https://bugzilla.mindrot.org/show_bug.cgi?id=1844
Mark <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[email protected] --- Comment #11 from Mark <[email protected]> --- checking in to add a vote for this :) I am inclined to disagree with comment #10 referring to the unfinished IETF spec because the IETF of course favours rough consensus and running code... and openBSD implemented it as-is many years ago and Fedora / RHEL have shipped this patch for 6 years now. :-) [1] As a result there is lots of documentation out there referring to "-m" workaround. To now suggest updating the specification to introduce this new idea of senders and receivers "applying rules" and also muddying what the existing umask flag does... is not the solution to the use case mentioned in this ticket. Infact it's not the solution to any real world problem as far as I can tell. The use case is just for an sftp server to ignore incoming permissions. Not to reinterpret the sent umask, not to apply rules. Just to set its own unilaterally. The patch does it, it has been running in production on RHEL for a long time and it is documented widely online. At this point, merging it at this point seems to me like a bit of a trivial decision. [1] https://src.fedoraproject.org/rpms/openssh/blob/f22/f/openssh-6.7p1-sftp-force-permission.patch -- You are receiving this mail because: You are watching the assignee of the bug. _______________________________________________ openssh-bugs mailing list [email protected] https://lists.mindrot.org/mailman/listinfo/openssh-bugs
