Skip to content

Sql: db close() failing isn't fatal#6161

Merged
ckamm merged 1 commit into2.4from
dont_fatal_on_sqlite_close_fail
Nov 14, 2017
Merged

Sql: db close() failing isn't fatal#6161
ckamm merged 1 commit into2.4from
dont_fatal_on_sqlite_close_fail

Conversation

@ckamm
Copy link
Copy Markdown
Contributor

@ckamm ckamm commented Nov 14, 2017

The assert was made fatal when looking at asserts for #5429 #5518
without having a particular problem in mind. Recent reports weakly
suggest that this might lead to occasional crashes for users when
sqlite_close fails in ways that look ignorable.

The assert was made fatal when looking at asserts for #5429 #5518
without having a particular problem in mind. Recent reports weakly
suggest that this might lead to occasional crashes for users when
sqlite_close fails in ways that look ignorable.
@ckamm ckamm added this to the 2.4.0 milestone Nov 14, 2017
@ckamm ckamm self-assigned this Nov 14, 2017
@ckamm ckamm requested a review from ogoffart November 14, 2017 09:41
@ckamm ckamm merged commit 820ebf4 into 2.4 Nov 14, 2017
@ckamm ckamm deleted the dont_fatal_on_sqlite_close_fail branch November 14, 2017 11:10
@guruz
Copy link
Copy Markdown
Contributor

guruz commented Nov 15, 2017

@ckamm Do you think this fixes #6049 (comment) ?

@guruz
Copy link
Copy Markdown
Contributor

guruz commented Nov 15, 2017

...or maybe not, the ENFORCE is after the sqlite3 function

@ckamm
Copy link
Copy Markdown
Contributor Author

ckamm commented Nov 15, 2017

@guruz I had hoped so, but the sentry backtrace linked from #6049 (comment) suggests that it actually crashes in sqlite_close

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants