Where report problems with the debian startup system (more then one package affected)
Hi,
during the last weeks I found some really surprising effects with various
packages which all seems to be associated with the debian startup system.
To be more precise - I assume the problems seems to be related either with the
order the services start or with the time between the startups of the affected
services.
(No, the value of the variable CONCURRENCY in /etc/init.d/rc each time was
'none' and has never changed.)
For example I get - each time a machine starts up - an entry in exims panic log
file
'Exim configuration error in line 645 of /var/lib/exim4/config.autogenerated.tmp
I reinstalled exim and all (possible) related packages with no effect.
But stopping exim, removing/clearing the panic log and restarting exim works
with no errors. So I assume that exim needs the availability of various
services/daemons but they are not available at the time exim starts during the
boot process.
Another example seems to be related to nfs-common and/or nfs-kernel-server:
Sometimes (sorry I can't be more precise) I get lots of messages like
'rpc.statd[xxx]: recv_rply: can't decode RPC message!'
Restarting nfs-common and nfs-kernel-server seems to solve this problem.
Finally a third example:
on one machine - the user is preferring gnome instead the mainly used KDE
desktop - I get lots of messages like
'gnome-keyring-daemon[xxx]: Scheduling hal init retry'
and
'gnome-keyring-daemon[xxx]: failed to initialize a HAL context: (null)'
As far as I could see this has something to do with the start sequence of the
services hal and dhcdbd. In the syslog files I can find the following 2
messages each time the machine boots.
'dhcdbd: dbus_svc_init failed: org.freedesktop.DBus.Error.NoReply Did not
receive a reply. Possible causes include: the remote application did not send a
reply, the message
bus security policy blocked the reply, the reply timeout expired, or the
network connection was broken.'
'dhcdbd: Failed to initialise D-Bus service.'
It seems that a variation of the startup sequence could solve this problem.
Abstract:
With some fortune - at least I hope so - I can fix the problems for my machines
and for now. But - I'm sure - the next update happens an then - the same
procedure starts again.
Please could somebody help me to forward this problem to the right people.
BTW, all the problems reported above appear only with machines which were
started at least once per day. Machines which are up the whole time are not
affected.
TIA
Cheers, Christian
--
Name: Christian Andretzky | Address: TU Chemnitz, Fakultaet fuer |
Phone: ++49 +371 531 32130 | Elektrotechnik/Informationstechnik |
FAX: ++49 +371 531 832130 | Reichenhainer Str. 70 |
mail: Christian.Andretzky@MB.TU-Chemnitz.DE | D-09107 Chemnitz |
PGP: public key avaliable via public key server |
In a world without walls and fences, there is no need for windows and gates |
"The Box said 'Windows 95 or better'. - So I installed Linux." |
Reply to: