FTBFS on artful (can't cd to python in override_dh_install)

[Christian Ehrhardt ]

Hi,
while working on a new DPDK I was wondering how I'd broken the non DPDK arches as well.
TL;DR it is currently FTBFS as-is in artful.

Full build log:
https://launchpadlibrarian.net/320055040/buildlog_ubuntu-artful-amd64.openvswitch_2.7.0-0ubuntu2_BUILDING.txt.gz

Error:
Copying ovs.egg-info to /<<PKGBUILDDIR>>/debian/python3-openvswitch/usr/lib/python3/dist-packages/ovs-2.7.0.egg-info
Skipping SOURCES.txt
running install_scripts
/bin/sh: 2: cd: can't cd to python
debian/rules:84: recipe for target 'override_dh_install' failed
make[1]: *** [override_dh_install] Error 2
make[1]: Leaving directory '/<<PKGBUILDDIR>>'
debian/rules:6: recipe for target 'binary' failed

[Christian Ehrhardt  (paelzer)]

Also:

x86_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -g -O2 -fdebug-prefix-map=/<<PKGBUILDDIR>>=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -fPIC -I/usr/include/python2.7 -c ovs/_json.c -o build/temp.linux-x86_64-2.7/ovs/_json.o
ovs/_json.c:1:20: fatal error: Python.h: No such file or directory
 #include "Python.h"

[Christian Ehrhardt  (paelzer)]

Ok, the gcc error seems to be "normal"?
Maybe it just went undetected so far.

The good build has this as well:
x86_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -g -O2 -fdebug-prefix-map=/<<PKGBUILDDIR>>=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -fPIC -I/usr/include/python2.7 -c ovs/_json.c -o build/temp.linux-x86_64-2.7/ovs/_json.o
ovs/_json.c:1:20: fatal error: Python.h: No such file or directory
 #include "Python.h"
x86_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -g -O2 -fdebug-prefix-map=/<<PKGBUILDDIR>>=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -fPIC -I/usr/include/python3.5m -c ovs/_json.c -o build/temp.linux-x86_64-3.5/ovs/_json.o
ovs/_json.c:1:20: fatal error: Python.h: No such file or directory
 #include "Python.h"

Once per python version, the openvswitch build automatically does:
WARNING: The C extension could not be compiled, speedups are not enabled.
Failure information, if any, is above.
Retrying the build without the C extension.

Need to discuss if that is ok or another issue in itself, but that means the FTBFS atm is only about:
GOOD
running install_scripts
install -D _dpdk/vswitchd/ovs-vswitchd \
        /<<PKGBUILDDIR>>/debian/openvswitch-switch-dpdk/usr/lib/openvswitch-switch-dpdk/ovs-vswitchd-dpdk

BAD
running install_scripts
/bin/sh: 2: cd: can't cd to python
debian/rules:84: recipe for target 'override_dh_install' failed

[Christian Ehrhardt  (paelzer)]

To ensure debugging isn't too easy a local SBUILD just worked as-is, while reliably failing on Launchpad.

[Christian Ehrhardt  (paelzer)]

Local repro worked, it seems it is due to two python's being in proposed at once.
The first cd works, but the second fails ... trying a fix now.

[Christian Ehrhardt  (paelzer)]

Confirmed, pyversions provides two values for
PYTHON3S:=$(shell py3versions -vr)

$ py3versions -vr
3.6 3.5

[Christian Ehrhardt  (paelzer)]

Fix is working, but I need to fix up more around DPDK and Arm until I push mine together with the others.

Please let me know if that is any more urgent for you.

[Christian Ehrhardt  (paelzer)]

FYI - still waiting on upstream fix in regard to arm support to be accepted to be "one" upload for the changes.

[Launchpad Janitor (janitor)]

This bug was fixed in the package openvswitch - 2.7.0-0ubuntu2

---------------
openvswitch (2.7.0-0ubuntu2) artful; urgency=medium

  * d/control, d/rules: Enable DPDK support on Arm (LP: #1691659)
  * d/rules: Fix FTBFS if multiple python interpreters are
    installed (LP: #1691658)
  * d/p/ovs-dev-v3-netdev-dpdk-round-up-mbuf_size-to-cache_line_size.patch:
    round up mbuf_size to cache_line_size to fix some DPDK PMDs e.g. vNIC
    thunderx PMD (LP: #1694221)

 -- Christian Ehrhardt <email address hidden> Wed, 17 May 2017 15:45:13 +0200