Re: +2/+1 approach and Re: ServiceUnavailableException

Flavio Fernandes <ffernand@...>

On Feb 13, 2015, at 12:09 PM, Edward Warnicke <hagbard@...> wrote:

I'm not hugely comfortable requiring two committers to review a patch as a matter of policy, because some things (like the .gitignore) are pretty harmless and unalloyed goods and some thing are actually time critical (like fixing build breakage).

That said, I think for consequential things its probably a good idea to get more eyes if possible... but I think that should be at the discretion of the reviewer.

Does that make sense?

it does to me, Ed. I think using common sense on determining whether the complexity requires more than 1 pair of eyes is good.

you will def never have me +2 and submit something, unless I’m very comfortable with it.

— flavio


On Fri, Feb 13, 2015 at 9:36 AM, Ryan Moats <rmoats@...> wrote:

There is a commit to address this now [1], which brings up the question of +2/+1 behavior:

I'd like to propose that we have two committers look at each patch - the first +1s it and the second +2s it

That way we've got at least half the folks reviewing each patch...




Flavio Fernandes <ffernand@...> wrote on 02/13/2015 06:43:10 AM:

> From: Flavio Fernandes <ffernand@...>

> To: Ryan Moats/Omaha/IBM@IBMUS
> Cc: neutron-dev@...
> Date: 02/13/2015 06:44 AM
> Subject: ServiceUnavailableException
> Hi Ryan,
> I just noticed that in some files you changed yesterday [1], the
> error code 503 is not mentioned in the
>     @StatusCodes({
>        ...
>       @ResponseCode(code = 503, condition = “CONDITION") })  <— missing
> block. Is that a reason for concern? I’m not familiar with the
> requirements for having all these codes
> spelled out in there. Sorry for not caching that sooner.
> Best,
> — flavio
> [1]:

neutron-dev mailing list

neutron-dev mailing list

Join { to automatically receive all group messages.