Re: [sage-release] Re: Sage 8.1.rc4 released

[Dima Pasechnik]

While I agree that the Cygwin/Windows port is somewhat neglected in this 
current release cycle, how about a Cygwin patchbot/buildbot?

I 
see 
https://patchbot.sagemath.org/?machine=CYGWIN_NT-10.0=2017-04-01%2020:47=x86_64=2.8.0(0.309/5/3)=windows-10-prof=needs_review

but is it alive and kicking?


On Tuesday, December 5, 2017 at 5:24:19 PM UTC, Erik Bray wrote:
>
> On Tue, Dec 5, 2017 at 6:16 PM, Emmanuel Charpentier 
>  wrote: 
> > 
> > 
> > Le mardi 5 décembre 2017 18:03:19 UTC+1, Dima Pasechnik a écrit : 
> >> 
> >> The skylake wait was due to a bunch of active Sage users and devs 
> needing 
> >> it and asking for it. 
> >> 
> >> Perhaps we should aim at 8.1.1 with Cygwin fix? 
> > 
> > 
> > The attention to Cygwin is indeed  justified by the hope of getting 
> *new* 
> > Sage users (namely Windows 10 users). 
> > But I note that Trac#24121 has been positively reviewed for 4 weeks, 
> whereas 
> > Trac#21233 has got positive review "only" 13 days ago. 
>
> It originally had "positive review" 5 months ago, then Volker set it 
> back to "needs work" due to a seemingly-related test that was failing. 
> In fact as far as I can tell that test was not directly related to my 
> fix, and was already broken due to a random nature of it, and it just 
> happened to fail (possibly my changes made it more likely to fail due 
> to entropy :).  I immediately corrected this and set it back to "needs 
> review", and then nothing... 
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Erik Bray]

On Tue, Dec 5, 2017 at 6:16 PM, Emmanuel Charpentier
 wrote:
>
>
> Le mardi 5 décembre 2017 18:03:19 UTC+1, Dima Pasechnik a écrit :
>>
>> The skylake wait was due to a bunch of active Sage users and devs needing
>> it and asking for it.
>>
>> Perhaps we should aim at 8.1.1 with Cygwin fix?
>
>
> The attention to Cygwin is indeed  justified by the hope of getting *new*
> Sage users (namely Windows 10 users).
> But I note that Trac#24121 has been positively reviewed for 4 weeks, whereas
> Trac#21233 has got positive review "only" 13 days ago.

It originally had "positive review" 5 months ago, then Volker set it
back to "needs work" due to a seemingly-related test that was failing.
In fact as far as I can tell that test was not directly related to my
fix, and was already broken due to a random nature of it, and it just
happened to fail (possibly my changes made it more likely to fail due
to entropy :).  I immediately corrected this and set it back to "needs
review", and then nothing...

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Emmanuel Charpentier]

Le mardi 5 décembre 2017 18:03:19 UTC+1, Dima Pasechnik a écrit :
>
> The skylake wait was due to a bunch of active Sage users and devs needing 
> it and asking for it. 
>
> Perhaps we should aim at 8.1.1 with Cygwin fix?
>

The attention to Cygwin is indeed  justified by the hope of getting *new* 
Sage users (namely Windows 10 users).
But I note that Trac#24121  has 
been positively reviewed for 4 weeks, whereas Trac#21233 
 has got positive review "only" 13 
days ago.

--
Emmanuel Charpentier

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Erik Bray]

On Tue, Dec 5, 2017 at 6:05 PM, Emmanuel Charpentier
 wrote:
>
>
> Le mardi 5 décembre 2017 10:52:14 UTC+1, Jeroen Demeyer a écrit :
>>
>> On 2017-12-04 19:18, Emmanuel Charpentier wrote:
>> > I already reported
>> > 
>> > (twice
>> > 
>> > !) that 8.1.rc(whatever) won't build without Trac#24121
>>
>> If you find that ticket so important, why didn't you change its priority
>> to "blocker"?
>
>
> 'cause I' a newbie : I am aware that I do not know many things about the
> development process : I report what I see or can achieve, and let people
> (presumably) more knowledgeable than I am take the "right" decisions. I do
> not want to make of myself a bigger pain in the @$$ than I am already :-).
>
>>
>> I see that this was done very recently, but if you had
>> done that from the beginning, it would already have been merged.
>
>
> Indeed. But I had some difficulty to convince others that a change breaking
> the last update to Windows 10 was importent enough : I thank that the
> critical fact is that Microsoft **pushes** this change to users. teling "do
> not upgrade now" is useless...

In fairness the one issue we're discussing here only affects building
Sage from source, and so as far as I'm concerned does not impact
"users", and the people building Sage from source on Windows can be
counted by a dog.  That said, as a matter of principal, it should have
been fixed, and I don't understand how the release manager chooses
which "positive_review" tickets are good enough to merge, and which
ones can be ignored.  The "priority" tag obviously has some influence
but when the default priority is "major" that still doesn't seem too
meaningful.  In my mind this was not a "blocker", but still important
to fix, and it has been available to merge for some time.

In retrospect I should have supplied a buildbot on Cygwin; though in
the case of this issue a buildbot might not have even caught it since,
as Emmanuel pointed out, it depends heavily on what Windows updates
are installed.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Emmanuel Charpentier]

Le mardi 5 décembre 2017 10:52:14 UTC+1, Jeroen Demeyer a écrit :
>
> On 2017-12-04 19:18, Emmanuel Charpentier wrote: 
> > I already reported 
> >  
> > (twice 
> >  
> > !) that 8.1.rc(whatever) won't build without Trac#24121 
>
> If you find that ticket so important, why didn't you change its priority 
> to "blocker"?


'cause I' a newbie : I am aware that I do not know many things about the 
development process : I report what I see or can achieve, and let people 
(presumably) more knowledgeable than I am take the "right" decisions. I do 
not want to make of myself a bigger pain in the @$$ than I am already :-).
 

> I see that this was done very recently, but if you had 
> done that from the beginning, it would already have been merged. 
>

Indeed. But I had some difficulty to convince others that a change breaking 
the last update to Windows 10 was importent enough : I thank that the 
critical fact is that Microsoft **pushes** this change to users. teling "do 
not upgrade now" is useless...

--
Emmanuel Charpentier

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Erik Bray]

On Tue, Dec 5, 2017 at 10:52 AM, Jeroen Demeyer  wrote:
> On 2017-12-04 19:18, Emmanuel Charpentier wrote:
>>
>> I already reported
>> 
>> (twice
>> 
>> !) that 8.1.rc(whatever) won't build without Trac#24121
>
>
> If you find that ticket so important, why didn't you change its priority to
> "blocker"? I see that this was done very recently, but if you had done that
> from the beginning, it would already have been merged.

Also it's less about "so important" is it is not about releasing
software with broken tests (even in minor ways).  The build procedure
for the binaries ensures that all the tests pass--obviously this can
be bypassed if really needed but that's basically cheating.  To that
end we should also add a buildbot for Cygwin if that's how Volker
determines whether or not a release is ready.  I will work on that
(I've only since come to understand that the 'patchbots' as opposed to
'buildbots' are merely advisory).

Furthermore I can't release "SageMath 8.1" if it's not the source
release tagged "8.1".  At best I can release SageMath 8.1-1 or
somesuch to indicate a patch level.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Erik Bray]

On Tue, Dec 5, 2017 at 10:52 AM, Jeroen Demeyer  wrote:
> On 2017-12-04 19:18, Emmanuel Charpentier wrote:
>>
>> I already reported
>> 
>> (twice
>> 
>> !) that 8.1.rc(whatever) won't build without Trac#24121
>
>
> If you find that ticket so important, why didn't you change its priority to
> "blocker"? I see that this was done very recently, but if you had done that
> from the beginning, it would already have been merged.

I think directly e-mailing the release manager would also be "good
enough", but that was done and ignored *shrug*

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Erik Bray]

On Mon, Dec 4, 2017 at 10:49 PM, Nicolas M. Thiery
 wrote:
> On Sat, Dec 02, 2017 at 08:04:50AM -0800, Volker Braun wrote:
>> Unfortunately, Eric's happiness will have to wait for 8.2.beta like
>> all the other tickets ;-)
>
> It's not so much about Erik's happiness than good Cygwin support and
> therefore our Window's user happiness; a rather high priority in my
> view.
>
> Though I do care about Erik's happiness as well :-)
>
> And I also see the tension between release often and wait-until-ready
> schedules. At some point, we should have a discussion about what model
> we want for SageMath, and what prerequisites in terms of continuous
> integration this would take.

Plus the wait-until-ready did occur while waiting on the fix for
skylake.  In the meantime of that waiting the fixes I requested for
Windows were already sitting there waiting...

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Jeroen Demeyer]

On 2017-12-04 19:18, Emmanuel Charpentier wrote:

I already reported

(twice

!) that 8.1.rc(whatever) won't build without Trac#24121


If you find that ticket so important, why didn't you change its priority 
to "blocker"? I see that this was done very recently, but if you had 
done that from the beginning, it would already have been merged.


--
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Thierry]

Hi,

let me try to make everyone happy by pointing some obvious points: 

- realeasing and binary building are different concepts, not to be
  confused:

- "release often" is about source code development

  - there is a cost in delaying the release: the longer time Sage source
code stays in a frozen state, the longer get the list of positively
reviewed unmerged tickets, the more rebasing work there will be.

  - no windows user are expected to compile Sage from the released source
code (yet), they will not be affected by such source-code releases.

- "wait until ready" is about public distribution

  - the last windows binary is 8.0, among the 90% of the undergrad
students, only a small part might care about the latest features
introduced in 8.1.

  - the patch of #24121 is ready, nothing prevents to apply this ticket
when building 8.1 binary for windows, right after its release so that
windows user will not have to wait for 8.2 either.

- cygwin port could still be considered as not fully mature yet (those are
  not criticism, the work that was done already is very huge):

  - many optional packages can not be provided yet (i am not sure, please
correct me if i am wrong), they do not go to experimental.

  - sage-windows has only one release, that is labelled 0.2.

  - the last modification of psutil was done 8 months ago (#21598,
#22625), but #24121 was reported only 5 weeks ago (and set to blocker
9 hours ago)

  - the cygwin architecture is not the one used by developers, hence there
are not enough patchbots/buildbots/devs to test it on a regular basis,
and detect such issues earlier.

  - the single cygwin patchbot hardly succeeds to pass ticket 0.

Let me witness about the releasing of Sage Debian Live, which have a lot
of similarities with cygwin port:

  - its release depends on a single person
  - it is 32 bits (in order to run on a wider range of computers)
  - no dev runs a 32bit machine with all optional packages installed
  - there have been releases that i either had to skip, or to apply some
unmerged tickets by hand to let things work, sometimes much after the
source release, see e.g. [1,2]
  - i used to feel some frustration when some 32bit ticket was set to
not-a-blocker (i do not anymore, no idea why actually)
  - i did not release 8.0 even if Sage compilation was OK, i prefered to
be sure not to miss something in the migration jessie->stretch
  - i am now running a 32bit patchbot (and on the way to run an army of
them with different OSes and sets of optional packages installed)

Let me suggest some concrete things for improving the situation in the
future:

- we should have an explicit (and collectively discussed) list of what
  should be considered as blocker and what should not.

- this check-list should be all-green right before a rc0 is triggered.

- we should have more patchbots running weird architectures, and
  combinations of optional packages.

Ciao,
Thierry

[1] https://groups.google.com/forum/#!msg/sage-release/9isZshOxcjg/8lyCtK0NJAAJ
[2] https://groups.google.com/forum/#!msg/sage-release/1lpNhYM22v8/yFzpyO2GlfoJ



On Mon, Dec 04, 2017 at 10:49:56PM +0100, Nicolas M. Thiery wrote:
> On Sat, Dec 02, 2017 at 08:04:50AM -0800, Volker Braun wrote:
> > Unfortunately, Eric's happiness will have to wait for 8.2.beta like
> > all the other tickets ;-)
> 
> It's not so much about Erik's happiness than good Cygwin support and
> therefore our Window's user happiness; a rather high priority in my
> view.
> 
> Though I do care about Erik's happiness as well :-)
> 
> And I also see the tension between release often and wait-until-ready
> schedules. At some point, we should have a discussion about what model
> we want for SageMath, and what prerequisites in terms of continuous
> integration this would take.
> 
> Cheers,
>   Nicolas
> --
> Nicolas M. Thiéry "Isil" 
> http://Nicolas.Thiery.name/
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "sage-release" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-release+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-release@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-release.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Emmanuel Charpentier]

Le lundi 4 décembre 2017 15:16:39 UTC+1, Jeroen Demeyer a écrit :
>
> On 2017-12-04 14:35, Erik Bray wrote: 
> > I don't see why.  What's the harm in merging a couple working branches? 
>
> The first rc version of Sage 8.1 is now almost 1 month ago. We don't 
> want to delay the release of 8.1 forever. 
>

I have no problem waiting for the adoption of new or enhanced features. I 
have a problem with the release of a partially broken system.

Shouldn't we adopt, at least partially,  the Debian motto "it will be 
released when it is ready" ?

I already reported 
 (
twice 
 !) 
that 8.1.rc(whatever) won't build without Trac#24121 
 on Cygwin running on recent 
versions of Windows 10 (the "Fall Creators Update", which broke psutil). 
This version wiil be *_pervasive_*, Microsoft ***pushing*** this update to 
Windows 10 users (and pushing Windows 7 and 8 users towards Windows 10). 
Therefore, it seems to me that this fix is necessary, unless we drop 
Windows-under-Cygwin as a fully supported platform.

Which would be, IMSAO, a *_major_* strategic blunder : Windows 10 is (alas) 
the platform used by 90% of the undergrad students, engineers and 
non-mathematician potential Sage users, which should be our target of 
choice. And Erik's magic tarball is a *_hell_* of a lot easier to install 
for a newbie (i. e. 90% of our potential users) that our current VM 
contraption.

Similarly Erik has reported and fixed a bug that randomly break ptestlong. 
Should we release with a warning such as "such and such test may fail : 
don't break your pretty little head on it..." ?

I think not...

--
Emmanuel Charpentier

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Erik Bray]

On Mon, Dec 4, 2017 at 3:16 PM, Jeroen Demeyer  wrote:
> On 2017-12-04 14:35, Erik Bray wrote:
>>
>> I don't see why.  What's the harm in merging a couple working branches?
>
>
> The first rc version of Sage 8.1 is now almost 1 month ago. We don't want to
> delay the release of 8.1 forever.

???

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Jeroen Demeyer]

On 2017-12-04 14:35, Erik Bray wrote:

I don't see why.  What's the harm in merging a couple working branches?


The first rc version of Sage 8.1 is now almost 1 month ago. We don't 
want to delay the release of 8.1 forever.


--
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Re: Sage 8.1.rc4 released

[Erik Bray]

On Sat, Dec 2, 2017 at 5:04 PM, Volker Braun  wrote:
> On Saturday, December 2, 2017 at 1:54:02 PM UTC+1, tsc...@ucdavis.edu wrote:
>>
>> What about https://trac.sagemath.org/ticket/24297? I don't know if this
>> problem materializes on other platforms, but I know Erik will be happy with
>> all tests passing on Cygwin. ;)
>
>
> Unfortunately, Eric's happiness will have to wait for 8.2.beta like all the
> other tickets ;-)

I don't see why.  What's the harm in merging a couple working branches?

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[sage-release] Re: Sage 8.1.rc4 released

[Volker Braun]

We time it by actually connecting, not sending an ICMP packet for the 
record. So if you have a not-so-transparent proxy in front of you then its 
quite possible that the timings differ.

 

On Sunday, December 3, 2017 at 4:17:29 PM UTC+1, Ralf Stephan wrote:
>
> What could be the reason that for me the mirror search shows >5s ping for 
> many mirrors, also for switch.ch which is always 20ms from here (and a 
> manual ping confirms this)?
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[sage-release] Re: Sage 8.1.rc4 released

[Dima Pasechnik]

In Oxford, a.t.m. the ping to the usual UK Sage mirror in 5ms from the 
university network; faster than usual.

On Sunday, December 3, 2017 at 5:05:05 PM UTC, Eric Gourgoulhon wrote:
>
>
> Le dimanche 3 décembre 2017 16:17:29 UTC+1, Ralf Stephan a écrit :
>>
>> What could be the reason that for me the mirror search shows >5s ping for 
>> many mirrors, also for switch.ch which is always 20ms from here (and a 
>> manual ping confirms this)?
>>
>>
> Maybe it's some issue with your internet access. Here is what I get from 
> here (Paris):
>
> [patch-2.7.5] Searching fastest mirror
> [patch-2.7.5]   277ms: http://files.sagemath.org/
> [patch-2.7.5]   409ms: http://ftp.iitm.ac.in/sage/
> [patch-2.7.5]  1126ms: http://ftp.kaist.ac.kr/sage/
> [patch-2.7.5]   511ms: http://ftp.leg.uct.ac.za/pub/packages/sage/
> [patch-2.7.5]   480ms: http://ftp.ntua.gr/pub/sagemath/
> [patch-2.7.5]   951ms: http://ftp.riken.jp/sagemath/
> [patch-2.7.5]   926ms: http://ftp.tsukuba.wide.ad.jp/software/sage/
> [patch-2.7.5]   711ms: http://ftp.yz.yamagata-u.ac.jp/pub/math/sage/
> [patch-2.7.5]   921ms: http://linorg.usp.br/sage/
> [patch-2.7.5]   716ms: http://mirror.aarnet.edu.au/pub/sage/
> [patch-2.7.5]  1127ms: http://mirror.hust.edu.cn/sagemath/
> [patch-2.7.5]   242ms: http://mirror.switch.ch/mirror/sagemath/
> [patch-2.7.5]   676ms: http://mirror.ufs.ac.za/sagemath/
> [patch-2.7.5]   122ms: 
> http://mirror.yandex.ru/mirrors/sage.math.washington.edu/
> [patch-2.7.5]   286ms: http://mirrors-usa.go-parts.com/sage/sagemath/
> [patch-2.7.5]   151ms: http://mirrors.mit.edu/sage/
> [patch-2.7.5]  2101ms: http://mirrors.tuna.tsinghua.edu.cn/sagemath/
> [patch-2.7.5]  1535ms: http://mirrors.ustc.edu.cn/sagemath/
> [patch-2.7.5]   410ms: http://mirrors.xmission.com/sage/
> [patch-2.7.5]   268ms: http://sage.mirror.garr.it/mirrors/sage/
> [patch-2.7.5]  1061ms: http://sagemath.c3sl.ufpr.br/
> [patch-2.7.5]   613ms: http://sagemath.mirror.ac.za/
> [patch-2.7.5]   128ms: http://www-ftp.lip6.fr/pub/math/sagemath/
> [patch-2.7.5]   894ms: http://www.cecm.sfu.ca/sage/
> [patch-2.7.5]   107ms: 
> http://www.mirrorservice.org/sites/www.sagemath.org/
> [patch-2.7.5]   133ms: https://mirror.koddos.net/sagemath/
> [patch-2.7.5]   180ms: https://mirrors.up.pt/pub/sage/
> [patch-2.7.5] Fastest mirror: 
> http://www.mirrorservice.org/sites/www.sagemath.org/
>
> and ping mirror.swith.ch yields 35 ms.
>
> Best regards,
>
> Eric.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[sage-release] Re: Sage 8.1.rc4 released

[Eric Gourgoulhon]

Le dimanche 3 décembre 2017 16:17:29 UTC+1, Ralf Stephan a écrit :
>
> What could be the reason that for me the mirror search shows >5s ping for 
> many mirrors, also for switch.ch which is always 20ms from here (and a 
> manual ping confirms this)?
>
>
Maybe it's some issue with your internet access. Here is what I get from 
here (Paris):

[patch-2.7.5] Searching fastest mirror
[patch-2.7.5]   277ms: http://files.sagemath.org/
[patch-2.7.5]   409ms: http://ftp.iitm.ac.in/sage/
[patch-2.7.5]  1126ms: http://ftp.kaist.ac.kr/sage/
[patch-2.7.5]   511ms: http://ftp.leg.uct.ac.za/pub/packages/sage/
[patch-2.7.5]   480ms: http://ftp.ntua.gr/pub/sagemath/
[patch-2.7.5]   951ms: http://ftp.riken.jp/sagemath/
[patch-2.7.5]   926ms: http://ftp.tsukuba.wide.ad.jp/software/sage/
[patch-2.7.5]   711ms: http://ftp.yz.yamagata-u.ac.jp/pub/math/sage/
[patch-2.7.5]   921ms: http://linorg.usp.br/sage/
[patch-2.7.5]   716ms: http://mirror.aarnet.edu.au/pub/sage/
[patch-2.7.5]  1127ms: http://mirror.hust.edu.cn/sagemath/
[patch-2.7.5]   242ms: http://mirror.switch.ch/mirror/sagemath/
[patch-2.7.5]   676ms: http://mirror.ufs.ac.za/sagemath/
[patch-2.7.5]   122ms: 
http://mirror.yandex.ru/mirrors/sage.math.washington.edu/
[patch-2.7.5]   286ms: http://mirrors-usa.go-parts.com/sage/sagemath/
[patch-2.7.5]   151ms: http://mirrors.mit.edu/sage/
[patch-2.7.5]  2101ms: http://mirrors.tuna.tsinghua.edu.cn/sagemath/
[patch-2.7.5]  1535ms: http://mirrors.ustc.edu.cn/sagemath/
[patch-2.7.5]   410ms: http://mirrors.xmission.com/sage/
[patch-2.7.5]   268ms: http://sage.mirror.garr.it/mirrors/sage/
[patch-2.7.5]  1061ms: http://sagemath.c3sl.ufpr.br/
[patch-2.7.5]   613ms: http://sagemath.mirror.ac.za/
[patch-2.7.5]   128ms: http://www-ftp.lip6.fr/pub/math/sagemath/
[patch-2.7.5]   894ms: http://www.cecm.sfu.ca/sage/
[patch-2.7.5]   107ms: http://www.mirrorservice.org/sites/www.sagemath.org/
[patch-2.7.5]   133ms: https://mirror.koddos.net/sagemath/
[patch-2.7.5]   180ms: https://mirrors.up.pt/pub/sage/
[patch-2.7.5] Fastest mirror: 
http://www.mirrorservice.org/sites/www.sagemath.org/

and ping mirror.swith.ch yields 35 ms.

Best regards,

Eric.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[sage-release] Re: Sage 8.1.rc4 released

[Ralf Stephan]

What could be the reason that for me the mirror search shows >5s ping for 
many mirrors, also for switch.ch which is always 20ms from here (and a 
manual ping confirms this)?

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[sage-release] Re: Sage 8.1.rc4 released

[Eric Gourgoulhon]

With Ubuntu 16.04 x86_64 running on Intel Core i7-6700HQ + 16 GB RAM, from 
a fresh git clone + pull develop, parallel (-j8) build OK and make 
ptestlong successful ("All tests passed!").

Eric.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[sage-release] Re: Sage 8.1.rc4 released

[Andrew]

Builds on a macbook pro with a *skylake* chip running 10.12.6. All tests 
pass, although there were 22 warnings for slow tests (can post details if 
wanted). Thanks Dima, Jeroen and everyone else who contributed to #24085.

Andrew


On Saturday, 2 December 2017 12:23:02 UTC+1, Volker Braun wrote:
>
> Against all odds the mpir patch finally materialized (thanks, Dima!). This 
> should now really be the final release candidate, please give it a test!
>
> As always, you can get the latest beta version from the "develop" git 
> branch. Alternatively, the self-contained source tarball is at 
> http://www.sagemath.org/download-latest.html
>
> 6453938218 (tag: 8.1.rc4) Updated SageMath version to 8.1.rc4
> 8506011b59 Trac #24085: mpir-3.0.0 does not compile on skylake chips on OSX
> 20007d522c (tag: 8.1.rc3) Updated SageMath version to 8.1.rc3
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

[sage-release] Re: Sage 8.1.rc4 released

[tscrim]

What about https://trac.sagemath.org/ticket/24297? I don't know if this 
problem materializes on other platforms, but I know Erik will be happy with 
all tests passing on Cygwin. ;)

Best,
Travis

PS - I saw that https://trac.sagemath.org/ticket/24085 was not closed.


On Saturday, December 2, 2017 at 5:23:02 AM UTC-6, Volker Braun wrote:
>
> Against all odds the mpir patch finally materialized (thanks, Dima!). This 
> should now really be the final release candidate, please give it a test!
>
> As always, you can get the latest beta version from the "develop" git 
> branch. Alternatively, the self-contained source tarball is at 
> http://www.sagemath.org/download-latest.html
>
> 6453938218 (tag: 8.1.rc4) Updated SageMath version to 8.1.rc4
> 8506011b59 Trac #24085: mpir-3.0.0 does not compile on skylake chips on OSX
> 20007d522c (tag: 8.1.rc3) Updated SageMath version to 8.1.rc3
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.