write_file: fsync() before renaming the file in atomic mode #30
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 was noticed when a btrfs went out of space while the NixOS activation script attempted to write the UID map file. It left an empty file on disk, destroying an integral piece of state which then caused the system to fail booting because the content was not valid JSON. I cannot verify whether this patch fixes that issue in particular because it is hard to reproduce. The previous behaviour is wrong in even just theory though as there is no guarantee that the content write and rename happen in the same transaction AFAICT. The code now explicitly waits for the content to have hit the disk before renaming. I've verified that the order of syscalls as observed by `strace` is correct.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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 was noticed when a btrfs went out of space while the NixOS activation
script attempted to write the UID map file. It left an empty file on disk,
destroying an integral piece of state which then caused the system to fail
booting because the content was not valid JSON.
I cannot verify whether this patch fixes that issue in particular because it is
hard to reproduce. The previous behaviour is wrong in even just theory though as
there is no guarantee that the content write and rename happen in the same
transaction AFAICT.
The code now explicitly waits for the content to have hit the disk before
renaming. I've verified that the order of syscalls as observed by
straceiscorrect.