Re: Poudriere testport failure but manual jailed build success
On Tue, Mar 03, 2015 at 05:24:15PM -0800, Chris H wrote: > On Tue, 3 Mar 2015 23:37:30 +0100 Marin Bernardwrote > > > Hi, > > > > I've been banging my head for several days on what follows and I've come to > > the point where I have to get some help. Here's the point. > > > > I'm trying to port LizardFS (a distributed file system for Unix/Linux) on > > FreeBSD and I built a port candidate I would like to submit. But first I > > needed to be sure everything was OK, so I ran some tests. As of now: > > - The port builds fine on FreeBSD 10.1-RELEASE amd64 host. > > - portlint does not report any issue (on the same host as above) > > - port test (from porttools) happily validates the port (on the same host > > as above) - BUT poudriere fails to build the port. > > > > I'm using poudriere 3.1.1 on FreeBSD 11-Current, and failure occurs within a > > FreeBSD 10.1-RELEASE amd64 jail. > > > > What basically happens is that the build process runs fine until it reaches > > man page generation. There, a2x throws an error because xlstproc returns > > with > > return code 5 (= "error in the stylesheet"), whereas it shouldn't. What > > kills > > me here is that if I enter the jail after the failure and try to build the > > port manually, everything builds fine! You'll find poudriere log at the end > > of this message. > > > Any reason you couldn't simply lower the risk of failure based > on tools you have no control over; by simply creating a valid > man page to begin with? In other words; if the man is already > properly formatted groff/troff/mandoc (take your pick). You > wouldn't ever need to worry again. :) > > Just a thought, and hope it helps. I just ran into this one. Here is the fix. (months late) You need a LIBDEPENDS on libxslt.so:textproc/libxslt This probably should be documented in the porters handbook. > > --Chris > .. > > Diane -- - d...@freebsd.org d...@db.net http://www.db.net/~db ___ freebsd-ports@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
RE: Poudriere testport failure but manual jailed build success
Hi, I've been banging my head for several days on what follows and I've come to the point where I have to get some help. Here's the point. I'm trying to port LizardFS (a distributed file system for Unix/Linux) on FreeBSD and I built a port candidate I would like to submit. But first I needed to be sure everything was OK, so I ran some tests. As of now: - The port builds fine on FreeBSD 10.1-RELEASE amd64 host. - portlint does not report any issue (on the same host as above) - port test (from porttools) happily validates the port (on the same host as above) - BUT poudriere fails to build the port. I'm using poudriere 3.1.1 on FreeBSD 11-Current, and failure occurs within a FreeBSD 10.1-RELEASE amd64 jail. What basically happens is that the build process runs fine until it reaches man page generation. There, a2x throws an error because xlstproc returns with return code 5 (= error in the stylesheet), whereas it shouldn't. What kills me here is that if I enter the jail after the failure and try to build the port manually, everything builds fine! You'll find poudriere log at the end of this message. Any reason you couldn't simply lower the risk of failure based on tools you have no control over; by simply creating a valid man page to begin with? In other words; if the man is already properly formatted groff/troff/mandoc (take your pick). You wouldn't ever need to worry again. :) Hi Chris. Many thanks for answering me. I'm not sure I understand: are you suggesting to ship the final manpage with the port files instead of building it from scratch ? If so, is this considered good practice? I thought the build process of a port had to be as much dynamic as possible: was I wrong? -Marin ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Poudriere testport failure but manual jailed build success
Hi, I've been banging my head for several days on what follows and I've come to the point where I have to get some help. Here's the point. I'm trying to port LizardFS (a distributed file system for Unix/Linux) on FreeBSD and I built a port candidate I would like to submit. But first I needed to be sure everything was OK, so I ran some tests. As of now: - The port builds fine on FreeBSD 10.1-RELEASE amd64 host. - portlint does not report any issue (on the same host as above) - port test (from porttools) happily validates the port (on the same host as above) - BUT poudriere fails to build the port. I'm using poudriere 3.1.1 on FreeBSD 11-Current, and failure occurs within a FreeBSD 10.1-RELEASE amd64 jail. What basically happens is that the build process runs fine until it reaches man page generation. There, a2x throws an error because xlstproc returns with return code 5 (= error in the stylesheet), whereas it shouldn't. What kills me here is that if I enter the jail after the failure and try to build the port manually, everything builds fine! You'll find poudriere log at the end of this message. Here are the things I tried: 1. As the port uses cmake I thought something broke in the makefiles. It appears I'm able to manually build the port using either: - The makefiles left behind by poudriere in the /wrkdirs tree (this confirms that cmake worked as expected) - A new set of Makefiles in another dir 2. I found several examples of xlstproc throwing errors when it can't get some resources from the internet (mainly DTD files). I made a patch to force xsltproc to run with the --nonet option, but the build failed the same. 3. I also tried to disable parallel make jobs, without success. I'll try to build the port from FreeBSD 10.1-RELEASE tomorrow. I don't think it will make a difference but I'll try anyway. Does anybody have a clue about what's going on? Thanks for your time! Marin. BEGIN POUDRIERE LOG build started at Tue Mar 3 22:41:59 CET 2015 port directory: /usr/ports/sysutils/lizardfs-admin building for: FreeBSD 101Ramd64-development 10.1-RELEASE-p6 FreeBSD 10.1-RELEASE-p6 amd64 maintained by: ma...@olivarim.com ident warning: no id keywords in /usr/local/poudriere/data/.m/101Ramd64-development/ref//usr/ports/sysutils/lizardfs-admin/Makefile Makefile ident: Poudriere version: 3.1.1 Host OSVERSION: 1100062 Jail OSVERSION: 1001000 ---Begin Environment--- PKGREPOSITORY=/tmp/pkgs PACKAGES=/tmp/pkgs OSVERSION=1001000 UNAME_v=FreeBSD 10.1-RELEASE-p6 UNAME_r=10.1-RELEASE-p6 BLOCKSIZE=K MAIL=/var/mail/root STATUS=1 SAVED_TERM=xterm NO_WARNING_PKG_INSTALL_EOL=yes MASTERMNT=/usr/local/poudriere/data/.m/101Ramd64-development/ref FORCE_PACKAGE=yes PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/games:/usr/local/sbin:/usr/local/bin:/root/bin POUDRIERE_BUILD_TYPE=bulk OLDPWD=/usr/local/poudriere/ports/development/sysutils PWD=/ MASTERNAME=101Ramd64-development USER=root HOME=/root POUDRIERE_VERSION=3.1.1 TRYBROKEN=yes LOCALBASE=/usr/local PACKAGE_BUILDING=yes ---End Environment--- ---Begin OPTIONS List--- === The following configuration options are available for lizardfs-admin-2.5.4: DOCS=on: Build and/or install documentation === Use 'make config' to modify these settings ---End OPTIONS List--- --CONFIGURE_ARGS-- --End CONFIGURE_ARGS-- --CONFIGURE_ENV-- XDG_DATA_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work XDG_CONFIG_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work TMPDIR=/tmp XDG_DATA_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work XDG_CONFIG_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work TMPDIR=/tmp SHELL=/bin/sh CONFIG_SHELL=/bin/sh --End CONFIGURE_ENV-- --MAKE_ENV-- XDG_DATA_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work XDG_CONFIG_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work TMPDIR=/tmp XDG_DATA_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work XDG_CONFIG_HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work HOME=/wrkdirs/usr/ports/sysutils/lizardfs-admin/work TMPDIR=/tmp NO_PIE=yes SHELL=/bin/sh NO_LINT=YES PREFIX=/usr/local LOCALBASE=/usr/local LIBDIR=/usr/lib CC=cc CFLAGS=-O2 -pipe -fstack-protector -fno-strict-aliasing CPP=cpp CPPFLAGS= LDFLAGS= -fstack-protector LIBS= CXX=c++ CXXFLAGS=-O2 -pipe -fstack-protector -fno-strict-aliasing MANPREFIX=/usr/local BSD_INSTALL_PROGRAM=install -s -o root -g wheel -m 555 BSD_INSTALL_LIB=install -s -o root -g wheel -m 444 BSD_INSTALL_SCRIPT=install -o root -g wheel -m 555 BSD_INSTALL_DATA=install -o root -g wheel -m 0644 BSD_INSTALL_MAN=install -o root -g wheel -m 444 --End MAKE_ENV-- --PLIST_SUB-- DOCS= NO_DOCS=@comment CMAKE_BUILD_TYPE=release OSREL=10.1 PREFIX=%D LOCALBASE=/usr/local RESETPREFIX=/usr/local PORTDOCS= PORTEXAMPLES= LIB32DIR=lib
Re: Poudriere testport failure but manual jailed build success
On Tue, 3 Mar 2015 23:37:30 +0100 Marin Bernard ma...@olivarim.com wrote Hi, I've been banging my head for several days on what follows and I've come to the point where I have to get some help. Here's the point. I'm trying to port LizardFS (a distributed file system for Unix/Linux) on FreeBSD and I built a port candidate I would like to submit. But first I needed to be sure everything was OK, so I ran some tests. As of now: - The port builds fine on FreeBSD 10.1-RELEASE amd64 host. - portlint does not report any issue (on the same host as above) - port test (from porttools) happily validates the port (on the same host as above) - BUT poudriere fails to build the port. I'm using poudriere 3.1.1 on FreeBSD 11-Current, and failure occurs within a FreeBSD 10.1-RELEASE amd64 jail. What basically happens is that the build process runs fine until it reaches man page generation. There, a2x throws an error because xlstproc returns with return code 5 (= error in the stylesheet), whereas it shouldn't. What kills me here is that if I enter the jail after the failure and try to build the port manually, everything builds fine! You'll find poudriere log at the end of this message. Any reason you couldn't simply lower the risk of failure based on tools you have no control over; by simply creating a valid man page to begin with? In other words; if the man is already properly formatted groff/troff/mandoc (take your pick). You wouldn't ever need to worry again. :) Just a thought, and hope it helps. --Chris .. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org