@pmatilai commented on this pull request.
> }
- if (rpmMkdirs(rpmtxnRootDir(txn), "%{_keyringpath}")) {
+ free(dir);
+ return rc;
+}
+
+/*****************************************************************************/
+
+static rpmRC acquire_write_lock(rpmtxn txn)
...ie it's not the existence of 'writelock' file that is the lock, but a
flock() on that file. Mere file existence would be broken lock mechanism
because if the process dies in the middle it de-facto leaves a stale lock
behind, whereas locks on the file are cleaned up by the OS.
--
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/3437#discussion_r1836593518
You are receiving this because you are subscribed to this thread.
Message ID: <rpm-software-management/rpm/pull/3437/review/[email protected]>_______________________________________________
Rpm-maint mailing list
[email protected]
http://lists.rpm.org/mailman/listinfo/rpm-maint
