Systemd service in package is restarted on upgrade - setuptools

I use debhelper with python setuptools in order to build my packages.
I recently updated the compatibility level 9 to 11 in order to use the systemd timers.
From that moment every time I upgrade the package the service contained is restarted.
I tried to build with the following rules:
#! /usr/bin/make -f
#export DH_VERBOSE = 1
export PYBUILD_NAME=my_pkg
export DH_ALWAYS_EXCLUDE=CVS:.svn:.git:.vscode*
export PYBUILD_INTERPRETERS=python3
%:
dh $# --with python3 --buildsystem=pybuild
override_dh_installinit:
dh_installinit --no-stop-on-upgrade --no-restart-on-upgrade --no-restart-after-upgrade --no-start
override_dh_systemd_enable:
dh_systemd_enable --name=my_pkg
override_dh_systemd_start:
dh_systemd_start --no-stop-on-upgrade --no-restart-on-upgrade --no-restart-after-upgrade --no-start
python3 setup.py clean --all
According to the documentation those tags should do what I look for but probably there is something I'm missing:
dh_systemd_start
dh_installinit
Every time I update it the service contained is restarted.
The service is running the update itself so when is restarted the update is left uncompleted.

As expected I was looking in the wrong direction.
The correct option to use is:
dh_installsystemd
change the rule file into:
#! /usr/bin/make -f
#export DH_VERBOSE = 1
export PYBUILD_NAME=my_pkg
export DH_ALWAYS_EXCLUDE=CVS:.svn:.git:.vscode*
export PYBUILD_INTERPRETERS=python3
%:
dh $# --with python3 --buildsystem=pybuild
override_dh_systemd_enable:
dh_systemd_enable --name=my_pkg
override_dh_installsystemd:
dh_installsystemd --no-restart-after-upgrade
override_dh_systemd_start:
python3 setup.py clean --all
I hope this will help someone else in the same situation.

Related

Viewing the cpanm error log in Bitbucket pipelines when installing a Perl module fails

I have a BitBucket Pipeline that installs a bunch of Perl modules using cpanm. One of them fails and this is the snippet I can see in the log:
Fetching http://www.cpan.org/authors/id/P/PE/PETDANCE/ack-v3.0.2.tar.gz ... OK
Configuring ack-v3.0.2 ... OK
==> Found dependencies: File::Next
--> Working on File::Next
Fetching http://www.cpan.org/authors/id/P/PE/PETDANCE/File-Next-1.16.tar.gz ... OK
Configuring File-Next-1.16 ... OK
Building and testing File-Next-1.16 ... OK
Successfully installed File-Next-1.16
! Installing App::Ack failed. See /root/.cpanm/work/1562605191.55/build.log for details. Retry with --force to force install it.
! Installing the dependencies failed: Module 'App::Ack' is not installed
! Bailing out the installation for ..
Building and testing ack-v3.0.2 ... FAIL
How could I access the build.log that was created by the installation process?
Apparently recently Bitbucket added a feature called "after-script" so I could add the following and that would print the content of the log files.
after-script:
- ls -1 /root/.cpanm/work/*/build.log | xargs cat
or maybe even this:
after-script:
- cat /root/.cpanm/work/*/build.log
and the following, I think, will only show the content of the log files if the build failed:
after-script:
- $BITBUCKET_EXIT_CODE && cat /root/.cpanm/work/*/build.log
Read more here: https://bitbucket.org/blog/after-scripts-now-available-for-bitbucket-pipelines

"cpanm PHP" fails

What should I change in order to install the module? php is 5.3.3 withyum install php-devel in place.
PHP.c: In function ‘PHP_set_php_input’:
PHP.c:818: warning: passing argument 2 of ‘Perl_sv_2pv_flags’ from incompatible pointer type
/home/mpapec/.plenv/versions/5.20.0/lib/perl5/5.20.0/x86_64-linux/CORE/proto.h:3931: note: expected ‘STRLEN * const’ but argument is of type ‘int *’
cc -c -I/usr/include/php -I/usr/include/php/main -I/usr/include/php/TSRM -I/usr/include/php/Zend -I/usr/include/php/ext -I/usr/include/php/ext/date/lib -Wall -O2 -DVERSION=\"0.15\" -DXS_VERSION=\"0.15\" -fPIC "-I/home/mpapec/.plenv/versions/5.20.0/lib/perl5/5.20.0/x86_64-linux/CORE" array.c
In file included from /usr/include/php/main/php.h:33,
from /usr/include/php/sapi/embed/php_embed.h:23,
from PHP.h:14,
from array.c:9:
/usr/include/php/main/php_config.h:2417:1: warning: "_GNU_SOURCE" redefined
In file included from /home/mpapec/.plenv/versions/5.20.0/lib/perl5/5.20.0/x86_64-linux/CORE/perl.h:28,
from PHP.h:9,
from array.c:9:
/home/mpapec/.plenv/versions/5.20.0/lib/perl5/5.20.0/x86_64-linux/CORE/config.h:1825:1: warning: this is the location of the previous definition
rm -f blib/arch/auto/PHP/PHP.so
cc -shared -O2 -L/usr/local/lib -fstack-protector PHP.o array.o -o blib/arch/auto/PHP/PHP.so \
\
chmod 755 blib/arch/auto/PHP/PHP.so
"/home/mpapec/.plenv/versions/5.20.0/bin/perl5.20.0" -MExtUtils::Command::MM -e 'cp_nonempty' -- PHP.bs blib/arch/auto/PHP/PHP.bs 644
Manifying 1 pod document
Running Mkbootstrap for PHP ()
chmod 644 "PHP.bs"
PERL_DL_NONLAZY=1 "/home/mpapec/.plenv/versions/5.20.0/bin/perl5.20.0" "-Iblib/lib" "-Iblib/arch" test.pl
1..79
not ok 1 - use_ok PHP
# Failed test 'use_ok PHP'
# at test.pl line 11.
not ok 2 - require PHP;
# Failed test 'require PHP;'
# at test.pl line 18.
# Tried to require 'PHP'.
# Error: Attempt to reload PHP.pm aborted.
# Compilation failed in require at (eval 6) line 2.
not ok 3 - eval
# Failed test 'eval'
# at test.pl line 49.
Module PHP failed to load at blib/lib/PHP.pm line 80.
Module PHP failed to load at blib/lib/PHP.pm line 80.
END failed--call queue aborted at test.pl line 50.
# Looks like you planned 79 tests but ran 3.
# Looks like you failed 3 tests of 3 run.
# Looks like your test exited with 22 just after 3.
make: *** [test_dynamic] Error 22
-> FAIL Installing PHP failed. See /home/mpapec/.cpanm/work/1440522239.12833/build.log for details. Retry with --force to force install it.
(
PHP is kind of fragile. It probably won't work out of the box with your system php installation, and may have trouble with 64-bit or multi-threaded versions of perl.
I have only ever gotten it to work on Linux. The latest version I have tried to use is 5.3.8 (back in 2013), though I remember things going smoothly from 5.2.x to 5.3.8.
I always build php from source, with this configuration:
./configure --enable-embed --with-zlib --with-openssl --with-mysql \
--with-libdir=lib/i386-linux-gnu
--enable-embed is absolutely required, as the pod mentions, to build a PHP interpreter with the SAPI extension, and which then allows perl to manipulate the PHP interpreter through XS code. The other extensions were for other requirements of my project; they may be optional, but I haven't experimented with building the PHP interpreter or the PHP module with any other configuration. The pod also says to never use the --with-apxs argument, which I was never tempted to do anyway.
The build process of the PHP module will look for and require a program called php-config. You may need to hack your $PATH, if only during the build process, so that the PHP module runs the correct php-config. After that the module will know where to look for the rest of your php installation.
I had fun working with this module for a while (writing a Catalyst and then a Mojolicious wrapper around WordPress), but it has fallen into disrepair and disrepute. Share whatever you learn trying to build it and we'll put it in the docs, making this module that much easier to use.

Unable to install Image::ExifTool due to 'make test' hanging

I have a Netgear ReadyNAS NV+ on which I'm trying to do some things using perl.
Since the perl installation from the vendor is compiled without the uselargefiles-flag I've tried to compile/install my own version to run in parallell with the system one.
When I first tried to compile perl on the machine itself I ran into trouble as the system is lacking alot of regular system tools which prevents the Configure-script from running properly.
# ./Configure -des -Dprefix=$HOME/perl-5.16.2
First let's make sure your kit is complete. Checking...
./Configure: line 1986: split: command not found
cat: /root/dload/perl-5.16.2/UU/x??: No such file or directory
Locating common programs...
I don't know where 'comm' is, and my life depends on it.
Go find a public domain implementation or fix your PATH setting!
To get around that I installed and used this cross-compiler to build perl on another machine.
I could not figure out how to cross-compile using the standard configure-script from perl, so I used this.
After that I could just copy my new perl-folder onto the NAS and have thing like the following work:
# ~/sparc-perl-5.16.2/bin/perl -version
This is perl 5, version 16, subversion 2 (v5.16.2) built for sparc-linux
Copyright 1987-2012, Larry Wall
Perl may be copied only under the terms of either the Artistic License or the
GNU General Public License, which may be found in the Perl 5 source kit.
Complete documentation for Perl, including FAQ lists, should be found on
this system using "man perl" or "perldoc perl". If you have access to the
Internet, point your browser at http://www.perl.org/, the Perl Home Page.
# ~/sparc-perl-5.16.2/bin/perl -e 'print "A string!\n";'
A string!
So in some sense I feel I've been successful in my quest to get "my" perl working on the machine.
I get in trouble when I try to install extra modules though.
I've tried installing Image::ExifTool using cpanp (among others), but the installation never finishes.
Today I resorted to trying the installation manually.
wget http://search.cpan.org/CPAN/authors/id/E/EX/EXIFTOOL/Image-ExifTool-9.12.tar.gz
tar xzf Image-ExifTool-9.12.tar.gz
cd Image-ExifTool-9.12
~/sparc-perl-5.16.2/bin/perl Makefile.pl
make
All work as they're supposed to.
when I get to make test however it hangs indefinitely like so:
# make test
PERL_DL_NONLAZY=1 /root/sparc-perl-5.16.2/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/AFCP.t ...........
Looking at the running processes I think these are the ones representing the command:
# ps aux | grep per[l]
root 26187 0.5 0.7 12000 7808 pts/0 S+ 10:31 0:07 /root/sparc-perl-5.16.2/bin/perl -MExtUtils::Command::MM -e test_harness(0, 'blib/lib', 'blib/arch') t/AFCP.t t/AIFF.t t/APE.t t/ASF.t t/BigTIFF.t t/BMP.t t/CanonRaw.t t/Canon.t t/CanonVRD.t t/Casio.t t/DICOM.t t/DjVu.t t/DNG.t t/DV.t t/EXE.t t/ExifTool.t t/FLAC.t t/FlashPix.t t/Flash.t t/Font.t t/FotoStation.t t/FujiFilm.t t/Geotag.t t/GeoTiff.t t/GE.t t/GIF.t t/GIMP.t t/GPS.t t/HTML.t t/InDesign.t t/IPTC.t t/ITC.t t/Jpeg2000.t t/JVC.t t/Kodak.t t/KyoceraRaw.t t/Lang.t t/LNK.t t/M2TS.t t/Matroska.t t/MIE.t t/MIFF.t t/Minolta.t t/MP3.t t/MWG.t t/MXF.t t/Nikon.t t/Olympus.t t/OpenEXR.t t/Panasonic.t t/PDF.t t/Pentax.t t/PGF.t t/PhotoCD.t t/PhotoMechanic.t t/Photoshop.t t/PICT.t t/PNG.t t/PostScript.t t/PPM.t t/PSP.t t/QuickTime.t t/Radiance.t t/Real.t t/Ricoh.t t/RIFF.t t/RTF.t t/Sanyo.t t/Sigma.t t/Sony.t t/Unknown.t t/Vorbis.t t/Writer.t t/XMP.t t/ZIP.t
root 26189 1.7 0.0 0 0 pts/0 Z+ 10:31 0:24 [perl] <defunct>
This is completely consistent with what happens when I try the same installation using cpanp.
I guess I could just run make install and hope that it won't be to much of a problem that the test is unable to run/complete, but that makes me feel a bit uneasy. I'd much rather figure out what is wrong.
I also find it odd that the test hangs the way it does, I could accept that it fails, but that it just keeps running in a non-productive manner feels weird.
Since it is the first time I've built perl from source and the first time I've used a cross-compiler I would not be surprised to learn I've done something very wrong, any help identifying what is very welcome.
What am I doing wrong?
EDIT - compiling on the mahcine:
Figured out how to get the Configure script working natively on the machine.
Will try to build there to see if that will work better. It'll take some time though, the machine is slow.
To be able to run the configure script I needed:
GCC
apt-get install libc6-dev gcc gdb libtag1-dev uuid-dev
coreutils
apt-get install coreutils
-Dcc=gcc
./Configure -des -Dprefix=$HOME/perl-5.16.2 -Dcc=gcc
EDIT2 - compile failed:
Most shortlived success ever:
# make
`sh cflags "optimize='-O2'" perlmini.o` -DPERL_IS_MINIPERL -DPERL_EXTERNAL_GLOB perlmini.c
CCCMD = gcc -DPERL_CORE -c -fno-strict-aliasing -pipe -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -Wall
In file included from perl.c:33:
perl.h:699: error: conflicting types for `syscall'
/usr/include/unistd.h:939: error: previous declaration of `syscall'
ake: *** [perlmini.o] Error 1
EDIT3 - testdriving fork:
Tried the following to verify that fork works as it should. I believe that the test is successful, but I've never tried fork before, so please let me know if I read the output wrong.
# cat test_forkwait.px
#!/root/sparc-perl-5.16.2/bin/perl
$pid = fork();
die if $pid < 0;
$SIG{CHLD} = sub { warn "SIGCHLD\n"; };
if($pid) {
$status = waitpid($pid, 0);
warn "wait status: $status\n";
} else {
warn "Child starting\n";
sleep 1;
warn "Child terminating\n";
}
# ./test_forkwait.px
Child starting
Child terminating
SIGCHLD
wait status: 29913
I'm a bit late with this comment, and don't have any ideas about your "make test" hang, but I wanted to point out the Image::ExifTool is pure Perl, so it does not require compilation. The install step is as simple as copying the distribution files to the proper directories.
Phil
Gave up on figuring out why the test fails. Ended up using the module anyways. Seems to work OK with my files.

Trying to install meteor on centos 5

OS: Red Hat Enterprise Linux Server release 5.5 (Tikanga)
Trying to run this command to install meteor:
curl install.meteor.com | /bin/sh
============= Output Error ===============
Installing /tmp/meteor-install-Nv47529/meteor-0.3.6-1.x86_64.rpm
error: Failed dependencies:
libc.so.6(GLIBC_2.6)(64bit) is needed by meteor-0.3.6-1.x86_64
libc.so.6(GLIBC_2.7)(64bit) is needed by meteor-0.3.6-1.x86_64
libc.so.6(GLIBC_2.9)(64bit) is needed by meteor-0.3.6-1.x86_64
libstdc++.so.6(GLIBCXX_3.4.9)(64bit) is needed by meteor-0.3.6-1.x86_64
Installation failed.
============= Output Error ===============
glib is missing showing but check glib using
rpm -qa --qf="%{n}-%{v}-%{r}.%{arch}\n" | grep glibc | sort
showing me
compat-glibc-2.3.4-2.26.i386
compat-glibc-2.3.4-2.26.x86_64
compat-glibc-headers-2.3.4-2.26.x86_64
glibc-2.5-49.el5_5.7.i686
glibc-2.5-49.el5_5.7.x86_64
glibc-common-2.5-49.el5_5.7.x86_64
glibc-devel-2.5-49.el5_5.7.i386
glibc-devel-2.5-49.el5_5.7.x86_64
glibc-headers-2.5-49.el5_5.7.x86_64
glibc-utils-2.5-49.el5_5.7.x86_64
so any idea what is the issue?
GLIBC_2.6 is required by the RPM, you only have GLIBC_2.5 so you need to upgrade.
Alternatively, this dependency may be more strict than it needs to be; try packing Meteor yourself.

How to set node path for nodejs (Ubuntu)

I'm trying to setup nodejs to access a postgres database. What I've done so far is the following (https://gist.github.com/579814):
echo 'export PATH=$HOME/local/bin:$PATH' >> ~/.bashrc
. ~/.bashrc
mkdir ~/local
mkdir ~/node-latest-install
cd ~/node-latest-install
curl http://nodejs.org/dist/node-latest.tar.gz | tar xz --strip-components=1
./configure --prefix=~/local
make install # ok, fine, this step probably takes more than 30 seconds...
curl http://npmjs.org/install.sh | sh
then
git clone git://github.com/isaacs/npm.git
make
make install
so far, so good. However, when I try to install the postgres driver
npm install pg
I get the following:
node-waf configure build || true
Checking for program g++ or c++ : /usr/bin/g++
Checking for program cpp : /usr/bin/cpp
Checking for program ar : /usr/bin/ar
Checking for program ranlib : /usr/bin/ranlib
Checking for g++ : ok
Checking for node path : not found
Checking for node prefix : ok /usr/local
Checking for program pg_config : /usr/bin/pg_config
'configure' finished successfully (0.066s)
Waf: Entering directory `/home/christian/node_modules/pg/build'
[1/2] cxx: src/binding.cc -> build/default/src/binding_1.o
../src/binding.cc:3:25: fatal error: node_events.h: No such file or directory
compilation terminated.
Waf: Leaving directory `/home/christian/node_modules/pg/build'
Build failed: -> task failed (err #1):
{task: cxx binding.cc -> binding_1.o}
I've been looking around for setting the node path, although haven't found anything of help so far - probably also because I'm totally new to nodejs, so I'd be happy about any hint.
Now, you have NodeJS installed in your Ubuntu. You should set /etc/environment and load nodeJS path that can be executed by another users. For example:
NODE="/home/ubuntu/local/node"
NODE_PATH="/usr/local/lib/node_modules"
PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:$NODE/bin:$NODE/lib/node_modules"
#PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games"
Do this in bash:
echo 'export NODE_PATH=~/local/:~/local/node_modules' >> ~/.bashrc
before things are fixed you must use node 0.5.1 (you can use gitk to revert the tree to this version)
Auteur: Ryan Dahl <ry#tinyclouds.org> 2011-07-19 10:46:38
Auteur du commit: Ryan Dahl <ry#tinyclouds.org> 2011-07-19 10:46:38
Parent: 0a3fc1d9c8becc32c63ae736ca2b3719a3d03c5b (Remove StatWatcher's dep on C++ EventEmitter)
Enfant: 061ce7b0ac370c8a5ae93d95ab7da171cbd488f0 (net_uv: Fix simple/test-http-expect-continue.js)
Branche: master, remotes/origin/master
Suit: v0.5.1
Précède: v0.5.2
Finally remove node::EventEmitter
I had the same problem.
The issue was that I was specifying a old version of PG in my package.js
After I removed the old version dependancy I was able to install PG without issue.