fsck

[Bob Martin]

Tim,
What gets mounted, and where, is determined by the file /ete/fstab. 
Unless you have entries in that file for the /dev/da1 disk, it will not 
be mounted. fsck is just reading a little entry off the disk that says 
where the slice was last mounted... A truly priceless thing to know if 
your building a phoenix out of someone else's ashes.

I have seen fsck take all night to run. More over, I've seen your soft 
update errors more often than I would have liked. At least one directory 
on this slice is horribly corrupted. Unless there are files on there you 
just can't live without, (e.g. email that's in the spool, etc.), I would 
just newfs the slice, and start over. If that's not an option, run fsck 
-y /dev/da1s1f. Wait (all night if needs be) If it gives soft update 
errors, cold reboot the box (make sure the drives come to a full stop) 
and repeat the process until the errors go away. Watch the output. If 
fsck runs a ways, fixes some errors then hangs, kill it and restart it. 
If it's not making any head way at all, just be really patient. (The -y 
makes it go a little faster)

One foot note to running newfs. While most of the stuff on var can be 
recreated, you'll loose your ports database. Something to keep in mind.

Since you have a backup (you do have a backup don't you?) you can just 
newfs the file system and restore... I mention this now as it was an 
event just like this that drove me to doing regular, complete backups. I 
even back up my home system daily. (There is magic server pixy dust... 
It's called dump :) )

Bob Martin

Tim McCullagh wrote:
> Hi Kurt
> 
> Thanks for the response.
> 
> I have run the fsck /dev/da1s1f and waited over an hour.  When I do a ctrl
> +t  it gets to somewhere between 1440K and 2880K of a 16GB partition and
> seems to just stop.
> 
> Is it possible that because the partition was mounted as /var on /dev/da0s1f
> on the original machine and when fsck reads it it wants to mount it as /var,
> but is not able to because on the new host machine on which it is mounted as
> /dev/da1s1f that the new host machine has a /var on /dev/da0s1f.   Would
> this have any affect at all?
> 
> Regards
> 
> Tim
> 
> ----- Original Message -----
> From: "Kurt Jaeger" <lists at complx.LF.net>
> To: "Tim McCullagh" <timbo at halenet.com.au>
> Cc: "Putinas Piliponis" <putinas.piliponis at icnspot.net>;
> <freebsd-isp at FreeBSD.ORG>
> Sent: Wednesday, January 21, 2004 4:30 AM
> Subject: Re: fsck
> 
> 
> 
>>Hi!
>>
>>
>>>Message from syslogd at cathy at Tue Jan 20 22:05:56 2004 ...
>>>cathy /kernel: <<<<<<<<<<<<<<<< Dump Card State Ends >>>>>>>>>>>>>>>>>>
>>>
>>>Then It freezes or continues to give
>>>cathy /kernel: <<<<<<<<<<<<<<<< Dump Card State Ends >>>>>>>>>>>>>>>>>>
>>>errors
>>>
>>>Should I just leave it and wait?
>>
>>Yes, try to wait a while (10 minutes?). I've seen these messages
>>as well, and as of know they seem harmless and do not stop the
>>system.
>>
>>But: Your system may vary...
>>
>>--
>>MfG/Best regards, Kurt Jaeger                                  16 years to
> 
> go !
> 
>>LF.net GmbH        fon +49 711 90074-23  pi at LF.net
>>Ruppmannstr. 27    fax +49 711 90074-33
>>D-70565 Stuttgart  mob +49 171 3101372
>>
> 
> 
> _______________________________________________
> freebsd-isp at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-isp
> To unsubscribe, send any mail to "freebsd-isp-unsubscribe at freebsd.org"