[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: /etc/resolv.conf replaced!

Hal Vaughan wrote:
On Tuesday 06 December 2005 05:25 pm, Hal Vaughan wrote:

During the past two days I've had power flickers and outages from snow.  I
have not yet been able to attach a new system to a UPS (too many things to
move), and it's lost power a few times.  When it reboots, it cannot
communicate with the network.  I checked, and /etc/resolv.conf is no longer
there.  It has been replaced by a link to /etc/resolvconf/run/resolv.conf,
and /etc/resolvconf/run is linked to /dev/shm/resolvconf.  There is no
corresponding file in /dev/shm/resolvconf to work with.

I've found references to the other files on Google, but nothing clear
telling what is going on.  There are references that make me think some
program *thinks* it is supposed to do this (so I doubt it's a virus), but I
need to find out what is going on so I can either stop it or make sure it
does it right.

So how can I stop this from happening at reboot?  What is doing it?  Is it
a boot thing, or a re-configure thing?

Any help is appreciated.


Thanks for the responses. Since every response suggested the problem was with resolvconf (the package, not the file), I checked, even though I had previously seen aptitude report it as not installed. I checked with dpkg, and it reported the same, but I tried 'aptitude purge resolvconf' and it said it was purging 1 package. While I had asked about someting similar on another mailing list, this system is my "money" system that pays the bills, so I do NOT experiment on it. I did not install resolvconf on it, and if anything depended on it, aptitude would have removed it as well. So I have two mysteries: 1) How did the package end up there in the first place, and 2) Why didn't aptitude and dpkg report it as installed, yet aptitude still found something to remove (and it removed the sub directory).

I haven't rebooted, and likely won't until later this week when I have to shut everything down to move the systems to where the old ones were (so they can be connected to my KVM and the UPS). When that happens, I'll see if there's any problem.

But more interesting: what happened after the power out?

Those happen here quite frequently (no snow, lots of sun :-) ) and what you have to do is closely watch the e2fsck run after the powerout and during the reboot: he will indicate what he found creamed.

E.g during an oops a few days ago I lost the capability to reboot. (Problem with fbsplash) During the red button reboot e2fsck noted he found a problem with chrony.rtc. Sure enough it was gone.
That is the thing to watch during reboot.


Reply to: