Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hegedus Gabor wrote:
> How can I test the drbd system? I mean, eg a disk faulire, network
> faulire?
>
Disk failures, well the manpage tells:
on-io-error handler
handler is taken, if the lower level device reports
io-error to the upper layers.
handler may be pass_on, panic, or detach.
pass_on: Report the io-error to the upper layers. On
Primary report it to the mounted file system. On Secondary ignore it.
panic: The node leaves the cluster by doing a kernel panic.
detach: The node drops its low level device, and continues
in disk less mode.
Yesterday one of my scsi cabinets got busfree errors and drbd detached
it and became a disklessclient. This is the default behaviour.
Network failures can be simulated by adding deny rule using iptables or
if you use a dedicated interface for the replication just ifdown.
-- Leroy