incubator-cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alex Huang <Alex.Hu...@citrix.com>
Subject RE: [ACS41] Where to merge features?
Date Tue, 29 Jan 2013 21:49:47 GMT
> 
> I'm suggesting that there is risk, and that we need to agree on a
> mitigation for that risk.  I'm not suggesting anything beyond that.
> 
> What happens if it takes you longer than a day or two to get Javelin
> and master merged?  What do we do then?  Do you want to delay the
> 4.1.0 release schedule day for day?
> 
> Or perhaps this isn't enough of a risk (I need your input here) to
> even bother talking about mitigations?
> 
> Thanks!

Chip,

It's difficult to assess the risk until all the code that needs to be merged in is in.  From
experience, we did four merges from master, each merge was progressively easier and faster
than the previous.  From that perspective, I think the risk should be low.  

I'm fine with the risk suggestion. There obviously is one.  I do think that we move javelin
to the end of the code freeze to help the entire community so that instead of everyone merging
javelin into their branch, javelin takes the hit of merging everyone else's work.  So, it
doesn't make sense that there is the option of dropping javelin if it couldn't be merged in
one or two days.  If it takes a couple of weeks again, then we can consider such an option.
 But if it's within that, we should consider day-to-day slip.  

--Alex

Mime
View raw message