On 2015-08-31 at 20:37, Jonathan de Boyne Pollard wrote:
> The Wanderer:
>
>> No, but I believe I still have my laptop configured in a way which
>> gets this behavior. If you want, I can reboot it and do a detailed
>> examination; I'm probably about due for a reboot of that laptop,
>> anyway.
>
> It's mainly the final part about other people's login sessions
> causing messages on unrelated terminals that is perplexing.
Actually, that part is speculative on my part (and I tried to label it
that way, by saying "if I'm interpreting the situation correctly, [this
will probably happen]"); my understanding of the mechanism by which
logind dumps those messages into the console in the first place
indicates that it will dump them into any and every active text console,
so that whoever is logged in gets the notification that a new session
(or maybe that's "seat"? I'm not clear on the official terminology) has
activated.
Part of the "detailed examination" (which I do think I want to do at
some point, regardless) would be to verify, if I can, whether or not
this does happen. If it does not, then that's my bad.
> The only mechanisms that immediately spring to mind for that are ones
> like {r,}syslog configured to take logind messages from the journal
> and write them to the console, splatting over whatever kernel virtual
> terminal happens to be active, which presumably you would have
> already thought of and checked.
I believe that's roughly how it works, yes - and I believe rsyslog is
intentionally set up that way, so that various system messages which
would appear in the active console if the journal were not present will
still appear there. It's just that now there are _more_ messages, which
would not have existed in the absence of
systemd-the-collection-of-binaries-which-orbit-the-PID1-binary. (These
unambiguous names get kind of unwieldy...)
If there's a way to disable doing this for just those messages, without
disabling it for the other messages (which I do want and expect to
appear in the console, at least in some cases), I haven't found it.
--
The Wanderer, hoping he hasn't flubbed something after the long day
The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man. -- George Bernard Shaw
Attachment:
signature.asc
Description: OpenPGP digital signature