Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-09 Thread Martin v. Löwis
Am 08.01.14 16:03, schrieb Nick Coghlan:
 On 9 January 2014 00:43, Bob Hanson d2mp...@newsguy.com wrote:
 When I read this comment of yours, Guido, I immediately started
 wondering about this. You may well be right -- indeed, I have a
 very old install (c.2007) which has not been updated (other than
 one or three new MS drivers).

 Perhaps the Python 3.4.0b2 MSI installer uses a new capability,
 which, as you say, causes the installer to at least attempt to
 upgrade...?
 
 I believe the pip bootstrapping involves an MSI feature we haven't
 previously used (MvL would be able to confirm). If so, then MSI may be
 looking for a new version to interpret that new setting.

That's not true. The pip bootstrapping uses a custom action, and
we already have one that is similar (compile to pyc), although that
isn't run by default.

My guess is that it might try verifying signatures, and somehow tries
to obtain the CA certificates (although it's puzzling that it would get
them from akamai - perhaps MS is hosting the CA bundle there).

Regards,
Martin

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-09 Thread Martin v. Löwis
Am 06.01.14 17:26, schrieb Michael Urman:
 Here's some more guesswork. Does it seem possible that msiexec is
 trying to verify the revocation status of the certificate used to sign
 the python .msi file? Per
 http://blogs.technet.com/b/pki/archive/2006/11/30/basic-crl-checking-with-certutil.aspx
 it looks like crl.microsoft.com is the host; this is hosted on akamai:
crl.microsoft.com is an alias for crl.www.ms.akadns.net.
crl.www.ms.akadns.net is an alias for a1363.g.akamai.net.

I think that could be close. The MSI file has two signatures in it: the
PSF code signing signature, and a Verisign timestamping signature.

For the PSF certificate, the CRL is at csc3-2010-crl.verisign.com,
which is (here) a CNAME for crl.ws.symantec.com.edgekey.net, which
in turn is a CNAME for e6845.ce.akamaiedge.net.

The timestamping signature has its CRL at ts-crl.ws.symantec.com,
which is a CNAME for crl.ws.symantec.com.edgekey.net again.

So the most plausible reason is indeed that it tries to download
CRLs, though not Microsoft ones, but Verisign/Symantic ones.

Regards,
Martin


___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-08 Thread Bob Hanson
[Top-post fixed (use-case is an exception to the GvR rule ;-) )
and some attributions restored with my additional comments
following for the ease of future readers.]

TL;DR: Outbound-connection attempts seem to be happening only to
me, therefore, most likely not a Python problem -- but some
problem at my end. Thanks to all.

On Mon, 6 Jan 2014 05:43:38 -1000, Guido van Rossum wrote:

 On Mon, Jan 6, 2014 at 5:29 AM, Bob Hanson d2mp...@newsguy.com wrote:

  [For the record: I'm running 32bit Windows XP (Pro) SP2 and
  installing for all users.]
 
  TL;DR: No matter what I tried this morning re uninstalling and
  reinstalling 3.4.0b2, pip or no pip, MSI still tried to connect
  to the Akamai URLs.
 
  On Sun, 05 Jan 2014 23:06:49 -0500, R. David Murray wrote:
 
   On Sun, 05 Jan 2014 19:32:15 -0800, Bob Hanson wrote:
   
Still wondering why [...] msiexec.exe [is] trying to connect out while
installing 3.4.0b2 from my harddrive...?
  
   The ensurepip developers will have to say for sure, but my understanding
   is that it does *not* go out to the network.  On the other hand, it is
   conceivable that pip 1.5, unlike the earlier version in Beta1, is doing
   some sort of up to date check that it shouldn't be doing in the
   ensurepip scenario.
  
   I presume you did have the installer install pip.
  
  To be honest, I forgot all about pip [...] didn't
  even notice a checkbox for that option.
 
   If you haven't already, You might try reinstalling and unchecking
   that option, and see if it msiexec still tries to go out to the
   network.  That would confirm it is ensurepip that is the issue
   (although that does seem most likely).
 
  [...snip synopsis of various uninstall-reinstall dances...]
 
  So, whatever I have tried -- pip or no pip -- msiexec.exe still
  attempts to connect to those Akamai URLs.
 
 Since MSIEXEC.EXE is a legit binary (not coming from our packager) and
 Akamai is a legitimate company (MS most likely has an agreement with
 them), at this point I would assume that there's something that
 MSIEXEC.EXE wants to get from Akamai, which is unintentionally but
 harmlessly triggered by the Python install. Could it be checking for
 upgrades?

When I read this comment of yours, Guido, I immediately started
wondering about this. You may well be right -- indeed, I have a
very old install (c.2007) which has not been updated (other than
one or three new MS drivers). 

Perhaps the Python 3.4.0b2 MSI installer uses a new capability,
which, as you say, causes the installer to at least attempt to
upgrade...?

In any event, as there's been no other reports, this seems to be
something happening only to me. As such, it seems to be not a
Python problem, but some misconfiguration on my own system, say.

If I retain interest in investigating this, and if I *do* find an
actual problem with Python, I'll post again.

Thanks go to you, Guido, as well as to Tim and all the others who
helped me with this.

Regards,
Bob Hanson

-- 
Write once, read many.

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-08 Thread Nick Coghlan
On 9 January 2014 00:43, Bob Hanson d2mp...@newsguy.com wrote:
 When I read this comment of yours, Guido, I immediately started
 wondering about this. You may well be right -- indeed, I have a
 very old install (c.2007) which has not been updated (other than
 one or three new MS drivers).

 Perhaps the Python 3.4.0b2 MSI installer uses a new capability,
 which, as you say, causes the installer to at least attempt to
 upgrade...?

I believe the pip bootstrapping involves an MSI feature we haven't
previously used (MvL would be able to confirm). If so, then MSI may be
looking for a new version to interpret that new setting.

Cheers,
Nick.

-- 
Nick Coghlan   |   ncogh...@gmail.com   |   Brisbane, Australia
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-06 Thread Bob Hanson
[For the record: I'm running 32bit Windows XP (Pro) SP2 and
installing for all users.]

TL;DR: No matter what I tried this morning re uninstalling and
reinstalling 3.4.0b2, pip or no pip, MSI still tried to connect
to the Akamai URLs.

On Sun, 05 Jan 2014 23:06:49 -0500, R. David Murray wrote:

  Still wondering why, all of a sudden after years of using a
  firewalled msiexec.exe, I get it now trying to connect out while
  installing 3.4.0b2 from my harddrive...?
 
 The ensurepip developers will have to say for sure, but my understanding
 is that it does *not* go out to the network.  On the other hand, it is
 conceivable that pip 1.5, unlike the earlier version in Beta1, is doing
 some sort of up to date check that it shouldn't be doing in the
 ensurepip scenario.
 
 I presume you did have the installer install pip.

To be honest, I forgot all about pip until after I become a wee
bit alarmed by the installer going out to the interwebs -- didn't
even notice a checkbox for that option.

 If you haven't already, You might try reinstalling and unchecking
 that option, and see if it msiexec still tries to go out to the
 network.  That would confirm it is ensurepip that is the issue
 (although that does seem most likely).

Working again on this, this morning: Uninstalled and then
reinstalled 3.4.0b2. No check box for pip, but there was that
strange tree of collapsed options which included a pip one and
which appeared to default to install pip. Left pip on as was
the default.

Curiously, although I hadn't remembered this happening yesterday,
this morning, the installer said there was already a 3.4x
installed and do I want to overwrite it. (Uninstalling 3.4.0b2
had left behind my addition of sympy and another file or two of
my own from 3.4.0b1.) I told MSI to go ahead and overwrite.

Sure enough, the installer tried to connect to the same two IPs
(Akamai I'm now told) with the installer left at default options.

---

Next, I uninstalled 3.4.0b2 again, this time removing the entire
dir after uninstalling. Reinstalling (still default settings on
installer) this time gave me a bunch of *new* additions to
site-packages including pip, setuptools, easy_install.py -- all
of which were *not* installed into site-packages when priorly
overinstalling. 

Again, this time, msiexec.exe still attempted to connect (two
bursts -- each time, twice to each of the aforementioned URLs).

---

Finally, I uninstalled 3.4.0b2, removed the dir, and reinstalled
yet again, this time selecting the don't install pip option in
the funky Windows option tree in the MSI installer. 

Yet *still* again, there were several sets of attempts by
msiexec.exe to connect to the same two Akamai URLs -- but, no pip
or other cruft in site-packages nor any pip-things in Scripts
after the install finished.

So, whatever I have tried -- pip or no pip -- msiexec.exe still
attempts to connect to those Akamai URLs.

[Hopefully, I kept accurate notes this morning and didn't typo
them above.]

At any rate, the attempts to connect to the network seem like
undesirable behavior to this man. If pip is necessary, then some
Window users may well end up without it -- and then not know why
something later doesn't work.

Got to go now, but will check in this evening or in the morning,
tomorrow.

Again, many thanks to all for the help --

Bob Hanson

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-06 Thread Paul Moore
On 6 January 2014 15:29, Bob Hanson d2mp...@newsguy.com wrote:
 At any rate, the attempts to connect to the network seem like
 undesirable behavior to this man. If pip is necessary, then some
 Window users may well end up without it -- and then not know why
 something later doesn't work.

I have installed python 3.4b2 on Windows (7, 64-bit) and seen no
network connections like this. I didn't check too closely, and I don't
know that my (corporate) firewall would necessarily report this to me.
But it seemed fine to me. I'll see if I can try an install on a VM
with no network access at some point, and see what that does.

One possibility which might be worth investigating - some Windows
software can insert itself into the network stack and trigger extra
net calls (I believe it's common with things like parental control
software, and I once ended up with a thoroughly broken network because
ZoneAlarm did something nasty to me). As no-one else seems to be
having the issues you are, could it be that something else is
intercepting part of the install process, unrelated to Python?

It's also worth noting that the Python MSI is just a database of
files and settings to install (plus some post-install scripts that
would behave the same on all systems, and don't connect to the net
AIUI). The MSI is interpreted, as you note, by the OS-supplied
msiexec.exe. Is it possible that you have some sort of patched msiexec
(there's lots of opportunity for OEM customisation in Windows, maybe
your hardware supplier added something to get a logo/advert from their
website when installs run)?

I'm clutching at straws here, certainly, but it does look like it's an
issue local to your setup.
Paul
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-06 Thread Michael Urman
On Mon, Jan 6, 2014 at 9:43 AM, Guido van Rossum gu...@python.org wrote:
 Since MSIEXEC.EXE is a legit binary (not coming from our packager) and
 Akamai is a legitimate company (MS most likely has an agreement with
 them), at this point I would assume that there's something that
 MSIEXEC.EXE wants to get from Akamai, which is unintentionally but
 harmlessly triggered by the Python install. Could it be checking for
 upgrades?

Here's some more guesswork. Does it seem possible that msiexec is
trying to verify the revocation status of the certificate used to sign
the python .msi file? Per
http://blogs.technet.com/b/pki/archive/2006/11/30/basic-crl-checking-with-certutil.aspx
it looks like crl.microsoft.com is the host; this is hosted on akamai:
   crl.microsoft.com is an alias for crl.www.ms.akadns.net.
   crl.www.ms.akadns.net is an alias for a1363.g.akamai.net.

There are various things you could try to verify this. You could test
with simpler .msi files where one is signed and another is not signed
(I'll leave it up to you to find such things, but ORCA is a common
test .msi file). Or you could take a verbose log of the installation
process (msiexec /l*v python.log python.msi OR
http://support.microsoft.com/kb/223300), sit on the prompt for network
access so you can uniquely identify the log's timestamps, and try to
identify at what point of the installation the network access occurs.
Once that is known, more steps can be taken to identify and resolve
any actual issues.

Michael
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


[Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Larry Hastings


On behalf of the Python development team, I'm pleased to announce
the second beta release of Python 3.4.

This is a preview release, and its use is not recommended for
production settings.

Python 3.4 includes a range of improvements of the 3.x series, including
hundreds of small improvements and bug fixes.  Major new features and
changes in the 3.4 release series include:

* PEP 428, a pathlib module providing object-oriented filesystem paths
* PEP 435, a standardized enum module
* PEP 436, a build enhancement that will help generate introspection
   information for builtins
* PEP 442, improved semantics for object finalization
* PEP 443, adding single-dispatch generic functions to the standard library
* PEP 445, a new C API for implementing custom memory allocators
* PEP 446, changing file descriptors to not be inherited by default
   in subprocesses
* PEP 450, a new statistics module
* PEP 451, standardizing module metadata for Python's module import system
* PEP 453, a bundled installer for the *pip* package manager
* PEP 454, a new tracemalloc module for tracing Python memory allocations
* PEP 456, a new hash algorithm for Python strings and binary data
* PEP 3154, a new and improved protocol for pickled objects
* PEP 3156, a new asyncio module, a new framework for asynchronous I/O

Python 3.4 is now in feature freeze, meaning that no new features will be
added.  The final release is projected for late February 2014.


To download Python 3.4.0b2 visit:

http://www.python.org/download/releases/3.4.0/


Please consider trying Python 3.4.0b2 with your code and reporting any
new issues you notice to:

 http://bugs.python.org/


Enjoy!

--
Larry Hastings, Release Manager
larry at hastings.org
(on behalf of the entire python-dev team and 3.4's contributors)
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Bob Hanson
On Sun, 05 Jan 2014 13:20:50 -0800, Larry Hastings wrote:

 On behalf of the Python development team, I'm pleased to announce
 the second beta release of Python 3.4.

Thanks, Larry and all the devs, your hard work is appreciated.

However, why does this new version look like adware or other
malware when installing?

This is the first time I ever installed a version of Python which
caused something called MSIEXEC.EXE to try to access some
commercial dot-com site.

Naturally, my firewall stopped it, but what's going on? 

(A command prompt Windows box opened, followed by my firewall
firing off a set of warnings about having blocked attempted
unauthorized outbound connections to blah-blah-blah-dot-com.) 

When I attempted to run the python interpreter it appeared to
open normally, but I haven't tried doing anything with it.

If this has to do with ensure_pip or whatever it's called,
perhaps some other solution is called for which is more
user-friendly and not as likely to incite unease and mistrust by
attempting to silently access a commercial site while Python is
installing. At the very least a warning seems to be called for
(possibly along with an opt-out). Also, a more friendly site
(python.org?) to connect to would help to restore some faith.

If it *is* about the pip thing, I can only imagine the
frustration of Window users: Having to reconfigure firewalls to
(properly?) upgrade Python, or ignoring the firewall-warning
glitch assuming it's okay. Or, possibly, a silent failure of
something if they're firewalled with warnings turned off?

If it's not pip, what is it?

Finally, is my install now broken? Are offline installs now not
possible. Does one now need an always-on internet connection to
use Python?

Thanks --

Bob Hanson

-- 
Snowden reveals that George Orwell was an extreme optimist.

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Benjamin Peterson


-- 
Regards,
Benjamin

On Sun, Jan 5, 2014, at 04:19 PM, Bob Hanson wrote:
 On Sun, 05 Jan 2014 13:20:50 -0800, Larry Hastings wrote:
 
  On behalf of the Python development team, I'm pleased to announce
  the second beta release of Python 3.4.
 
 Thanks, Larry and all the devs, your hard work is appreciated.
 
 However, why does this new version look like adware or other
 malware when installing?
 
 This is the first time I ever installed a version of Python which
 caused something called MSIEXEC.EXE to try to access some
 commercial dot-com site.
 
 Naturally, my firewall stopped it, but what's going on? 
 
 (A command prompt Windows box opened, followed by my firewall
 firing off a set of warnings about having blocked attempted
 unauthorized outbound connections to blah-blah-blah-dot-com.) 

Well, what is blah-blah-blah.com exactly?

 
 When I attempted to run the python interpreter it appeared to
 open normally, but I haven't tried doing anything with it.
 
 If this has to do with ensure_pip or whatever it's called,
 perhaps some other solution is called for which is more
 user-friendly and not as likely to incite unease and mistrust by
 attempting to silently access a commercial site while Python is
 installing. At the very least a warning seems to be called for
 (possibly along with an opt-out). Also, a more friendly site
 (python.org?) to connect to would help to restore some faith.
 
 If it *is* about the pip thing, I can only imagine the
 frustration of Window users: Having to reconfigure firewalls to
 (properly?) upgrade Python, or ignoring the firewall-warning
 glitch assuming it's okay. Or, possibly, a silent failure of
 something if they're firewalled with warnings turned off?
 
 If it's not pip, what is it?
 
 Finally, is my install now broken? Are offline installs now not
 possible. Does one now need an always-on internet connection to
 use Python?
 
 Thanks --
 
 Bob Hanson
 
 -- 
 Snowden reveals that George Orwell was an extreme optimist.
 
 ___
 Python-Dev mailing list
 Python-Dev@python.org
 https://mail.python.org/mailman/listinfo/python-dev
 Unsubscribe:
 https://mail.python.org/mailman/options/python-dev/benjamin%40python.org
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Tim Peters
[Benjamin Peterson]
 ...
 This is the first time I ever installed a version of Python which
 caused something called MSIEXEC.EXE

msiexec.exe is not part of the Python download..  msiexec.exe is part
of the Windows operating system, and is precisely the program that
installs .msi files (which the Python installer is).

 to try to access some commercial dot-com site.

 Naturally, my firewall stopped it, but what's going on?

Possible:  you have a virus or trojan that replaced the system
msiexec.exe with its own malware.  Run a full virus scan ASAP, try
Malwarebyte's anti-malware program, etc etc etc.
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Bob Hanson
[Bob Hanson]
  This is the first time I ever installed a version of Python which
  caused something called MSIEXEC.EXE

[Tim Peters]
 msiexec.exe is not part of the Python download..  msiexec.exe is part
 of the Windows operating system, and is precisely the program that
 installs .msi files (which the Python installer is).

That is correct. ;-)

[Bob Hanson]
  to try to access some commercial dot-com site.
 
  Naturally, my firewall stopped it, but what's going on?

[Tim Peters]
 Possible:  you have a virus or trojan that replaced the system
 msiexec.exe with its own malware.  Run a full virus scan ASAP, try
 Malwarebyte's anti-malware program, etc etc etc.

Didn't think this likely, but I have now quintuple-checked
everything again. Everything says I have the real McCoy
msiexec.exe in its proper location -- just upgraded another app
which used MSI installers and it went as per normal.

I'm presuming, still, that it is something to do with the ensure
that pip is present on Windows thing? [See the bottom of my
original post.]

Do Nick or Martin (or other dev) have any comments?

Bob Hanson

-- 
Sent from my Smart shoephone.

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Tim Peters
[Bob Hanson]
 ...
 Didn't think this likely, but I have now quintuple-checked
 everything again. Everything says I have the real McCoy
 msiexec.exe in its proper location -- just upgraded another app
 which used MSI installers and it went as per normal.

That sounds most likely to me too ;-)


 I'm presuming, still, that it is something to do with the ensure
 that pip is present on Windows thing? [See the bottom of my
 original post.]

As Benjamin asked, could you please flesh out what
blah-blah-blah-dot-com means - what, exactly, was the site your
firewall warned you about?  My firewall didn't complain when I
installed 3.4.0b2 on Windows.
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Ethan Furman

On 01/05/2014 06:02 PM, Bob Hanson wrote:


I'm presuming, still, that it is something to do with the ensure
that pip is present on Windows thing?


Perhaps you could help us out by telling us what site was trying to be accessed?

--
~Ethan~
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Bob Hanson
On Sun, 5 Jan 2014 20:09:23 -0600, Tim Peters wrote:

 As Benjamin asked, could you please flesh out what
 blah-blah-blah-dot-com means - what, exactly, was the site your
 firewall warned you about? 

Forgive me, but I'm an old man with very poor vision. Using my
magnifying glass, I see it is two very long URLs ending with
something like after the blah-blah:  ... akametechnology.com

More precisely, these two IP addresses:
23.59.190.113:80
23.59.190.106:80

 My firewall didn't complain when I installed 3.4.0b2 on Windows.

I don't use the Windows firewall, and I have mine sent to block
all apps connecting in or out unless I make specific rules for
them. I have never authorized C:/Windows/System32/msiexec.exe to
connect in or out (and it didn't ;-) ).

Bob Hanson

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Tim Peters
[Bob Hanson]
 Forgive me, but I'm an old man with very poor vision. Using my
 magnifying glass, I see it is two very long URLs ending with
 something like after the blah-blah:  ... akametechnology.com

 More precisely, these two IP addresses:
 23.59.190.113:80
 23.59.190.106:80

So:

C:\Codeping -a  23.59.190.113

Pinging a23-59-190-113.deploy.static.akamaitechnologies.com
[23.59.190.113] with 32 bytes of data:
...
C:\Codeping -a  23.59.190.106

Pinging a23-59-190-106.deploy.static.akamaitechnologies.com
[23.59.190.106] with 32 bytes of data:

So it's just Akamai caching content.  Common as mud.  Can't say
specifically what was being cached, but it _could_ be that your ISP
contracts with Akamai.


 My firewall didn't complain when I installed 3.4.0b2 on Windows.

 I don't use the Windows firewall, and I have mine sent to block
 all apps connecting in or out unless I make specific rules for
 them. I have never authorized C:/Windows/System32/msiexec.exe to
 connect in or out (and it didn't ;-) ).

Same here on both counts.  We're getting nowhere with admirable speed ;-)
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Bob Hanson
On Sun, 5 Jan 2014 21:09:53 -0600, Tim Peters wrote:

 [Bob Hanson]
  Forgive me, but I'm an old man with very poor vision. Using my
  magnifying glass, I see it is two very long URLs ending with
  something like after the blah-blah:  ... akametechnology.com
 
  More precisely, these two IP addresses:
  23.59.190.113:80
  23.59.190.106:80
 
 So:
 
 C:\Codeping -a  23.59.190.113
 
 Pinging a23-59-190-113.deploy.static.akamaitechnologies.com
 [23.59.190.113] with 32 bytes of data:

 [...]

 So it's just Akamai caching content.  Common as mud.  Can't say
 specifically what was being cached, but it _could_ be that your ISP
 contracts with Akamai.

Still not following *why* this should be happening. I was
installing from my harddrive -- nothing needed to be cached as
far as I was concerned. Indeed, I would normally think I could
install while offline -- and often do on my PCs which are
air-gapped.

Still wondering why, all of a sudden after years of using a
firewalled msiexec.exe, I get it now trying to connect out while
installing 3.4.0b2 from my harddrive...?

   My firewall didn't complain when I installed 3.4.0b2 on Windows.
 
  I don't use the Windows firewall, and I have mine sent to block
  all apps connecting in or out unless I make specific rules for
  them. I have never authorized C:/Windows/System32/msiexec.exe to
  connect in or out (and it didn't ;-) ).
 
 Same here on both counts.  We're getting nowhere with admirable speed ;-)

So, we just need to make some distance -- our speed is good? ;-)

Less non-seriously, thanks for all the help, Tim (and others) --

Bob Hanson

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread R. David Murray
On Sun, 05 Jan 2014 19:32:15 -0800, Bob Hanson d2mp...@newsguy.com wrote:
 On Sun, 5 Jan 2014 21:09:53 -0600, Tim Peters wrote:
  So it's just Akamai caching content.  Common as mud.  Can't say
  specifically what was being cached, but it _could_ be that your ISP
  contracts with Akamai.
 
 Still not following *why* this should be happening. I was
 installing from my harddrive -- nothing needed to be cached as
 far as I was concerned. Indeed, I would normally think I could
 install while offline -- and often do on my PCs which are
 air-gapped.
 
 Still wondering why, all of a sudden after years of using a
 firewalled msiexec.exe, I get it now trying to connect out while
 installing 3.4.0b2 from my harddrive...?

The ensurepip developers will have to say for sure, but my understanding
is that it does *not* go out to the network.  On the other hand, it is
conceivable that pip 1.5, unlike the earlier version in Beta1, is doing
some sort of up to date check that it shouldn't be doing in the
ensurepip scenario.

I presume you did have the installer install pip.  If you haven't
already, You might try reinstalling and unchecking that option, and see
if it msiexec still tries to go out to the network.  That would confirm
it is ensurepip that is the issue (although that does seem most likely).

--David
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Donald Stufft
ensurepip uses —no-index so it shouldn’t be hitting the network at all.

On Jan 5, 2014, at 11:06 PM, R. David Murray rdmur...@bitdance.com wrote:

 On Sun, 05 Jan 2014 19:32:15 -0800, Bob Hanson d2mp...@newsguy.com wrote:
 On Sun, 5 Jan 2014 21:09:53 -0600, Tim Peters wrote:
 So it's just Akamai caching content.  Common as mud.  Can't say
 specifically what was being cached, but it _could_ be that your ISP
 contracts with Akamai.
 
 Still not following *why* this should be happening. I was
 installing from my harddrive -- nothing needed to be cached as
 far as I was concerned. Indeed, I would normally think I could
 install while offline -- and often do on my PCs which are
 air-gapped.
 
 Still wondering why, all of a sudden after years of using a
 firewalled msiexec.exe, I get it now trying to connect out while
 installing 3.4.0b2 from my harddrive...?
 
 The ensurepip developers will have to say for sure, but my understanding
 is that it does *not* go out to the network.  On the other hand, it is
 conceivable that pip 1.5, unlike the earlier version in Beta1, is doing
 some sort of up to date check that it shouldn't be doing in the
 ensurepip scenario.
 
 I presume you did have the installer install pip.  If you haven't
 already, You might try reinstalling and unchecking that option, and see
 if it msiexec still tries to go out to the network.  That would confirm
 it is ensurepip that is the issue (although that does seem most likely).
 
 --David
 ___
 Python-Dev mailing list
 Python-Dev@python.org
 https://mail.python.org/mailman/listinfo/python-dev
 Unsubscribe: 
 https://mail.python.org/mailman/options/python-dev/donald%40stufft.io


-
Donald Stufft
PGP: 0x6E3CBCE93372DCFA // 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA



signature.asc
Description: Message signed with OpenPGP using GPGMail
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Stephen J. Turnbull
Bob Hanson writes:
  On Sun, 5 Jan 2014 20:09:23 -0600, Tim Peters wrote:
  
   As Benjamin asked, could you please flesh out what
   blah-blah-blah-dot-com means - what, exactly, was the site your
   firewall warned you about? 
  
  Forgive me, but I'm an old man with very poor vision. Using my
  magnifying glass, I see it is two very long URLs ending with
  something like after the blah-blah:  ... akametechnology.com

I suppose you tried cutting and pasting?  Note that you don't need to
be exact as long as you're pretty sure you got the whole thing -- your
readers who have better eyesight can parse out the URL easily enough.

  More precisely, these two IP addresses:
  23.59.190.113:80
  23.59.190.106:80

Somebody who doesn't know the rules of capitalization (see
ww1.akamitechnologies.com) appears to be spoofing Akamai (the web
caching/distribution service used by President Obama among other
prominent users).

The domain referenced is presumably some variation on
IP address.deploy.static.akamitechnologies.com (according to host IP),
and the long URL is rooted at /ses/ so it's trying to convince you
it's a session (whether that is actually true or not I don't know,
that's just what I would guess if I were trying to reverse engineer an
honest URL, which this sure doesn't seem to be).

So your alarm seems to be verified, but why this happened to a Python
download I don't know.  It could be DNS hacking between you and
python.org, as well as something in the Python MSI.

HTH

Steve
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Tim Peters
[Bob Hanson]
 ... magnifying glass, I see it is two very long URLs ending with
 something like after the blah-blah:  ... akametechnology.com

[Stephen J. Turnbull]
 I suppose you tried cutting and pasting?  Note that you don't need to
 be exact as long as you're pretty sure you got the whole thing -- your
 readers who have better eyesight can parse out the URL easily enough.

I don't think this was cut 'n paste.  Looking up the IP addresses
returns legit Akamai URLs:

 More precisely, these two IP addresses:
 23.59.190.113:80
 23.59.190.106:80

C:\Codeping -a 23.59.190.113

Pinging a23-59-190-113.deploy.static.akamaitechnologies.com
[23.59.190.113] with 32 bytes of data:
...
C:\Codeping -a 23.59.190.106

Pinging a23-59-190-106.deploy.static.akamaitechnologies.com
[23.59.190.106] with 32 bytes of data:
...

Bob's  ... akametechnology.com just looks like compounded typos.

 ...
 So your alarm seems to be verified, but why this happened to a Python
 download I don't know.  It could be DNS hacking between you and
 python.org, as well as something in the Python MSI.

Honestly, for all we _know_, this firewall alert may have been
triggered by some other program that just happened to wake up while
Bob was installing Python.  Sure, that's unlikely.  But so is
everything else about this ;-)
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Christian Heimes

On 06.01.2014 05:12, Donald Stufft wrote:

ensurepip uses —no-index so it shouldn’t be hitting the network at all.


Do you have a test to ensure that ensurepip doesn't try to use network 
connections? You could e.g. mock socket.create_connection() and 
socket.socket() in a custom socket module. The subprocess makes it a 
little bit more complicated to test its behavior.


Christian
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Bob Hanson
On Sun, 5 Jan 2014 23:54:41 -0600, Tim Peters wrote:

[Bob Hanson]
  ... magnifying glass, I see it is two very long URLs ending with
  something like after the blah-blah:  ... akametechnology.com
 
[Stephen J. Turnbull]
  I suppose you tried cutting and pasting? [...]

Tried, but was unsuccessful.

[Tim Peters] 
 I don't think this was cut 'n paste.  Looking up the IP addresses
 returns legit Akamai URLs:

 [...] 
 
 C:\Codeping -a 23.59.190.113
 
 Pinging a23-59-190-113.deploy.static.akamaitechnologies.com
 [23.59.190.113] with 32 bytes of data:
 ...
 C:\Codeping -a 23.59.190.106
 
 Pinging a23-59-190-106.deploy.static.akamaitechnologies.com
 [23.59.190.106] with 32 bytes of data:
 ...
 
 Bob's  ... akametechnology.com just looks like compounded typos.

Typos or blindos. ;-)

Took a screenshot just now and zoomed in -- I can now verify that
the URLs are as Tim has 'em above.

[Stephen J. Turnbull]
  So your alarm seems to be verified, but why this happened to a Python
  download I don't know.  It could be DNS hacking between you and
  python.org, as well as something in the Python MSI.

[Tim Peters]
 Honestly, for all we _know_, this firewall alert may have been
 triggered by some other program that just happened to wake up while
 Bob was installing Python.  Sure, that's unlikely.  But so is
 everything else about this ;-)

Unlikely as the firewall alert has the full correct path for
*msiexec.exe*. I also keep tabs on all processes running, watch
my firewall routinely, etc. And -- I'm almost paranoid enough to
be a computer security guy. ;-) 

Wanted to add this tiny bit of info, but now I need to retire for
the night. I'll check further on things in the morning.

Bob Hanson

___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] [RELEASED] Python 3.4.0b2

2014-01-05 Thread Stephen J. Turnbull
Stephen J. Turnbull writes:

  IP address.deploy.static.akamitechnologies.com (according to host IP),

Ignore this; *my* aging eyes dropped the A in akamAitechnologies.com.

Sorry for the noise.


___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com