general problem -> Re: Fedora 16 and the 3.3.1-3 kernel

[Bill Davidsen]

Reindl Harald wrote:
>
> Am 23.04.2012 01:02, schrieb Bill Davidsen:
>>> udevd[187]: timeout '/sbin/blkid -o udev -p /dev/sda3'
>>> udevd[187]: timeout: killing '/sbin/blkid -o udev -p /dev/sda3' [332]
>>> udevd[187]: timeout: killing '/sbin/blkid -o udev -p /dev/sda3' [332]
>>> udevd[187]: '/sbin/blkid -o udev -p /dev/sda3' [332] terminated by signal 9
>>> (Killed)
>>> udevd[185]: timeout '/sbin/blkid -o udev -p /dev/dm-0'
>>> udevd[185]: timeout: killing '/sbin/blkid -o udev -p /dev/dm-0' [385]
>>> udevd[185]: '/sbin/blkid -o udev -p /dev/dm-0' [385] terminated by signal 9
>>> (Killed)
>>>
>>> I can say that the previous kernels boot just fine (3.3.0-8) and I have no
>>> trouble with them. Any ideas on what I can look at? Maybe file a bug report?
>>>
>> I would try the current kernel first, 3.3.1-5, although I did not have any such problem as you report with 3.3.1-3,
>> x86_64.
>
> no 3.3.1 is the current one, 3.3.2 is current
>
If you say so... my kernel is 3.3.1-5 and doing a software update assures me 
that "All software is up to date" so I have no idea where you get new kernels, 
unless you are running from testing or rawhide. In any case, a quick look at the 
subject suggests that the O.P. isn't even running updated production software 
versions.

> but the whole 3.3 line until now elaborates on the same problem
> not at each boot, mostly only slower but terrible at all
>
> https://bugzilla.redhat.com/show_bug.cgi?id=806548
> https://bugzilla.redhat.com/show_bug.cgi?id=805317
>
> i post a copy to the devel-list
> maybe someone cosiders "CONFIG_RCU_FAST_NO_HZ=n" because
> this is the change in 3.3 and seems to be the root cause
> of all this troubles
>
> my first bug report is now more than a month old and
> there are built one kernel after the next with the
> same problem
>
> -------- Original-Nachricht --------
> Betreff: Re: [systemd-devel] systemd / udev / kernel 3.3 / vmware
> Datum: Sun, 22 Apr 2012 13:01:04 +0200
> Von: Lennart Poettering<lennart at poettering.net>
> Organisation: Red Hat, Inc.
> An: Reindl Harald<h.reindl at thelounge.net>
> Kopie (CC): Mailing-List systemd<systemd-devel at lists.freedesktop.org>
>
> On Sat, 21.04.12 13:25, Reindl Harald (h.reindl at thelounge.net) wrote:
>
>> can someone please take a look at this F16 report?
>> https://bugzilla.redhat.com/show_bug.cgi?id=806548
>>
>> * boot sometimes terrible slow
>> * randomly boot fails with udev crashing / timeout
>> * shutdown terrible slow
>> * partly also on physical machines but not sooo slow
>>
>> since most of the boot time is consumed by udev and
>> the terrible slow shutdown of>  30 seconds happens
>> while unmounting "old..." i guess this is not only
>> the kernel
>>
>> feels like interaction between kernel/udev/systemd
>>
>> on shutdown / reboot i can randomly see each single
>> unmounting line for 1-2 seconds, this is odd
>>
>> i have virtual machines which booted in 5-6 seconds
>> and are now degraded up to 1 minute
>
> The bug report suggests this was about a kernel feature
> (CONFIG_RCU_FAST_NO_HZ=y) and not related to systemd, no?
>


-- 
Bill Davidsen <davidsen at tmr.com>
   "We have more to fear from the bungling of the incompetent than from
the machinations of the wicked."  - from Slashdot