Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: (ASCEND) 5.0Ap33 woes ....





On Fri, 21 Nov 1997, Kevin Smith wrote:

> At 05:12 PM 11/21/97 -0500, Jason Nealis wrote:
> >
> >
> >On Fri, 21 Nov 1997, Kevin Smith wrote:
> >> 
> >> There is *NO* regression testing until the formal release - like 6.0 -
> that's
> >> been made clear numerous times, and is stated in the FTP site README files.
> >
> > Wait, I think saying there is no regression testing , That can not
> >be correct, Apparently, I was under the impression that when a Patch
> >Release is cut, They spend a couple days checking simple items like
> >routing, auth, acct, etc.. 
> >
> > I've got to be reading this incorrectly. Right Kevin?
> 
> Sorry, you are correct. I went a little too far. I meant to point out that
> *FULL REGRESSION TESTING* is not done except for full releases. We do a
> reduced regression test/sanity check for the intermediate releases in both
> the patch and incrementals branch. Please read the following from the FTP
> site for details of my intended 'message' :
> 
> Incremental Releases - An Incremental Release encompasses all new feature
> enhancements and bug fixes made since the last full release.  Since these
> releases have not been fully regression tested (only sanity checked) they
> are not as stable as the patch releases.  Incremental Releases are usually
> released by Wednesday of each week for all product lines.  

To have instability in the Incremental's I can understand, Myself, I never
dare to venture in to that realm.

> 
> Patch Releases - A Patch Release contains only bug fixes made since the last
> full release which will not affect the stability of the release.  Patch
> Releases may not be released on a regular interval.

"full release which will not affect the stability of the release", Good
marketing lingo there, But this far from true, As any can note on this
list. One protocol comes to mind here, "Slip", It should be changed to

"full release which may affect the stability of the release."

I'd admit the Max does a million and one things, and yes maybe it's a
little hard to check every option, but when items like slip/V.FAST/V.FC/
accounting are in one patch release and not the next, then something is
obviously wrong here. Ascend is a billion dollar company, Go hire some
more regression testing monkeys to test these items. 


Jason Nealis



> What does this mean?
> If you have a customer needing a fix made in both the patch and the
> incremental release, and stability is what you want, then use the Patch
> Release.  If there is a new feature the customer desires or if they have a
> problem fixed in the Incremental Release that a Patch Release does not
> address, then they should use the Incremental Release.
> 
> Kevin
> 
> 
> ++ Ascend Users Mailing List ++
> To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
> To get FAQ'd:	<http://www.nealis.net/ascend/faq>
> 

++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>


References: