 2.6 has been EOL since 29-10-2013
deprecate python 2.6
…_empty Fix transferer crashing on empty file download attempt
|
this doesn't actually fix the issue, I fixed it a few days later for real |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Closes Backblaze/B2_Command_Line_Tool#554
Description
The original intent of making these changes was to fix the bug, which is mentioned above. Using B2 CLI, users were facing error in B2 SDK - like
b2.exception.TruncatedOutput: only 40719702112 of 79957501184 bytes read ERROR: only 40719702112 of 79957501184 bytes read. And as it was discovered, the most probable reason - failing download of some chunks. I.e., in case of big file we will have a lot of chunks to download. And due to unability to get some of them - we'll finally get less data than required. And the original issues was reported with B2 SDK version, which had number of retries equal to1(means no retries).As the first step in making changes - last state of
masterin original repo was merged into working branch (of my fork https://github.com/dbelaventsev-reef/b2-sdk-python). So, the rest of the changes were added on the top of that.Tests
Automated test suite is passing. Additional tests were added - downloads with retries and download with failing retries. Test case for failing retries will make 4 fails in a row (having 5 retry numbers in the code). Raw simulator was extended to support those things.
Fix
This issue doesn't require any changes in the code (since the latest B2 SDK was extended to have max number of retries equal to
5, which is more than enough to fight rare failing cases). Test cases was extended to fail 4 times in a row (having 5 retry attempts in the code). So we should be covered with all things.Possible Improvement
The only thing, which could be a good candidate to extend this changeset is to stop download of a big file (parallel downloader) if we have first failed chunk. That will save credits for a user. Let me exlain.
If a big file was split into 100 parts and we can't get 10 of them - user will get corrupted file at the end AND he'll loose credits, required for download of that "useless set of bytes".
So we could just abandon the download at all after first fail (which was happen after 5 retries).