Re: storage server fail-over
On 7/13/2012 8:40 AM, Muhammad Yousuf Khan wrote:
> thanks, i was also looking for that but i didn't have the idea if
> thats gonna work since i have zero knowledge about DRBD.
> anyways thanks for the tip it is very helpful.
DRBD simply mirrors disk blocks between two hosts' block devices (single
disk, mirror pair, hardware or software RAID device). That will allow
you a manual or scripted failover when one node fails. If you want an
active/active HA Samba cluster with both hosts serving requests, you'll
need to format the shared DRBD disk device with a cluster filesystem
such as GFS2 or OCFS2. Here's some decent docs for that:
http://doc.opensuse.org/products/draft/SLE-HA/SLE-ha-guide_sd_draft/cha.ha.samba.html
--
Stan
> On Fri, Jul 13, 2012 at 5:52 PM, Stan Hoeppner <stan@hardwarefreak.com> wrote:
>> On 7/13/2012 2:26 AM, Muhammad Yousuf Khan wrote:
>>> i have a storage server (samba) and now there is a requirement from my
>>> management that i should provide a centralized document storage server
>>> where people can store their working document. (so this means another
>>> layer of responsibility :*(, (never mind)).
>>>
>>> i already have raid 1 implemented on SAMBA Server and using it as a
>>> Data backup and archiving, so what if machine dies?
>>> so my question is do you guys think that the best way of doing this is
>>> SAMBA or ISCSI or any other network sharing service which also provide
>>> a good fail-over plan.
>>
>> http://www.drbd.org/
>>
>> That's the cheapest method.
>>
>> --
>> Stan
>>
>
>
Reply to: