drbd WAS working fine; but now fsck can't open /dev/drbd? ??

[Bob]

running debian etch; drbd v0.7; along with HA; under ext3 filesystems;

both system were running fine, and drbd work working great.
assuming there was a power outage, now when I bring up the primary
system I get (during boot):
>>
fsck.ext3: no such file or directory while trying to open /dev/drbd0
the superblock could not be read or does not describe a correct ext2
filesystem.
fsck died with exit status 8 ....[snip]... superblock may be
corrupt.
<<

/dev/drbd0 is drbd's device to /dev/sdb1; which DOES fsck PERFECTLY.

this system isn't even coming up to runlevel 2... going into maint.
mode....

i've tried ef2check -f and it shows filesys clean... (again on /dev/
sdb1)

/dev/drbd0 doesn't exist until drbd is loaded; but if I load drbd,
there's no way I can fsck /dev/drbd0 because "it's busy".

anyone have any idea's ??

TIA - Bob

--
To UNSUBSCRIBE, email to debian-us...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

[Bob]

As a followup to my own post -

I've since tried inserting drbd into /etc/modules, and /etc/initramfs-
tools/modules (and updated initramfs).

But nothing I do seems to cause the drbd module to load prior to fsck
running..
(I do have drbd0.7-module-source installed) and the module DOES load
& work fine, if I manually continue past the boot issue.

Any other thoughts/suggestions?

TIA - Bob

On Jul 14, 1:00 pm, Bob <bobg.h...@gmail.com> wrote:
> running debian etch; drbd v0.7; along with HA; under ext3 filesystems;
>
> both system were running fine, and drbd work working great.
> assuming there was a power outage, now when I bring up the primary
> system I get (during boot):
>
> fsck.ext3: no such file or directory while trying to open /dev/drbd0
> the superblock could not be read or does not describe a correct ext2
> filesystem.
> fsck died with exit status 8 ....[snip]... superblock may be
> corrupt.
> <<
>
> /dev/drbd0 is drbd's device to /dev/sdb1; which DOES fsck PERFECTLY.
>
> this system isn't even coming up to runlevel 2... going into maint.
> mode....
>
> i've tried ef2check -f and it shows filesys clean... (again on /dev/
> sdb1)
>
> /dev/drbd0 doesn't exist until drbd is loaded; but if I load drbd,
> there's no way I can fsck /dev/drbd0 because "it's busy".
>
> anyone have any idea's ??
>
> TIA - Bob
>
> --

> To UNSUBSCRIBE, email to debian-user-REQU...@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

[Lucas Mocellin]

Hi Bob,

I can't help you with your trouble, but I'm interested in drbd, can
you post your cenario? I'll test it.

Thanks,

Lucas.

2008/7/16, Bob <bobg...@gmail.com>:

[Bob]

;-) well - as a followup to my followup...

I've confirmed that the drbd module IS loading.

The problem fsck is having is that drbd is marking the partition as
secondary, and as such, it is not RW accesssible.

So now this is looking like a drbd config problem; but all does look
fine; so I'm still a bit lost...

Bob

[Bob]

Hi Lucas -

happy to share my little bit of knowledge :)

I've got 2 servers - a primary & a secondary. Each has 2 hard drives:
the first, /dev/sda, is a typical linux config'd drive. the second, /
dev/sdb, specifically /dev/sdb1 is setup to be my cloned partition
across the 2 systems.

I used package doc's, as well as some I found on howtoforge, to walk
thru setting up & configing the 2 systems. I was trying to figure out
why HA wasn't working correctly - I brought down the #1, and #2 saw &
acknkowledged this, but never mounted the sdb1 partition. The next
day, #1 would not boot up without complaining about fsck'g /dev/drbd0
(which is the drbd device for /dev/sdb1).

You can see in my last self-followup, where I'm at diagnosing the
problem.

not sure how much detail you'd like, but in case you do - here's my
config files: (lintest = #1; lintest2 = #2)
/etc/drbd.conf: ==>
resource varb {
protocol C;
incon-degr-cmd "echo '!DRBD! pri on incon-degr' | wall ; sleep 60 ;
halt -f";

startup {
wfc-timeout 10;
degr-wfc-timeout 30; # 2 minutes. -> 30 seconds
}

disk {
on-io-error detach;
}

net {
max-buffers 20000; # 2048 -> 20000 (play w/ this # for perf)
unplug-watermark 12000; # 128 -> 12000 (play w/ this # for
perf)
max-epoch-size 20000; # should equal max-buffers
}

syncer {
rate 10M;
group 1;
al-extents 257;
}

on lintest {
device /dev/drbd0;
disk /dev/sdb1;
address 192.168.1.59:7788;
meta-disk internal;
}

on lintest2 {
device /dev/drbd0;
disk /dev/sdb1;
address 192.168.1.58:7788;
meta-disk internal;
}
}
<==

on the HA side:
/etc/ha.d/authkeys:
auth 1
1 crc

/etc/ha.d/haresources:
lintest 192.168.1.55 Filesystem::/dev/drbd0::/mnt::ext3

/etc/ha.d/ha.cf: ==>
logfile /var/log/ha-log
logfacility local0
keepalive 2 # change to 1????
deadtime 30 # drop to 10????
warntime 10
initdead 120
udpport 694
bcast eth0 # Linux
auto_failback on
node lintest
node lintest2
respawn hacluster /usr/lib/heartbeat/ipfail
<==

On Jul 16, 3:30 pm, "Lucas Mocellin" <lucasmocel...@gmail.com> wrote:
> Hi Bob,
>
> I can't help you with your trouble, but I'm interested in drbd, can
> you post your cenario? I'll test it.
>
> Thanks,
>
> Lucas.
>

> 2008/7/16, Bob <bobg.h...@gmail.com>:

--

[Bob]

Hi Lucas -

disk {
on-io-error detach;
}

> Hi Bob,
>
> I can't help you with your trouble, but I'm interested in drbd, can
> you post your cenario? I'll test it.
>
> Thanks,
>
> Lucas.
>

> 2008/7/16, Bob <bobg.h...@gmail.com>:

--