[DRBD-user] Data scrubbing

Federico Simoncelli federico.simoncelli at gmail.com
Thu Sep 25 14:50:53 CEST 2008

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


On Wed, Sep 24, 2008 at 11:59 PM, Lars Ellenberg
<lars.ellenberg at linbit.com> wrote:
> On Wed, Sep 24, 2008 at 06:17:05PM +0200, Federico Simoncelli wrote:
>> Hi all, how do you perform data scrubbing on DRBD?
>
> for integrity check, you can use "verify" (in latest 8.2 releases).

I tried to run a verify on a two-primaries resource. It ran fine for a
couple of minutes and then I got these errors:

[...]
Sep 24 18:39:00 kernel: drbd0: [drbd0_worker/2780] sock_sendmsg time
expired, ko = 4294967267
Sep 24 18:39:06 kernel: drbd0: [drbd0_worker/2780] sock_sendmsg time
expired, ko = 4294967266
Sep 24 18:39:12 kernel: drbd0: [drbd0_worker/2780] sock_sendmsg time
expired, ko = 4294967265
[...]

On the drbd resource I have configured an lvm volume.

> drbd expects to run on reliable local io subsystems,
> basically we expect to sit (not necessarily directly)
> on top of some sort of hardware or software raid.

Basically this means that using drbd over a software/hardware raid is
almost mandatory in a production environment. Am I right?

Thanks in advance.
-- 
Federico.



More information about the drbd-user mailing list