Control: reopen -1
On Wed, 18 Jul 2018 16:38:21 +0200 Michael Biebl wrote: > > Yes, I confirm that issuing [...] > > # shutdown 5 [...] > > or [...] > > # shutdown [...] > > failed to show the broadcast message to the users. > > I tested this on a stable system, for a locally logged in user and one > logged in via SSH. > Both got the wall message > > Broadcast message from root@debian on tty1 (Wed 2018-07-18 16:37:14 CEST): > > The system is going down for poweroff at Wed 2018-07-18 16:38:14 CEST! > > > I thus conclude that this issue has been fixed in the mean time and will > close the bug accordingly. This issue has only partially been fixed. I can confirm that the wall message is now sent, when using "shutdown -r" (which has the implicit delay of 1 min). However, when using "reboot", "poweroff" (and "halt", I suppose), no wall message is sent. As I have already said in message [#17], when the "reboot", "halt", and "poweroff" commands are provided by package sysvinit-core, they do not take a delay time argument either, but they send a broadcast message to all logged users to inform them that the system is going down. I have always seen such message and I think it's really useful to understand why your SSH connection was closed. On the other hand, when the "reboot", "halt", and "poweroff" commands are provided by package systemd-sysv, they fail to send such broadcast message. Please note that the halt(8) man page shipped by package systemd-sysv even mentions a --no-wall option to suppress the broadcast message! Hence, I am under the impression that this broadcast message is supposed to be sent, but fails to be actually sent due to a bug. Although I agree with Martin Pitt that, on multi-user systems, it's polite to do something like "shutdown -r 5", I believe that the immediate commands ("reboot", "halt", and "poweroff") exist for all the situations where it's reasonable or needed to shut the system down as soon as possible. I think that informing users of what's happening should be done in any case. [#17]: <https://bugs.debian.org/810608#17> I am therefore reopening the bug report. -- http://www.inventati.org/frx/ There's not a second to spare! To the laboratory! ..................................................... Francesco Poli . GnuPG key fpr == CA01 1147 9CD2 EFDF FB82 3925 3E1C 27E1 1F69 BFFE
pgpasKA1_4Jto.pgp
Description: PGP signature