cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chip Childers <>
Subject Re: Drop OVM in 4.0?
Date Thu, 18 Oct 2012 14:11:34 GMT
On Wed, Oct 17, 2012 at 10:30 PM, Joe Brockmeier <> wrote:
> On Wed, Oct 17, 2012 at 04:50:41PM -0700, Chiradeep Vittal wrote:
>> +1 on fixing in a 4.0.x.
>> We need to make some tough decisions if we want "time based releases".
>> We already dropped AutoScale and Brocade even though they were essentially
>> code-complete for this reason.
>> This is an issue that is of uncertain size (weeks at least) to fix. There
>> is just too much uncertainty.
>> Also it seems that folks want the latest version of OVM anyway, which
>> would not be helped by fixing this in 4.0
> We do need to make tough decisions for time-based releases, but we're
> not quite there yet.
> IIRC one of the criteria for 4.0.0-incubating is that we'd support
> upgrades from prior releases of CloudStack[1] and AutoScale and Brocade
> were not in those releases - OVM was.

That certainly was the case, but we might need to rethink that
assumption.  There are really 2 issues: supporting the latest version
of OVM and / or fixing the previous OVM support.

I understand the point about supporting the latest OVM version being
"at least > 1 week".  I haven't seen any answer to "fixing" the issues
for the previously supported OVM version.  Does anyone have a good
handle on that effort?  If it's equal or similar to making it work on
the latest OVM, then perhaps we are talking about the same thing...
If they are significantly different efforts, then we have to make a
different decision.

> It might be easier for folks to make a decision if we had a clear
> timeline for a 4.0.x release with OVM support and some idea who's going
> to be picking that up. (e.g., something like "Edison will be working on
> this feature and we're expecting it to be in master within two weeks of
> shipping 4.0.0-incubating).

Or at least a reasonable estimate for the different versions of
"fixing" this problem (per my question above).

> Holding the release up for an additional week (or longer) is not an
> attractive option, but neither is letting down a whole set of users who
> depend on OVM.


> [1]CloudStack - not CloudPlatform
> --
> Joe Brockmeier
> Twitter: @jzb

View raw message