Обсуждение: Ready to branch 7.2/7.3 ?
I'm getting itchy to get to work on 7.3 development, and with so few
bug reports coming in for 7.2, it seems that now is not too soon to
split off the 7.2 maintenance branch in CVS. Comments?
regards, tom lane
Definitely agree on the itchiness, especially with the broken planner in v7.2? :) On Tue, 12 Feb 2002, Tom Lane wrote: > I'm getting itchy to get to work on 7.3 development, and with so few > bug reports coming in for 7.2, it seems that now is not too soon to > split off the 7.2 maintenance branch in CVS. Comments? > > regards, tom lane > > ---------------------------(end of broadcast)--------------------------- > TIP 5: Have you checked our extensive FAQ? > > http://www.postgresql.org/users-lounge/docs/faq.html >
I vote for a fork - 7.2 seems to be very stable (as it should be after a year!) Chris > -----Original Message----- > From: pgsql-hackers-owner@postgresql.org > [mailto:pgsql-hackers-owner@postgresql.org]On Behalf Of Marc G. Fournier > Sent: Wednesday, 13 February 2002 11:18 AM > To: Tom Lane > Cc: pgsql-hackers@postgresql.org > Subject: Re: [HACKERS] Ready to branch 7.2/7.3 ? > > > > Definitely agree on the itchiness, especially with the broken planner in > v7.2? :) > > On Tue, 12 Feb 2002, Tom Lane wrote: > > > I'm getting itchy to get to work on 7.3 development, and with so few > > bug reports coming in for 7.2, it seems that now is not too soon to > > split off the 7.2 maintenance branch in CVS. Comments? > > > > regards, tom lane > > > > ---------------------------(end of broadcast)--------------------------- > > TIP 5: Have you checked our extensive FAQ? > > > > http://www.postgresql.org/users-lounge/docs/faq.html > > > > > ---------------------------(end of broadcast)--------------------------- > TIP 4: Don't 'kill -9' the postmaster >
Tom Lane writes: > I'm getting itchy to get to work on 7.3 development, and with so few > bug reports coming in for 7.2, it seems that now is not too soon to > split off the 7.2 maintenance branch in CVS. Comments? The ecpg thing should probably be fixed first. Other than that it would be nice to have a new branch by the weekend. -- Peter Eisentraut peter_e@gmx.net
Peter Eisentraut <peter_e@gmx.net> writes:
> The ecpg thing should probably be fixed first. Other than that it would
> be nice to have a new branch by the weekend.
I musta missed something. What ecpg thing?
regards, tom lane
Tom Lane writes: > Peter Eisentraut <peter_e@gmx.net> writes: > > The ecpg thing should probably be fixed first. Other than that it would > > be nice to have a new branch by the weekend. > > I musta missed something. What ecpg thing? The sqlca warning -- Peter Eisentraut peter_e@gmx.net
Peter Eisentraut <peter_e@gmx.net> writes:
> Tom Lane writes:
>> I musta missed something. What ecpg thing?
> The sqlca warning
Um, we can probably manage to double-patch a removal of two lines ;-)
However, since there were no objections, what are you waiting for?
Commit it.
regards, tom lane
Okay, if the ecpg is the only issue, does everyone feel comfortable with a branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night based on the branch, with an announce going out on Monday? That way, those anxious to dive into v7.3 have the whole weekend :) On Thu, 14 Feb 2002, Tom Lane wrote: > Peter Eisentraut <peter_e@gmx.net> writes: > > Tom Lane writes: > >> I musta missed something. What ecpg thing? > > > The sqlca warning > > Um, we can probably manage to double-patch a removal of two lines ;-) > > However, since there were no objections, what are you waiting for? > Commit it. > > regards, tom lane > > ---------------------------(end of broadcast)--------------------------- > TIP 3: if posting/reading through Usenet, please send an appropriate > subscribe-nomail command to majordomo@postgresql.org so that your > message can get through to the mailing list cleanly >
"Marc G. Fournier" <scrappy@hub.org> writes:
> Okay, if the ecpg is the only issue, does everyone feel comfortable with a
> branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night
> based on the branch, with an announce going out on Monday?
I don't think it's time for 7.2.1 quite yet; we should probably wait
another week or two to see what comes in. I just want to branch now...
regards, tom lane
On Fri, 15 Feb 2002, Tom Lane wrote: > "Marc G. Fournier" <scrappy@hub.org> writes: > > Okay, if the ecpg is the only issue, does everyone feel comfortable with a > > branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night > > based on the branch, with an announce going out on Monday? > > I don't think it's time for 7.2.1 quite yet; we should probably wait > another week or two to see what comes in. I just want to branch now... Ahhhh, okay ... I was going by the old 'the way we've always done it' mentality, sorry :) But, this sounds cool ... most happy to do that ... I'll do it tonight, if nobody objects?
On Fri, 15 Feb 2002, Bruce Momjian wrote: > Tom Lane wrote: > > "Marc G. Fournier" <scrappy@hub.org> writes: > > > Okay, if the ecpg is the only issue, does everyone feel comfortable with a > > > branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night > > > based on the branch, with an announce going out on Monday? > > > > I don't think it's time for 7.2.1 quite yet; we should probably wait > > another week or two to see what comes in. I just want to branch now... > > Agreed. Let's get everything we can into 7.2.1. We normally don't push > out a minor this quickly unless we have a major thing to fix, which we > don't. Right, and we never branch until we're ready for the first minor ... so ...
"Marc G. Fournier" wrote: > > On Fri, 15 Feb 2002, Bruce Momjian wrote: > > > Tom Lane wrote: > > > "Marc G. Fournier" <scrappy@hub.org> writes: > > > > Okay, if the ecpg is the only issue, does everyone feel comfortable with a > > > > branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night > > > > based on the branch, with an announce going out on Monday? > > > > > > I don't think it's time for 7.2.1 quite yet; we should probably wait > > > another week or two to see what comes in. I just want to branch now... > > > > Agreed. Let's get everything we can into 7.2.1. We normally don't push > > out a minor this quickly unless we have a major thing to fix, which we > > don't. > > Right, and we never branch until we're ready for the first minor ... so > ... Hang on here. It sounds like we're following previous methodology to our (possible) slight detriment here. At present, we need 7.3 to start, so that people can begin working on it. We also need to have the 7.2 branch, so stuff for the to-be 7.2.x can be included there where appropriate. So, lets get it branched, because that's what needs to be done here and now, in this instance. Then people with a burn to do 7.3 stuff can do so, and things can be added to 7.2.x where needed. :-) Regards and best wishes, Justin Clift > > ---------------------------(end of broadcast)--------------------------- > TIP 6: Have you searched our list archives? > > http://archives.postgresql.org -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
On Fri, 15 Feb 2002, Bruce Momjian wrote: > Marc G. Fournier wrote: > > On Fri, 15 Feb 2002, Bruce Momjian wrote: > > > > > Tom Lane wrote: > > > > "Marc G. Fournier" <scrappy@hub.org> writes: > > > > > Okay, if the ecpg is the only issue, does everyone feel comfortable with a > > > > > branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night > > > > > based on the branch, with an announce going out on Monday? > > > > > > > > I don't think it's time for 7.2.1 quite yet; we should probably wait > > > > another week or two to see what comes in. I just want to branch now... > > > > > > Agreed. Let's get everything we can into 7.2.1. We normally don't push > > > out a minor this quickly unless we have a major thing to fix, which we > > > don't. > > > > Right, and we never branch until we're ready for the first minor ... so > > ... > > Were you asking a question? > > We can branch before the minor, right? Backpatching into the first > minor isn't a bigger deal than backpatching into later minors, and we > aren't patching much of anything now anyway. *rofl* And you've been arguing *against* this for how many releases now, and I've tried to get you guys to go along with this for how many releases? :)
We're planning on ... this is a 'sore point' that I've been arguing for several releases yet ... I still feel the branch should be made *on release date* and not several weeks later ... I think this is the first release where we will do the branch prior to the minor being released, so, I guess, as time goes on, we'll get closer to what I've wanted for ages :) On Sat, 16 Feb 2002, Justin Clift wrote: > "Marc G. Fournier" wrote: > > > > On Fri, 15 Feb 2002, Bruce Momjian wrote: > > > > > Tom Lane wrote: > > > > "Marc G. Fournier" <scrappy@hub.org> writes: > > > > > Okay, if the ecpg is the only issue, does everyone feel comfortable with a > > > > > branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night > > > > > based on the branch, with an announce going out on Monday? > > > > > > > > I don't think it's time for 7.2.1 quite yet; we should probably wait > > > > another week or two to see what comes in. I just want to branch now... > > > > > > Agreed. Let's get everything we can into 7.2.1. We normally don't push > > > out a minor this quickly unless we have a major thing to fix, which we > > > don't. > > > > Right, and we never branch until we're ready for the first minor ... so > > ... > > Hang on here. It sounds like we're following previous methodology to > our (possible) slight detriment here. > > At present, we need 7.3 to start, so that people can begin working on > it. > > We also need to have the 7.2 branch, so stuff for the to-be 7.2.x can be > included there where appropriate. > > So, lets get it branched, because that's what needs to be done here and > now, in this instance. > > Then people with a burn to do 7.3 stuff can do so, and things can be > added to 7.2.x where needed. > > :-) > > Regards and best wishes, > > Justin Clift > > > > > > > ---------------------------(end of broadcast)--------------------------- > > TIP 6: Have you searched our list archives? > > > > http://archives.postgresql.org > > -- > "My grandfather once told me that there are two kinds of people: those > who work and those who take the credit. He told me to try to be in the > first group; there was less competition there." > - Indira Gandhi >
"Marc G. Fournier" <scrappy@hub.org> writes:
>> We can branch before the minor, right? Backpatching into the first
>> minor isn't a bigger deal than backpatching into later minors, and we
>> aren't patching much of anything now anyway.
> *rofl* And you've been arguing *against* this for how many releases now,
> and I've tried to get you guys to go along with this for how many
> releases? :)
Hey guys: there is no black or white on this. It's a tradeoff ---
delaying next-version development versus effort wasted to do double
patching.
In this particular cycle, I'm for an early branch because we don't seem
to have a lot of bugs coming in, and we've got a lot of development work
that people are eager to get started on (or even to apply already-done
patches, in some cases). Both of these facts have doubtless got a lot
to do with the horrendously long release cycle for 7.2 --- which is
something I am *not* happy about.
I think the fact that we want to branch so soon after release is an
indication that we delayed the release too long. 7.2 should have been
out months ago.
regards, tom lane
I guess I don't understand the issue at hand. Why wasn't the branch created when the release was made? Why can't 7.3 development start immediately? After all, once the branch is made, any 7.2 patches can be applied to the branch and specific patches can be tagged as such on the branch. Likewise, merging them back to the truck isn't exactly hard to do once you're ready to migrate back into the 7.3 development process. While I've not checked for any tags in CVS, I have noticed that some code as changes since the 7.2 release so if you tag CVS based on float/truck, 7.2 CVS != 7.2 release. Is there a document that explain's the general CM approach around here? What gives? Sorry if this is explained elsewhere. Greg On Fri, 2002-02-15 at 21:16, Justin Clift wrote: [snip] > Hang on here. It sounds like we're following previous methodology to > our (possible) slight detriment here. > > At present, we need 7.3 to start, so that people can begin working on > it. > > We also need to have the 7.2 branch, so stuff for the to-be 7.2.x can be > included there where appropriate. > > So, lets get it branched, because that's what needs to be done here and > now, in this instance. > > Then people with a burn to do 7.3 stuff can do so, and things can be > added to 7.2.x where needed. [snip] Greg
I guess I jumped the gun in my previous message as I hadn't read this one yet. I completely agree with Marc on this topic. What is the logic for simply not snap-shoting what really is 7.2 and move on with 7.3 while allowing 7.2.x to occur on the branch for migration on a later day? Has this been problematic in the past? Greg On Fri, 2002-02-15 at 22:56, Marc G. Fournier wrote: > > We're planning on ... this is a 'sore point' that I've been arguing for > several releases yet ... I still feel the branch should be made *on > release date* and not several weeks later ... I think this is the first > release where we will do the branch prior to the minor being released, so, > I guess, as time goes on, we'll get closer to what I've wanted for ages :) > > > > On Sat, 16 Feb 2002, Justin Clift wrote: > > > "Marc G. Fournier" wrote: > > > > > > On Fri, 15 Feb 2002, Bruce Momjian wrote: > > > > > > > Tom Lane wrote: > > > > > "Marc G. Fournier" <scrappy@hub.org> writes: > > > > > > Okay, if the ecpg is the only issue, does everyone feel comfortable with a > > > > > > branch going in this evening? I'll do a v7.2.1 tar-ball up Sunday night > > > > > > based on the branch, with an announce going out on Monday? > > > > > > > > > > I don't think it's time for 7.2.1 quite yet; we should probably wait > > > > > another week or two to see what comes in. I just want to branch now... > > > > > > > > Agreed. Let's get everything we can into 7.2.1. We normally don't push > > > > out a minor this quickly unless we have a major thing to fix, which we > > > > don't. > > > > > > Right, and we never branch until we're ready for the first minor ... so > > > ... > > > > Hang on here. It sounds like we're following previous methodology to > > our (possible) slight detriment here. > > > > At present, we need 7.3 to start, so that people can begin working on > > it. > > > > We also need to have the 7.2 branch, so stuff for the to-be 7.2.x can be > > included there where appropriate. > > > > So, lets get it branched, because that's what needs to be done here and > > now, in this instance. > > > > Then people with a burn to do 7.3 stuff can do so, and things can be > > added to 7.2.x where needed. > > > > :-) > > > > Regards and best wishes, > > > > Justin Clift > > > > > > > > > > > > ---------------------------(end of broadcast)--------------------------- > > > TIP 6: Have you searched our list archives? > > > > > > http://archives.postgresql.org > > > > -- > > "My grandfather once told me that there are two kinds of people: those > > who work and those who take the credit. He told me to try to be in the > > first group; there was less competition there." > > - Indira Gandhi > > > > > ---------------------------(end of broadcast)--------------------------- > TIP 3: if posting/reading through Usenet, please send an appropriate > subscribe-nomail command to majordomo@postgresql.org so that your > message can get through to the mailing list cleanly
Greg Copeland <greg@CopelandConsulting.Net> writes: > I guess I jumped the gun in my previous message as I hadn't read this > one yet. I completely agree with Marc on this topic. What is the logic > for simply not snap-shoting what really is 7.2 and move on with 7.3 > while allowing 7.2.x to occur on the branch for migration on a later > day? If you don't branch immediately, then bugfixes only have to be done once. So it makes some sense to let things stabilize for a little while after a general release. -Doug -- Let us cross over the river, and rest under the shade of the trees. --T. J. Jackson, 1863
...
> Has this been problematic in the past?
No problem, but for most previous releases the primary developers were
tired after pushing for the release, and a modest code slowdown after
release was welcome by all. It usually lasted a month or less and
allowed us to focus on problem reports from the new release without
getting those lost in the overall noise of new development.
As others have pointed out, 7.2 dragged on longer than anyone expected,
and maybe at least some of the "release issues" were fixed before
release.
Hmm, maybe one reason that things dragged out was that we *didn't*
engage the new developers as much as we could have, otherwise y'all
would be tired too ;)
hth
- Thomas