[DRBD-user] DRBD suitable for Fast Failover ?

DRBD User DRBDUser at gmx.at
Fri May 29 11:38:38 CEST 2015

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


2556

So a Master/Slave Configuration would make no difference (without GFS2 ...) ?
I assume fencing is required too ...

Or may be using DRBD in standalone mode ?

Gesendet: Donnerstag, 28. Mai 2015 um 09:50 Uhr
Von: Digimer <lists at alteeve.ca>
An: "DRBD User" <DRBDUser at gmx.at>, drbd-user at lists.linbit.com
Betreff: Re: [DRBD-user] DRBD suitable for Fast Failover ?
On 28/05/15 02:56 AM, DRBD User wrote:
> My goal is to achieve a failover of about 1-2 seconds.

Almost impossible to do sanely. Fencing is just not that fast... Add in
the overhead of detecting the fault (drbd and corosync), initiating the
fence, confirming it succeeded and then cleaning up locks/replaying FS
journals, etc... 10 seconds would be impressive.

--
Digimer
Papers and Projects: https://alteeve.ca/w/
What if the cure for cancer is trapped in the mind of a person without
access to education?



More information about the drbd-user mailing list