<p dir="ltr">My machine is 64bit (VMware guest). I can do any test/check on it when the problem appears, since it's a development machine and no critical job runs on it. If you have an idea.<br></p>
<p dir="ltr">-- <br>
Stefano</p>
<div class="gmail_quote">Il 15/ott/2015 19:48, "Michael Biebl" <<a href="mailto:biebl@debian.org">biebl@debian.org</a>> ha scritto:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Am 15.10.2015 um 19:15 schrieb James Bottomley:<br>
> On Thu, 2015-10-15 at 07:35 +0200, Stefano wrote:<br>
>> Same here. I have a git server with clients connecting every 10 minutes or<br>
>> so via SSH, and the problem is exactly the same.<br>
>><br>
>> About 20 hours from reboot logind goes nuts, then it must be restarted at<br>
>> very variable intervals, about a couple a day.<br>
><br>
> What architecture is this? I only see the problems on a 32 bit x86<br>
> system. I don't see them on a 64 bit x86 in spite of trying to trigger<br>
> them there (in a sample size of 1 of each, so it's not statistically<br>
> significant ... yet).<br>
<br>
Thanks, that is an interesting data point.<br>
<br>
> I also see that /run/systemd/sessions has about a hundred or so sessions<br>
> whose leaders are dead on the 32 bit system. On the 64 bit system it<br>
> appears only still living sessions are in here.<br>
><br>
> The final curious observation is that on the 32 bit system, the session<br>
> id never seems to go over 4k. On the 64 bit system, it does.<br>
<br>
Btw, in both cases (#770135 and #793814) the users have been restarting<br>
the dbus daemon. You mentioned that you did not do that.<br>
<br>
So I'm not sure if it's actually the same issue after all and your<br>
problem should be tracked as a separate issue.<br>
<br>
<br>
<br>
--<br>
Why is it that all of the instruments seeking intelligent life in the<br>
universe are pointed away from Earth?<br>
<br>
</blockquote></div>