6.1-stable amd64 lang/python/2.7

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

6.1-stable amd64 lang/python/2.7

Adam Jensen
$ sysctl -n kern.version
OpenBSD 6.1-stable (GENERIC.MP) #0: Sat Aug  5 19:07:13 EDT 2017
    [hidden email]:/sys/arch/amd64/compile/GENERIC.MP

$ cat /usr/ports/CVS/Tag
TOPENBSD_6_1

$ cd /usr/ports/lang/python/2.7
$ make 2>&1 | /usr/ports/infrastructure/bin/portslogger ~/portlog

$ cat /home/hanzer/portlog/default.log
+++ Sat Aug  5 21:45:44 EDT 2017
===> python-2.7.13p0 depends on: metaauto-* -> metaauto-1.0p1
===> python-2.7.13p0 depends on: autoconf-2.69 -> autoconf-2.69p2
===> python-2.7.13p0 depends on: bzip2-* -> bzip2-1.0.6p8
===> python-2.7.13p0 depends on: sqlite3-* -> sqlite3-3.17.0
===> python-2.7.13p0 depends on: libffi-* -> libffi-3.2.1p2
===> python-2.7.13p0 depends on: gettext-* -> gettext-0.19.8.1
===> python-2.7.13p0 depends on: gdbm-* -> gdbm-1.12
===> python-2.7.13p0 depends on: tk->=8.5,<8.6 -> tk-8.5.18p1
===> python-2.7.13p0 depends on: tcl->=8.5,<8.6 -> tcl-8.5.18p1
===> python-2.7.13p0 depends on: db->=4,<5|db->=4v0,<5v0 -> db-4.6.21p3v0
===>  Verifying specs:  bz2 c crypto expat ffi intl m ncursesw panelw pthread  readline sqlite3 ssl stdc++ util z   gdbm>=3 pthread python2.7  X11 pthread python2.7  tk85 tcl85 lib/db4/db>=4 pthread python2.7
===>  found bz2.10.4 c.89.3 crypto.41.1 expat.11.0 ffi.1.2 intl.6.0 m.10.0 ncursesw.14.0 panelw.6.0 pthread.23.0 readline.4.0 sqlite3.35.0 ssl.43.2 stdc++.57.0 util.12.1 z.5.0 gdbm.4.1 python2.7.0.0 X11.16.1 tk85.0.14 tcl85.1.7 lib/db4/db.5.0
--- Sat Aug  5 21:45:45 EDT 2017

$ cat /home/hanzer/portlog/Python-2.7.13.log
+++ Sat Aug  5 21:45:45 EDT 2017
===>  Checking files for Python-2.7.13
`/usr/ports/distfiles/Python-2.7.13.tgz' is up to date.
>> (SHA256) Python-2.7.13.tgz: OK
===>  Extracting for Python-2.7.13
rm -r /usr/ports/pobj/Python-2.7.13/Python-2.7.13/Modules/expat
===>  Patching for Python-2.7.13
===>   Applying OpenBSD patch patch-Lib_test_regrtest_py
Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|$OpenBSD: patch-Lib_test_regrtest_py,v 1.10 2016/12/19 20:47:26 rpointel Exp $
|
|- test_bsddb and test_bsddb3 are in the list below as they are only available
|  on i386 and amd64.
|
|--- Lib/test/regrtest.py.orig Sat Dec 17 21:05:06 2016
|+++ Lib/test/regrtest.py Sun Dec 18 08:21:50 2016
--------------------------
Patching file Lib/test/regrtest.py using Plan A...
Hunk #1 succeeded at 1533.
Hunk #2 succeeded at 1569.
done
===>   Applying OpenBSD patch patch-Lib_test_test_tempfile_py
Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|$OpenBSD: patch-Lib_test_test_tempfile_py,v 1.5 2016/05/24 09:57:50 sthen Exp $
|--- Lib/test/test_tempfile.py.orig Sun Nov 10 08:36:40 2013
|+++ Lib/test/test_tempfile.py Mon Nov 11 15:27:59 2013
--------------------------
Patching file Lib/test/test_tempfile.py using Plan A...
Hunk #1 succeeded at 28.
done
===>   Applying OpenBSD patch patch-Makefile_pre_in
Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|$OpenBSD: patch-Makefile_pre_in,v 1.10 2016/12/19 20:47:26 rpointel Exp $
|--- Makefile.pre.in.orig Sat Dec 17 21:05:06 2016
|+++ Makefile.pre.in Sun Dec 18 08:21:50 2016
--------------------------
Patching file Makefile.pre.in using Plan A...
Hunk #1 succeeded at 583.
done
===>   Applying OpenBSD patch patch-Modules_pyexpat_c
Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|$OpenBSD: patch-Modules_pyexpat_c,v 1.2 2011/07/06 12:05:56 jasper Exp $
|--- Modules/pyexpat.c.orig Sun Oct 24 16:16:05 2010
|+++ Modules/pyexpat.c Wed Jul  6 14:05:41 2011
--------------------------
Patching file Modules/pyexpat.c using Plan A...
Hunk #1 succeeded at 435.
done
===>   Applying OpenBSD patch patch-Python_random_c
Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|$OpenBSD: patch-Python_random_c,v 1.1 2016/12/19 20:47:26 rpointel Exp $
|
|Issue #28932: http://bugs.python.org/issue28932
|
|--- Python/random.c.orig Sat Dec 17 21:05:07 2016
|+++ Python/random.c Sun Dec 18 09:43:27 2016
--------------------------
Patching file Python/random.c using Plan A...
Hunk #1 succeeded at 3.
done
===>   Applying OpenBSD patch patch-configure_ac
Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|$OpenBSD: patch-configure_ac,v 1.7 2016/12/19 20:47:26 rpointel Exp $
|
|SOVERSION defaults to 1.0. SHARED_LIBS, however, could be changed
|at any point.  Ensure they are on sync in case INSTSONAME is used
|by a third party.
|
|--- configure.ac.orig Sat Dec 17 21:05:07 2016
|+++ configure.ac Sun Dec 18 08:21:52 2016
--------------------------
Patching file configure.ac using Plan A...
Hunk #1 succeeded at 917.
done
===>   Applying OpenBSD patch patch-setup_py
Hmm...  Looks like a unified diff to me...
The text leading up to this was:
--------------------------
|$OpenBSD: patch-setup_py,v 1.12 2016/07/03 20:10:33 rpointel Exp $
|--- setup.py.orig Sat Jun 25 23:49:32 2016
|+++ setup.py Fri Jul  1 10:35:01 2016
--------------------------
Patching file setup.py using Plan A...
Hunk #1 succeeded at 35.
Hunk #2 succeeded at 461.
Hunk #3 succeeded at 791.
Hunk #4 succeeded at 929.
Hunk #5 succeeded at 999.
Hunk #6 succeeded at 1063.
Hunk #7 succeeded at 1086.
Hunk #8 succeeded at 1244.
Hunk #9 succeeded at 1850.
Hunk #10 succeeded at 1905.
Hunk #11 succeeded at 1948.
done
/usr/bin/perl /usr/ports/infrastructure/bin/pkg_subst -DMODTK_VERSION=8.5 -DMODTK_BIN=/usr/local/bin/wish8.5 -DMODTCL_VERSION=8.5 -DMODTCL_BIN=/usr/local/bin/tclsh8.5 -DLIBpython2.7_VERSION=0.0 -DMACHINE_ARCH=amd64 -DARCH=amd64 -DHOMEPAGE=http://www.python.org/ -D^PREFIX=/usr/local -D^SYSCONFDIR=/etc -DFLAVOR_EXT= -DFULLPKGNAME=Python-2.7.13 -DMAINTAINER=Remi\ Pointel\ \<[hidden email]\> -D^BASE_PKGPATH=lang/python/2.7 -D^LOCALBASE=/usr/local -D^X11BASE=/usr/X11R6 -D^TRUEPREFIX=/usr/local -D^RCDIR=/etc/rc.d -D^LOCALSTATEDIR=/var -i -B /usr/ports/pobj/Python-2.7.13 /usr/ports/pobj/Python-2.7.13/Python-2.7.13/configure.ac
Running autoconf-2.69 in /usr/ports/pobj/Python-2.7.13/Python-2.7.13
Running autoheader-2.69 in /usr/ports/pobj/Python-2.7.13/Python-2.7.13
===>  Configuring for Python-2.7.13
Using /usr/ports/pobj/Python-2.7.13/config.site (generated)
configure: WARNING: unrecognized options: --disable-silent-rules, --disable-gtk-doc
configure: loading site script /usr/ports/pobj/Python-2.7.13/config.site
checking build system type... x86_64-unknown-openbsd6.1
checking host system type... x86_64-unknown-openbsd6.1
checking for --enable-universalsdk... no
checking for --with-universal-archs... 32-bit
checking MACHDEP... openbsd6
checking EXTRAPLATDIR...
checking for --without-gcc... no
checking for --with-icc... no
checking for gcc... cc
checking whether the C compiler works... yes
checking for C compiler default output file name... a.out
checking for suffix of executables...
checking whether we are cross compiling... configure: error: in `/usr/ports/pobj/Python-2.7.13/Python-2.7.13':
configure: error: cannot run C compiled programs.
If you meant to cross compile, use `--host'.
See `config.log' for more details
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2620 '/usr/ports/pobj/Python-2.7.13/.configure_done': @for d in /usr/ports/pobj/P...)
*** Error 1 in /usr/ports/lang/python/2.7 (/usr/ports/infrastructure/mk/bsd.port.mk:2344 'all')
--- Sat Aug  5 21:45:54 EDT 2017

$ cat /etc/mk.conf
SUDO=doas
CLEANDEPENDS=Yes


Reply | Threaded
Open this post in threaded view
|

Re: 6.1-stable amd64 lang/python/2.7

Daniel Jakots-3
Check your dmesg. I guess you lack the wxallowed flag on your pobj
mount point.

On Sat, 5 Aug 2017 21:56:22 -0400, Adam Jensen <[hidden email]>
wrote:

> $ sysctl -n kern.version
> OpenBSD 6.1-stable (GENERIC.MP) #0: Sat Aug  5 19:07:13 EDT 2017
>     [hidden email]:/sys/arch/amd64/compile/GENERIC.MP
>
> $ cat /usr/ports/CVS/Tag
> TOPENBSD_6_1
>
> $ cd /usr/ports/lang/python/2.7
> $ make 2>&1 | /usr/ports/infrastructure/bin/portslogger ~/portlog
>
> $ cat /home/hanzer/portlog/default.log
> +++ Sat Aug  5 21:45:44 EDT 2017
> ===> python-2.7.13p0 depends on: metaauto-* -> metaauto-1.0p1
> ===> python-2.7.13p0 depends on: autoconf-2.69 -> autoconf-2.69p2
> ===> python-2.7.13p0 depends on: bzip2-* -> bzip2-1.0.6p8
> ===> python-2.7.13p0 depends on: sqlite3-* -> sqlite3-3.17.0
> ===> python-2.7.13p0 depends on: libffi-* -> libffi-3.2.1p2
> ===> python-2.7.13p0 depends on: gettext-* -> gettext-0.19.8.1
> ===> python-2.7.13p0 depends on: gdbm-* -> gdbm-1.12
> ===> python-2.7.13p0 depends on: tk->=8.5,<8.6 -> tk-8.5.18p1
> ===> python-2.7.13p0 depends on: tcl->=8.5,<8.6 -> tcl-8.5.18p1
> ===> python-2.7.13p0 depends on: db->=4,<5|db->=4v0,<5v0 ->
> db-4.6.21p3v0 ===>  Verifying specs:  bz2 c crypto expat ffi intl m
> ncursesw panelw pthread  readline sqlite3 ssl stdc++ util z   gdbm>=3
> pthread python2.7  X11 pthread python2.7  tk85 tcl85 lib/db4/db>=4
> pthread python2.7 ===>  found bz2.10.4 c.89.3 crypto.41.1 expat.11.0
> ffi.1.2 intl.6.0 m.10.0 ncursesw.14.0 panelw.6.0 pthread.23.0
> readline.4.0 sqlite3.35.0 ssl.43.2 stdc++.57.0 util.12.1 z.5.0
> gdbm.4.1 python2.7.0.0 X11.16.1 tk85.0.14 tcl85.1.7 lib/db4/db.5.0
> --- Sat Aug  5 21:45:45 EDT 2017
>
> $ cat /home/hanzer/portlog/Python-2.7.13.log
> +++ Sat Aug  5 21:45:45 EDT 2017
> ===>  Checking files for Python-2.7.13  
> `/usr/ports/distfiles/Python-2.7.13.tgz' is up to date.
> >> (SHA256) Python-2.7.13.tgz: OK  
> ===>  Extracting for Python-2.7.13  
> rm -r /usr/ports/pobj/Python-2.7.13/Python-2.7.13/Modules/expat
> ===>  Patching for Python-2.7.13
> ===>   Applying OpenBSD patch patch-Lib_test_regrtest_py  
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --------------------------
> |$OpenBSD: patch-Lib_test_regrtest_py,v 1.10 2016/12/19 20:47:26
> rpointel Exp $ |
> |- test_bsddb and test_bsddb3 are in the list below as they are only
> available |  on i386 and amd64.
> |
> |--- Lib/test/regrtest.py.orig Sat Dec 17 21:05:06 2016
> |+++ Lib/test/regrtest.py Sun Dec 18 08:21:50 2016
> --------------------------
> Patching file Lib/test/regrtest.py using Plan A...
> Hunk #1 succeeded at 1533.
> Hunk #2 succeeded at 1569.
> done
> ===>   Applying OpenBSD patch patch-Lib_test_test_tempfile_py  
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --------------------------
> |$OpenBSD: patch-Lib_test_test_tempfile_py,v 1.5 2016/05/24 09:57:50
> sthen Exp $ |--- Lib/test/test_tempfile.py.orig Sun Nov 10
> 08:36:40 2013 |+++ Lib/test/test_tempfile.py Mon Nov 11
> 15:27:59 2013 --------------------------
> Patching file Lib/test/test_tempfile.py using Plan A...
> Hunk #1 succeeded at 28.
> done
> ===>   Applying OpenBSD patch patch-Makefile_pre_in  
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --------------------------
> |$OpenBSD: patch-Makefile_pre_in,v 1.10 2016/12/19 20:47:26 rpointel
> Exp $ |--- Makefile.pre.in.orig Sat Dec 17 21:05:06 2016
> |+++ Makefile.pre.in Sun Dec 18 08:21:50 2016
> --------------------------
> Patching file Makefile.pre.in using Plan A...
> Hunk #1 succeeded at 583.
> done
> ===>   Applying OpenBSD patch patch-Modules_pyexpat_c  
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --------------------------
> |$OpenBSD: patch-Modules_pyexpat_c,v 1.2 2011/07/06 12:05:56 jasper
> Exp $ |--- Modules/pyexpat.c.orig Sun Oct 24 16:16:05 2010
> |+++ Modules/pyexpat.c Wed Jul  6 14:05:41 2011
> --------------------------
> Patching file Modules/pyexpat.c using Plan A...
> Hunk #1 succeeded at 435.
> done
> ===>   Applying OpenBSD patch patch-Python_random_c  
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --------------------------
> |$OpenBSD: patch-Python_random_c,v 1.1 2016/12/19 20:47:26 rpointel
> Exp $ |
> |Issue #28932: http://bugs.python.org/issue28932
> |
> |--- Python/random.c.orig Sat Dec 17 21:05:07 2016
> |+++ Python/random.c Sun Dec 18 09:43:27 2016
> --------------------------
> Patching file Python/random.c using Plan A...
> Hunk #1 succeeded at 3.
> done
> ===>   Applying OpenBSD patch patch-configure_ac  
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --------------------------
> |$OpenBSD: patch-configure_ac,v 1.7 2016/12/19 20:47:26 rpointel Exp $
> |
> |SOVERSION defaults to 1.0. SHARED_LIBS, however, could be changed
> |at any point.  Ensure they are on sync in case INSTSONAME is used
> |by a third party.
> |
> |--- configure.ac.orig Sat Dec 17 21:05:07 2016
> |+++ configure.ac Sun Dec 18 08:21:52 2016
> --------------------------
> Patching file configure.ac using Plan A...
> Hunk #1 succeeded at 917.
> done
> ===>   Applying OpenBSD patch patch-setup_py  
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --------------------------
> |$OpenBSD: patch-setup_py,v 1.12 2016/07/03 20:10:33 rpointel Exp $
> |--- setup.py.orig Sat Jun 25 23:49:32 2016
> |+++ setup.py Fri Jul  1 10:35:01 2016
> --------------------------
> Patching file setup.py using Plan A...
> Hunk #1 succeeded at 35.
> Hunk #2 succeeded at 461.
> Hunk #3 succeeded at 791.
> Hunk #4 succeeded at 929.
> Hunk #5 succeeded at 999.
> Hunk #6 succeeded at 1063.
> Hunk #7 succeeded at 1086.
> Hunk #8 succeeded at 1244.
> Hunk #9 succeeded at 1850.
> Hunk #10 succeeded at 1905.
> Hunk #11 succeeded at 1948.
> done
> /usr/bin/perl /usr/ports/infrastructure/bin/pkg_subst
> -DMODTK_VERSION=8.5 -DMODTK_BIN=/usr/local/bin/wish8.5
> -DMODTCL_VERSION=8.5 -DMODTCL_BIN=/usr/local/bin/tclsh8.5
> -DLIBpython2.7_VERSION=0.0 -DMACHINE_ARCH=amd64 -DARCH=amd64
> -DHOMEPAGE=http://www.python.org/ -D^PREFIX=/usr/local
> -D^SYSCONFDIR=/etc -DFLAVOR_EXT= -DFULLPKGNAME=Python-2.7.13
> -DMAINTAINER=Remi\ Pointel\ \<[hidden email]\>
> -D^BASE_PKGPATH=lang/python/2.7 -D^LOCALBASE=/usr/local
> -D^X11BASE=/usr/X11R6 -D^TRUEPREFIX=/usr/local -D^RCDIR=/etc/rc.d
> -D^LOCALSTATEDIR=/var -i
> -B /usr/ports/pobj/Python-2.7.13 /usr/ports/pobj/Python-2.7.13/Python-2.7.13/configure.ac
> Running autoconf-2.69 in /usr/ports/pobj/Python-2.7.13/Python-2.7.13
> Running autoheader-2.69
> in /usr/ports/pobj/Python-2.7.13/Python-2.7.13 ===>  Configuring for
> Python-2.7.13 Using /usr/ports/pobj/Python-2.7.13/config.site
> (generated) configure: WARNING: unrecognized options:
> --disable-silent-rules, --disable-gtk-doc configure: loading site
> script /usr/ports/pobj/Python-2.7.13/config.site checking build
> system type... x86_64-unknown-openbsd6.1 checking host system type...
> x86_64-unknown-openbsd6.1 checking for --enable-universalsdk... no
> checking for --with-universal-archs... 32-bit checking MACHDEP...
> openbsd6 checking EXTRAPLATDIR... checking for --without-gcc... no
> checking for --with-icc... no checking for gcc... cc checking whether
> the C compiler works... yes checking for C compiler default output
> file name... a.out checking for suffix of executables... checking
> whether we are cross compiling... configure: error: in
> `/usr/ports/pobj/Python-2.7.13/Python-2.7.13': configure: error:
> cannot run C compiled programs. If you meant to cross compile, use
> `--host'. See `config.log' for more details *** Error 1 in .
> (/usr/ports/infrastructure/mk/bsd.port.mk:2620
> '/usr/ports/pobj/Python-2.7.13/.configure_done': @for d
> in /usr/ports/pobj/P...) *** Error 1 in /usr/ports/lang/python/2.7
> (/usr/ports/infrastructure/mk/bsd.port.mk:2344 'all') --- Sat Aug  5
> 21:45:54 EDT 2017
>
> $ cat /etc/mk.conf
> SUDO=doas
> CLEANDEPENDS=Yes
>
>

Reply | Threaded
Open this post in threaded view
|

Re: 6.1-stable amd64 lang/python/2.7

Adam Jensen
On Sat, 5 Aug 2017 22:08:13 -0400
Daniel Jakots <[hidden email]> wrote:

> Check your dmesg. I guess you lack the wxallowed flag on your pobj
> mount point.
>

Thanks, that seems to have fixed the build problem.

There might be an opportunity for someone in the dev/doc community to insert a blurb about this requirement into faq15 and/or faq14.

Reply | Threaded
Open this post in threaded view
|

Re: 6.1-stable amd64 lang/python/2.7

Diana Eichert
In reply to this post by Adam Jensen
I saw the same problem this morning, determined it was wxallowed
required on /usr.  After updating /usr mount with wxallowed
Python build continued.

I decided to search ports mailing list to see if anyone else had the
same issue, stumbled across this thread.

I installed 6.1 from ISO I downloaded right after 6.1 release
announcement.  I used installer default partition settings.

I did not create a new partition for ports, so it is a directory
on /usr.

As Adam stated perhaps an update to faq15?

diana

Reply | Threaded
Open this post in threaded view
|

Re: 6.1-stable amd64 lang/python/2.7

Ingo Schwarze
Hi Diana,

Diana Eichert wrote on Mon, Aug 07, 2017 at 02:41:36PM -0600:

> I saw the same problem this morning, determined it was wxallowed
> required on /usr.  After updating /usr mount with wxallowed
> Python build continued.

What a terrible idea.  You certainly don't want wxallowed on /usr.

> I did not create a new partition for ports, so it is a directory
> on /usr.

To use messy ports, a separate /usr/local is required, and to build
them, a separate /usr/ports/pobj.

> As Adam stated perhaps an update to faq15?

I think this ought to be documented in the manual pages.
It is a major trap for the unwary.

OK?
  Ingo


Index: packages.7
===================================================================
RCS file: /cvs/src/share/man/man7/packages.7,v
retrieving revision 1.41
diff -u -p -r1.41 packages.7
--- packages.7 27 Mar 2017 20:05:51 -0000 1.41
+++ packages.7 8 Aug 2017 16:25:35 -0000
@@ -85,6 +85,16 @@ security issues and
 will warn if those are installed and cannot be updated.
 This prevents a scenario where a bad guy would maintain a partial mirror
 with outdated packages.
+.Pp
+A small number of packages contain insecure code requiring
+.Xr mmap 2
+memory both writeable and executable.
+To use such insecurely written software, a separate
+.Pa /usr/local
+file system with the
+.Cm wxallowed
+.Xr mount 8
+option is needed.
 .Sh MANAGING FILES
 The package system offers some strong warranties.
 .Ss "Installing a package won't erase existing files"
Index: ports.7
===================================================================
RCS file: /cvs/src/share/man/man7/ports.7,v
retrieving revision 1.113
diff -u -p -r1.113 ports.7
--- ports.7 10 Jul 2017 22:48:00 -0000 1.113
+++ ports.7 8 Aug 2017 16:25:35 -0000
@@ -700,6 +700,12 @@ The default ports directory.
 Ports master Makefile.
 .It Pa /usr/ports/INDEX
 Ports index.
+.It Pa /usr/ports/pobj
+Build directories.
+A number of insecurely coded ports require a dedicated file system with the
+.Cm wxallowed
+.Xr mount 8
+option.
 .It Pa /usr/ports/infrastructure/mk/bsd.port.mk
 The ports main engine.
 .It Pa /usr/ports/infrastructure/templates/network.conf.template

Reply | Threaded
Open this post in threaded view
|

Re: 6.1-stable amd64 lang/python/2.7

Diana Eichert
Hi Ingo

Chuckle, I know mounting /usr with wxallowed is terrible and ugly,
but I figured if I stated such a stupid thing someone with commit
capability would come up with an idea to mitigate an idiot like
myself from shooting myself in the foot.  ;-)

Also I installed /usr/ports using the directions I've been using
for years, http://www.openbsd.org/anoncvs.html .  Perhaps a change
should be made to the "Getting the ports and xenocara trees" area
recommending the use of a separate partition for ports and ports
build space.

g.day

diana




On Tue, 8 Aug 2017, Ingo Schwarze wrote:

> Hi Diana,
>
> Diana Eichert wrote on Mon, Aug 07, 2017 at 02:41:36PM -0600:
>
>> I saw the same problem this morning, determined it was wxallowed
>> required on /usr.  After updating /usr mount with wxallowed
>> Python build continued.
>
> What a terrible idea.  You certainly don't want wxallowed on /usr.
>
>> I did not create a new partition for ports, so it is a directory
>> on /usr.
>
> To use messy ports, a separate /usr/local is required, and to build
> them, a separate /usr/ports/pobj.
>
>> As Adam stated perhaps an update to faq15?
>
> I think this ought to be documented in the manual pages.
> It is a major trap for the unwary.
>
> OK?
>  Ingo
>
>
> Index: packages.7
> ===================================================================
> RCS file: /cvs/src/share/man/man7/packages.7,v
> retrieving revision 1.41
> diff -u -p -r1.41 packages.7
> --- packages.7 27 Mar 2017 20:05:51 -0000 1.41
> +++ packages.7 8 Aug 2017 16:25:35 -0000
> @@ -85,6 +85,16 @@ security issues and
> will warn if those are installed and cannot be updated.
> This prevents a scenario where a bad guy would maintain a partial mirror
> with outdated packages.
> +.Pp
> +A small number of packages contain insecure code requiring
> +.Xr mmap 2
> +memory both writeable and executable.
> +To use such insecurely written software, a separate
> +.Pa /usr/local
> +file system with the
> +.Cm wxallowed
> +.Xr mount 8
> +option is needed.
> .Sh MANAGING FILES
> The package system offers some strong warranties.
> .Ss "Installing a package won't erase existing files"
> Index: ports.7
> ===================================================================
> RCS file: /cvs/src/share/man/man7/ports.7,v
> retrieving revision 1.113
> diff -u -p -r1.113 ports.7
> --- ports.7 10 Jul 2017 22:48:00 -0000 1.113
> +++ ports.7 8 Aug 2017 16:25:35 -0000
> @@ -700,6 +700,12 @@ The default ports directory.
> Ports master Makefile.
> .It Pa /usr/ports/INDEX
> Ports index.
> +.It Pa /usr/ports/pobj
> +Build directories.
> +A number of insecurely coded ports require a dedicated file system with the
> +.Cm wxallowed
> +.Xr mount 8
> +option.
> .It Pa /usr/ports/infrastructure/mk/bsd.port.mk
> The ports main engine.
> .It Pa /usr/ports/infrastructure/templates/network.conf.template
>
>

Reply | Threaded
Open this post in threaded view
|

Re: 6.1-stable amd64 lang/python/2.7

Ingo Schwarze
Hi Diana,

Diana Eichert wrote on Wed, Aug 09, 2017 at 11:33:45AM -0600:

> Also I installed /usr/ports using the directions I've been using
> for years, http://www.openbsd.org/anoncvs.html .  Perhaps a change
> should be made to the "Getting the ports and xenocara trees" area
> recommending the use of a separate partition for ports and ports
> build space.

I don't think that would be the right place to mention it.  Having
/usr/ports and /usr/xenocara as part of the /usr filesystem is
probably unusual, but i can't see anything seriously wrong with it.
Only /usr/ports/pobj needs a dedicated file system, and you can
easily do that after checking out the ports tree, even much later
once you run into the problem.

Yours,
  Ingo

Reply | Threaded
Open this post in threaded view
|

Re: 6.1-stable amd64 lang/python/2.7

Diana Eichert
Hello Ingo

Now I'm going more into generic bikeshed discussions.

I wasn't recommending change to /usr/xenocara just using the paragraph
title for reference.  Prior to the automatic partition setup in the
installer I did used to create a /usr/ports partition, but stopped
doing it when the installer asked if I wanted to use the partition
table with automatic defaults.  I assume the installer is trying to
create "sane" defaults, if so should not a /usr/ports partition also
be created.

I believe it is a false assumption to believe /usr/ports and
/usr/xenocara as part of /usr filesystem is unusual.

Anyway, this discussion is way beyond Python 2.7 build issue, I'll
leave it alone.

g.day

diana

On Fri, 11 Aug 2017, Ingo Schwarze wrote:

> Hi Diana,
>
> Diana Eichert wrote on Wed, Aug 09, 2017 at 11:33:45AM -0600:
>
>> Also I installed /usr/ports using the directions I've been using
>> for years, http://www.openbsd.org/anoncvs.html .  Perhaps a change
>> should be made to the "Getting the ports and xenocara trees" area
>> recommending the use of a separate partition for ports and ports
>> build space.
>
> I don't think that would be the right place to mention it.  Having
> /usr/ports and /usr/xenocara as part of the /usr filesystem is
> probably unusual, but i can't see anything seriously wrong with it.
> Only /usr/ports/pobj needs a dedicated file system, and you can
> easily do that after checking out the ports tree, even much later
> once you run into the problem.
>
> Yours,
>  Ingo
>
>