dm-thin and shared storage failover

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hello,

I am trying to come with an efficient dm-thin HA setup and there is very few
litterature on the subject since dm-thin is so new.

So I will ask my questions to this list so the responses could serve as later
reference for the interweb.

I clearly understood that dm-thin is not clustered and should not be used at the
same time on two nodes.
So I am thinking of an active/passive setup with shared SAS storage.

I have the following setup in mind.

A SAS storage bay would be shared between two active/passive nodes.

dm-thin and lvm modules would be loaded on the active node and not on the
passive node.

On an active node crash or power failure the passive node would load lvm modules
and the dm-thin module taking control of the SAS enclosure.

I can make sure that the upper layer is redundant between the node so no IOs is
lost during failover. This layer would open the dm-thin volume on the passive
node after takeover happened an redo IOs lost in the crash.

My questions boils down to:

How dm-thin cope with crash or powerfailure ?
What is the level of atomicity in dm-thin IOs ?
Are operations done via a journal or soft updates ?

Generally do you think this kind of setup would work ?

Best regards

Benoît Canet

--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel





[Index of Archives]     [DM Crypt]     [Fedora Desktop]     [ATA RAID]     [Fedora Marketing]     [Fedora Packaging]     [Fedora SELinux]     [Yosemite Discussion]     [KDE Users]     [Fedora Docs]

  Powered by Linux