Bitmessage/PyBitmessage-2025-02-16
Archived
2
0
Fork 0
Code Issues 167 Pull Requests 54 Actions Packages Projects Releases 5 Wiki Activity

Make writing keys.dat safer by creating a backup file #715

Merged
yurivict merged 3 commits from some_small_changes into master 2014-12-26 03:09:17 +01:00
Conversation 2 Commits 3 Files Changed 8 +50 -69
yurivict commented 2014-09-15 09:01:08 +02:00 (Migrated from github.com)
Copy Link

Just overwriting the file may cause its loss in case the overall disk write operation fails in the middle.
Created a .bak file for the duration of the write operation to prevent such possible file loss.

Just overwriting the file may cause its loss in case the overall disk write operation fails in the middle. Created a .bak file for the duration of the write operation to prevent such possible file loss.
ghost commented 2014-09-15 18:45:43 +02:00 (Migrated from github.com)
Copy Link

Your changes nicely show how much unnecessary repeated code is left in Bitmessage.

Your changes nicely show how much unnecessary repeated code is left in Bitmessage.
yurivict commented 2014-09-16 19:09:40 +02:00 (Migrated from github.com)
Copy Link

Yes, a lot of code duplication.

keys.dat contains keys for your accounts, that's why it should be very well protected. I also added timestamp on the backup file now because otherwise it can be overwritten too.

I just noticed another bad problem that errors from shared.writeKeysFile() not propagate to the user level and are silently ignored. I think error propagation is broken in bitmessage in general. Any important error during execution (like this one) should be shown to the user, and now they are only printed in the stdout or stderr where nobody is looking.

This PR is good anyways, just need further improvement.

Yes, a lot of code duplication. keys.dat contains keys for your accounts, that's why it should be very well protected. I also added timestamp on the backup file now because otherwise it can be overwritten too. I just noticed another bad problem that errors from shared.writeKeysFile() not propagate to the user level and are silently ignored. I think error propagation is broken in bitmessage in general. Any important error during execution (like this one) should be shown to the user, and now they are only printed in the stdout or stderr where nobody is looking. This PR is good anyways, just need further improvement.
This repo is archived. You cannot comment on pull requests.
Reviewers
No Reviewers
Labels
Clear labels
bug
Something isn't working as it's supposed to
build
Related to building or build system
dependencies
Pull requests that update a dependency file
developers
The issue is relevant mainly for developers rather than users
documentation
duplicate
enhancement
New feature
formatting
Code or UI text formatting
invalid
legal
mobile
Frontend for mobile devices (kivy)
obsolete
packaging
performance
protocol
Protocol change
question
refactoring
regression
Some issue, fixed in the past, appears again
security
test
translation
usability
wontfix
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
PeterSurda
No Assignees
1 Participants
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Bitmessage/PyBitmessage-2025-02-16#715
No description provided.
Delete Branch "some_small_changes"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?