On 2024-12-03 21:44:03 +1100, Craig Small wrote: > On Tue, 3 Dec 2024 at 21:21, Vincent Lefevre <vinc...@vinc17.net> wrote: > > > This particular one could be a bug in the kernel. But the access time > > (or modification time) is correct here with /dev/pts/* devices, though > > > They're handled differently so if the tty kernel code could still be > the issue. Also one user session can have many /dev/pts/* devices > and its not clear how to link a user to those devices reliably.
wtmpdb could be a solution as it is designed to replace wtmp/utmp. Note that "w" currently misses many tty devices. > > "w" gives an incorrect idle time. Another issue is that "w" does not > > output TTY information for me (that's bug 1080335). > > > That's a systemd issue that is impacting w. So this one should probably be reassigned to systemd. But note that this is a regression: there are no issues with Debian/stable. -- Vincent Lefèvre <vinc...@vinc17.net> - Web: <https://www.vinc17.net/> 100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/> Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)
