[DRBD-user] Hanging kernel

Theo Band theo.band at greenpeak.com
Thu Sep 23 14:57:41 CEST 2010

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.

 On 09/23/2010 02:32 PM, Alex Adriaanse wrote:
>  This looks like a complicated setup. Is this because you cannot run the
>> entire machine from a drbd disk? Is this intended for failover?
> Right.  This is a failover setup where all data is replicated to a
> backup server that's located offsite.  I'm not running the entire
> machine from a DRBD disk because that means the root filesystem would
> have to boot from the DRBD device.  Thus, the initrd image would have
> to configure the drbd device, and that's not supported on Debian AFAIK. 
If you create a virtual machine whose disk is a drbd device, you've
covered everything. That's the setup I use. Advantage is that you know
for sure that the machine will boot as nothing is left out. For
bandwidth it does not matter as only blocks that change will be replicated.
Anyway this has nothing to do with your hanging kernels.

>blocked for more than 120 seconds
When I got this this message in the past, I was sometimes able to solve
it by manually disconnecting and reconnecting. I also got it once while
reloading iptables (I added a additional port for a new drbd device). I
guess the netfilter stack can cause a hickup in the drbd tcp link. But
it never happened with the verify enabled.

More information about the drbd-user mailing list