Hi all!
We've seen quite a few references on the list to "100% binary compatibility with upstream". What I am curious about is, how precisely is this determined? All the ways I can think of for comparing how two systems might work seem flawed in some way (e.g., using some sort of checksum; unit testing; verifying build parameters). I did some searches both at centos.org and google, but couldn't find anything specific about the test(s) used to determine compatibility.
I have no particular reason for asking, other than wanting to be able to explain it to others if someone asks me.
--keith
On Fri, Apr 8, 2011 at 8:49 AM, Keith Keller kkeller@wombat.san-francisco.ca.us wrote:
Hi all!
We've seen quite a few references on the list to "100% binary compatibility with upstream". What I am curious about is, how precisely is this determined? All the ways I can think of for comparing how two systems might work seem flawed in some way (e.g., using some sort of checksum; unit testing; verifying build parameters). I did some searches both at centos.org and google, but couldn't find anything specific about the test(s) used to determine compatibility.
In this FAQ:
http://wiki.centos.org/FAQ/General/RebuildReleaseProcess
"Once built ... we use the tmverifyrpms against it from here:
http://mirror.centos.org/centos-4/4/build/distro/ "
Akemi
On Fri, 2011-04-08 at 08:58 -0700, Akemi Yagi wrote:
On Fri, Apr 8, 2011 at 8:49 AM, Keith Keller kkeller@wombat.san-francisco.ca.us wrote:
Hi all!
We've seen quite a few references on the list to "100% binary compatibility with upstream". What I am curious about is, how precisely is this determined? All the ways I can think of for comparing how two systems might work seem flawed in some way (e.g., using some sort of checksum; unit testing; verifying build parameters). I did some searches both at centos.org and google, but couldn't find anything specific about the test(s) used to determine compatibility.
In this FAQ:
http://wiki.centos.org/FAQ/General/RebuildReleaseProcess
"Once built ... we use the tmverifyrpms against it from here:
http://mirror.centos.org/centos-4/4/build/distro/ "
Akemi _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Thanks Keith, good question, that should have been on my list of "Questions to ask about CentOS building process", and thanks to Akemi for a quick answer :)
On 8.4.2011 18:03, Mister IT Guru wrote:
On Fri, 2011-04-08 at 08:58 -0700, Akemi Yagi wrote:
On Fri, Apr 8, 2011 at 8:49 AM, Keith Keller kkeller@wombat.san-francisco.ca.us wrote:
Hi all!
We've seen quite a few references on the list to "100% binary compatibility with upstream". What I am curious about is, how precisely is this determined? All the ways I can think of for comparing how two systems might work seem flawed in some way (e.g., using some sort of checksum; unit testing; verifying build parameters). I did some searches both at centos.org and google, but couldn't find anything specific about the test(s) used to determine compatibility.
In this FAQ:
http://wiki.centos.org/FAQ/General/RebuildReleaseProcess
"Once built ... we use the tmverifyrpms against it from here:
http://mirror.centos.org/centos-4/4/build/distro/ "
Akemi _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Thanks Keith, good question, that should have been on my list of "Questions to ask about CentOS building process", and thanks to Akemi for a quick answer :)
Given that its answered in a FAQ one could argue that it was not a good question.
Thanks Keith, good question, that should have been on my list of "Questions to ask about CentOS building process", and thanks to Akemi for a quick answer :)
Given that its answered in a FAQ one could argue that it was not a good question.
You know, there is a famous saying.. "If you have nothing nice to say, don't say anything at all".
On Fri, Apr 8, 2011 at 12:59 PM, Dvorkin, Asya dvorkias@umdnj.edu wrote:
Thanks Keith, good question, that should have been on my list of "Questions to ask about CentOS building process", and thanks to Akemi for a quick answer :)
Given that its answered in a FAQ one could argue that it was not a good question.
You know, there is a famous saying.. "If you have nothing nice to say, don't say anything at all".
Yes, and not to mention the giant warning on the top of that page: Comment from Karanbir Singh: "Just want to point out that CentOS does not use anything from that page - and details / scripts on that page have nothing to do with the CentOS process."
// Brian Mathis
On Fri, 2011-04-08 at 14:47 -0400, Brian Mathis wrote:
On Fri, Apr 8, 2011 at 12:59 PM, Dvorkin, Asya dvorkias@umdnj.edu wrote:
Thanks Keith, good question, that should have been on my list of "Questions to ask about CentOS building process", and thanks to Akemi for a quick answer :)
Given that its answered in a FAQ one could argue that it was not a good question.
You know, there is a famous saying.. "If you have nothing nice to say, don't say anything at all".
Yes, and not to mention the giant warning on the top of that page: Comment from Karanbir Singh: "Just want to point out that CentOS does not use anything from that page - and details / scripts on that page have nothing to do with the CentOS process."
// Brian Mathis
Okay, I'm getting a touch confused here. Are you saying that this page is a red herring?? Because if it is irrelevant, then does anyone have an issue if that page gets nuked? What's the point if it's not helpful?
Hi
On 04/08/2011 07:47 PM, Brian Mathis wrote:
Yes, and not to mention the giant warning on the top of that page: Comment from Karanbir Singh: "Just want to point out that CentOS does not use anything from that page - and details / scripts on that page have nothing to do with the CentOS process."
I just wanted to make sure that people realise that while the process on that page will get you to where one needs to be, the scripts etc are not what we use in the centos buildsystem.
- KB
On Saturday, April 09, 2011 05:32:34 AM Karanbir Singh wrote:
I just wanted to make sure that people realise that while the process on that page will get you to where one needs to be, the scripts etc are not what we use in the centos buildsystem.
And that's a more useful statement than the quote that's up on that page now.
After 6.0 is out if you, Johnny, Tru, or whomever would be so kind as to update the publicly accessible scripts I'm sure many folk would be appreciative.
So we now have C4.9 and C5.6 out; two down, one to go. Good progress, even if it did take a bit of time.
On Sat, 2011-04-09 at 08:52 -0400, Lamar Owen wrote:
On Saturday, April 09, 2011 05:32:34 AM Karanbir Singh wrote:
I just wanted to make sure that people realise that while the process on that page will get you to where one needs to be, the scripts etc are not what we use in the centos buildsystem.
And that's a more useful statement than the quote that's up on that page now.
After 6.0 is out if you, Johnny, Tru, or whomever would be so kind as to update the publicly accessible scripts I'm sure many folk would be appreciative.
So we now have C4.9 and C5.6 out; two down, one to go. Good progress, even if it did take a bit of time.
Does this deserve to be a new thread, seems we're talking more about the "build system/mechanisms used to build rpms", and that's a thread I would really like to follow
-
Akemi Yagi -
Brian Mathis -
Dvorkin, Asya -
Karanbir Singh -
Keith Keller -
Lamar Owen -
Markus Falb -
Mister IT Guru