[Lustre-discuss] Start the MGS first?

[Nathaniel Rutman]

James Beal wrote:
>
> On 25 Mar 2009, at 19:48, Nathaniel Rutman wrote:
>
>> As of Lustre 1.6, servers can be started in any order (after the 
>> initial registration at first startup).  Internally, this required 
>> significant bending of our connection rules, and with a move toward 
>> ZFS becomes even more burdensome.
>> So my question to the Lustre community is this: would anyone 
>> strenuously object to a startup ordering requirement that the MGS 
>> must be started before any other servers?
>> This would probably be in the Lustre 3.0 timeframe.  It is also 
>> likely that we will have to divorce the MGS and MDT onto separate 
>> devices -- no more "combo" MDT/MGSes.
>>
>>>> NR> I think the only reason to have a local config file is to be 
>>>> able to
>>>> NR> start a server in the absence of the MGS.  How much effort do 
>>>> we want
>>>> NR> to expend to be able to keep that ability?  I don't think it's 
>>>> a huge
>>>> NR> burden to say "MGS must be started first".
>
> Our current lustre heartbeat configurations rely on being able to 
> start servers in any order.
>
> While ensuring that the MGS is started before the MDT would be 
> relatively simple. Ensuring it starts before the OSTs would be 
> difficult as we a system where we have pairs of heartbeat systems and 
> they do not know the state of one and another. We have found small 
> heartbeat systems to be more reliable than large ones and this is the 
> reason we have multiple simple heartbeat systems.
>
Thanks - this is exactly the kind of feedback I was hoping for.

You wouldn't actually have to ensure the MGS starts first -- the OST and 
MDT mounts would block until the MGS was ready.  So if they are truly 
independent, everything will simply wait until the MGS has started, if I 
understand your setup correctly.