Re: Systemd debugging
On 2014-11-27 14:59 +0100, Haines Brown wrote:
> On Thu, Nov 27, 2014 at 12:06:20AM +0200, Andrei POPESCU wrote:
>
>> You can use this instead
>>
>> journalctl -alb
>
> Andrei, I did as you suggested, but I'm not sure if the result really
> adds anything substantive.
It is actually quite useful.
> Then the failed to insert module problem. I find that in fact I do have
> /lib/modules/3.12.0-4-686-pae/kernel/drivers/hwmon/it87.ko on this box.
> When it87 not inserted, were all further module insertions terminated?
I don't think so.
> If .../kernel/drivers/block/loop.ko was loaded, why not it87.ko?
>
> Nov 26 17:15:16 engels systemd-fsck[195]: /dev/sda1: clean,
> 15827/61056 files, 120381/243968 blocks
> Nov 26 17:15:16 engels systemd-modules-load[201]: Inserted module 'loop'
> Nov 26 17:15:16 engels kernel: loop: module loaded
> Nov 26 17:15:16 engels systemd-modules-load[201]: Inserted module
> 'coretemp'
> Nov 26 17:15:16 engels systemd-modules-load[201]: Failed to insert
> 'it87': No such device
I'm not familiar with the it87 module, but do you actually have the
hardware which it controls? If not, then it's not too surprising that
trying to load it fails with ENODEV.
> Then there's the report about not being able to mount /usb although in
> fact it is mounted and accessible. Was it initially seen as ro? Does
> this mean that the remount command succeeded to mount /usr on second
> try without ro?
No, it probably means that /usr was not mounted in the first place;
systemd-remount-fs expects that /usr is mounted by the initramfs which
is not the case for initramfs-tools in jessie. This is rather harmless,
though.
> Then there's something wrong when Console Kit and Policy Kit loaded.
> Both were ultimately activated, but something about page size failed:
>
> Nov 26 17:17:36 engels dbus[723]: [system] Activating via systemd:
> service name='org.freedesktop.ConsoleKit'
> unit='console-kit-daemon.service'
> Nov 26 17:17:36 engels console-kit-daemon[1310]: (process:1373):
> GLib-CRITICAL **: g_slice_set_config: assertion
> 'sys_page_size == 0' failed
This seems to be https://bugs.debian.org/718843. Maybe you should just
remove consolekit, it's deprecated and replaced by systemd-logind.
> This immediately followed by a problem getting IP address from my
> router's DHCP. The box has wifi card, but no ethernet cable. Before
> Jessie, wlan0 was assigned an address automatically, but with Jessie
> there's no assress and wlan-avahi instead gets a bogus address. I can
> do ifdown/ifup and get an IP address, but should not have to. Is DBus
> blocking the connection?
No, this does not have anything to to with dbus AFAICS.
> Nov 26 17:17:52 engels ntpd[896]: Deferring DNS for
> 1.debian.pool.ntp.org 1
> Nov 26 17:17:53 engels console-kit-daemon[1310]: (process:1540):
> GLib-CRITICAL **: g_slice_set_config: assertion
> 'sys_page_size == 0' failed
> Nov 26 17:17:55 engels dhclient[585]: DHCPDISCOVER on
> eth0 to 255.255.255.255 port 67 interval 15
> Nov 26 17:17:55 engels ifup[573]: DHCPDISCOVER on
> eth0 to 255.255.255.255 port 67 interval 15
> Nov 26 17:18:06 engels org.gtk.vfs.Daemon[1545]: A connection to
> the bus can't be made
> Nov 26 17:18:06 engels org.gtk.vfs.Daemon[1545]:
> g_dbus_connection_real_closed: Remote peer vanished with error:
> Underlying GIOStream returned 0 bytes on an async read
> (g-io-error-quark, 0). Exiting.
> Nov 26 17:18:06 engels console-kit-daemon[1310]: (process:1597):
> GLib-CRITICAL **: g_slice_set_config: assertion
> 'sys_page_size == 0' failed
> Nov 26 17:18:07 engels console-kit-daemon[1310]: (process:1599):
> GLib-CRITICAL **: g_slice_set_config: assertion
> 'sys_page_size == 0' failed
> Nov 26 17:18:10 engels dhclient[585]: No DHCPOFFERS received.
That's the problem, but without knowing how you connect to your router
(network-manager, ifupdown…) it's hard to tell why you don't get any
DHCPOFFERS from it.
Cheers,
Sven
Reply to: