Bug#798629: procps: w and who do not view any logged in users

[Martin Steigerwald]

Am Montag, 19. Dezember 2016, 00:12:38 CET schrieb Michael Biebl:
> Control: tags -1 + moreinfo unreproducible
> 
> On Sat, 12 Sep 2015 11:14:32 +0200 Martin Steigerwald
> 
> <martin at lichtvoll.de> wrote:
> > Am Samstag, 12. September 2015, 08:09:26 CEST schrieb Craig Small:
> > > On Fri, Sep 11, 2015 at 10:55:47AM +0200, Martin Steigerwald wrote:
> > > > I find that on my current system w and who are not displaying any 
users:
> > > [...]
> > > 
> > > > Today there was a systemd upgrade to 226-1. I am not sure whether it
> > > > is
> > > > related.
> > > > 
> > > > I have no idea what package to report this against, as it does not
> > > > appear
> > > > to be either procps (for w) or coreutils (who). Feel free to reassign
> > > > as
> > > > you see fit.
> > > 
> > > Yeah, I suspect its neither but both are victims of whatever is (not)
> > > updating /var/run/utmp
> > > 
> > > Run utmpdump /var/run/utmp and see if you have entries. If you do it's
> > > more likely w and who, I suspect there won't be much there and its a
> > > systemd-login problem not updating that file correctly.
> > 
> > Craig, it seems the utmp file is basically empty except the reboot and the
> > logged out already tty1 session.
> > 
> > merkaba:~> LANG=C utmpdump /var/run/utmp
> > Utmp dump of /var/run/utmp
> > [2] [00000] [~~  ] [reboot  ] [~           ] [4.2.0-tp520-btrfstrim+]
> > [0.0.0.0        ] [Sat Sep 12 11:05:25 2015    ] [1] [00053] [~~  ]
> > [runlevel] [~           ] [4.2.0-tp520-btrfstrim+] [0.0.0.0        ] [Sat
> > Sep 12 11:05:28 2015    ] [6] [01704] [tty1] [LOGIN   ] [tty1        ] [ 
> >                   ] [0.0.0.0        ] [Sat Sep 12 11:05:28 2015    ]
> > merkaba:~> w
> > 
> >  11:12:29 up 5 min,  0 users,  load average: 0,30, 0,67, 0,36
> > 
> > USER     TTY      FROM             LOGIN@   IDLE   JCPU   PCPU WHAT
> > merkaba:~> who
> > merkaba:~>
> > 
> > I suspected systemd-logind already, but I was not sure about it. Feel free
> > to reassign as you see fit.
> 
> Can you still reproduce the issue?
> Seems w and who is working fine here with v232 on an up-to-date system.

Appears to work here as well. Feel free to close.

Thanks,
-- 
Martin