For the last few versions of perl I am getting errors during the ebuild process and then the ebuild stops waitnig for CR. If I hit enter it continues and finishes without more errors. The system in general seems to be working fine after that (except for a few other bugs, that seem to be related somehow, see below). I get the same kind of problem with libperl too. This are the last few lines of output before the errors: ... <values.h> found. <gdbm.h> found. gdbm_open() found. Looking for extensions... What extensions do you wish to load dynamically? [B ByteLoader Cwd DB_File Data/Dumper Devel/DProf Devel/PPPort Devel/Peek Digest/MD5 Encode Fcntl File/Glob Filter/Util/Call GDBM_File I18N/Langinfo IO IPC/SysV List/Util MIME/Base64 NDBM_File Opcode POSIX PerlIO/encoding PerlIO/scalar PerlIO/via SDBM_File Socket Storable Sys/Hostname Sys/Syslog Time/HiRes Unicode/Normalize XS/APItest XS/Typemap attrs re threads threads/shared] What extensions do you wish to load statically? [none] Stripping down executable paths... Creating config.sh... Doing variable substitutions on .SH files... Extracting cflags (with variable substitutions) Extracting config.h (with variable substitutions) sed: can't read Configure: No such file or directory sed: can't read Makefile.SH: No such file or directory sed: can't read lib/ExtUtils/Install.pm: No such file or directory mv: cannot move `s' to `lib/ExtUtils/Install.pm': No such file or directory sed: can't read lib/ExtUtils/MM_Unix.pm: No such file or directory mv: cannot move `s' to `lib/ExtUtils/MM_Unix.pm': No such file or directory sed: can't read installperl: No such file or directory sed: can't read lib/Pod/Html.pm: No such file or directory mv: cannot move `s' to `lib/Pod/Html.pm': No such file or directory sed: can't read makedepend.SH: No such file or directory sed: can't read t/comp/cpp.aux: No such file or directory mv: cannot move `s' to `t/comp/cpp.aux': No such file or directory sed: can't read t/io/argv.t: No such file or directory mv: cannot move `s' to `t/io/argv.t': No such file or directory sed: can't read t/io/inplace.t: No such file or directory mv: cannot move `s' to `t/io/inplace.t': No such file or directory sed: can't read ext/GDBM_File/t/gdbm.t: No such file or directory mv: cannot move `s' to `ext/GDBM_File/t/gdbm.t': No such file or directory sed: can't read t/op/stat.t: No such file or directory mv: cannot move `s' to `t/op/stat.t': No such file or directory sed: can't read x2p/Makefile.SH: No such file or directory mv: cannot move `s' to `x2p/Makefile.SH': No such file or directory sed: can't read lib/termcap.pl: No such file or directory mv: cannot move `s' to `lib/termcap.pl': No such file or directory sed: can't read installman: No such file or directory sed: can't read lib/ExtUtils/Installed.pm: No such file or directory mv: cannot move `s' to `lib/ExtUtils/Installed.pm': No such file or directory sed: can't read lib/ExtUtils/Packlist.pm: No such file or directory mv: cannot move `s' to `lib/ExtUtils/Packlist.pm': No such file or directory sed: can't read t/io/iprefix.t: No such file or directory mv: cannot move `s' to `t/io/iprefix.t': No such file or directory sed: can't read ext/POSIX/Makefile.PL: No such file or directory mv: cannot move `s' to `ext/POSIX/Makefile.PL': No such file or directory sed: can't read lib/ExtUtils/t/Installed.t: No such file or directory mv: cannot move `s' to `lib/ExtUtils/t/Installed.t': No such file or directory USAGE: Configure [perlroot] [output-file] [embed.fnc] \nIf you want to use the 3b1 shared libraries, complete this script then read the header in 3b1cc. [Type carriage return to continue]\c sandbox: Caught signal 2 in pid 5988 ----------------------------------------------------- I would expect that the ebuild should not stop, and am concerned about the errors. I have re-emerged many times, including a complete emerge -e system && emerge -e world -- the error stays. This is one of a set of strange errors that nobody else seems to have. It always appears together with Bug 110182 and Bug 110188 and Bug 53778). Not sure how these are related, but all of my systems initially would build fine, and eventually after several weekly emerge -uD world, all four bugs appear on all of my machines. This has been happening for about a year or so now and is really driving me crazy :). If there is anything I can help with debuging these problems, I will absolutely do it. My emerge --info: Portage 2.0.53 (default-linux/x86/2005.0, gcc-3.4.4, glibc-2.3.5-r2, 2.6.14-gentoo-r2-p i686) ================================================================= System uname: 2.6.14-gentoo-r2-p i686 AMD Athlon(tm) XP 2100+ Gentoo Base System version 1.6.13 ccache version 2.3 [enabled] dev-lang/python: 2.3.5-r2, 2.4.2 sys-apps/sandbox: 1.2.12 sys-devel/autoconf: 2.13, 2.59-r6 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r1 sys-devel/binutils: 2.16.1 sys-devel/libtool: 1.5.20 virtual/os-headers: 2.6.11-r2 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CBUILD="i686-pc-linux-gnu" CFLAGS="-march=athlon-xp -O3 -pipe" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.4/env /usr/kde/3.4/share/config /usr/kde/3.4/shutdown /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /var/bind /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/init.d /etc/terminfo /etc/env.d" CXXFLAGS="-march=athlon-xp -O3 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoconfig ccache distlocks sandbox sfperms strict userpriv usersandbox" GENTOO_MIRRORS="http://128.213.5.34/gentoo/ http://mirror.datapipe.net/gentoo http://mirror.datapipe.net/gentoo http://gentoo.mirrors.pair.com/" LINGUAS="bg en ru" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="x86 3dnow X acl alsa apm arts artswrappersuid audiofile avi bash-completion berkdb bitmap-fonts bzip2 caps cdr crypt cups curl dvd dvdr ecc eds emboss encode esd exif expat fam ffmpeg flac flash foomaticdb gd gdbm gif glut gnome gphoto2 gpm gstreamer gtk gtk2 hal howl idea idn imagemagick imap imlib java jpeg kde lcms libg++ libwww mad maildir mikmod mmap mmx mng motif mozdevelop mozilla mozsvg mozxmlterm mp3 mpeg mysql ncurses nls nptl nptlonly ogg oggvorbis opengl oss pam pcre pda pdflib perl png ppds python qt quicktime quotas radeon readline real samba sasl sdl slang spell sse ssl svga tcpd tiff truetype truetype-fonts type1-fonts udev unicode usb vorbis winbind xine xml2 xmms xscreensaver xv xvid zlib linguas_bg linguas_en linguas_ru userland_GNU kernel_linux elibc_glibc" Unset: ASFLAGS, CTARGET, LANG, LC_ALL, LDFLAGS, PORTDIR_OVERLAY
Can you please confirm that it still happens with perl-5.8.7-r3 ?
Yes, still happening with perl-5.8.7-r3
Any chance you could try emerging it without userpriv and usersandbox? I'm just guessing...
Sorry to disappoint, same result -- waiting for CR.
Ok, please run: # ebuild /usr/portage/dev-lang/perl/perl-5.8.7-r3.ebuild clean fetch unpack # cd /var/tmp/portage/perl-5.8.7-r3/work/perl-5.8.7/ At this point I would like output of `ls -al` (please create as an attachment) # ebuild /usr/portage/dev-lang/perl/perl-5.8.7-r3.ebuild compile (Which should fail with your error, I guess) Then, still at /var/tmp/portage/perl-5.8.7-r3/work/perl-5.8.7/, I'd like you to: # sh -x ./config_h.SH > output 2>&1 At this point I'd like that output, and your config.sh Sorry for getting you through all of this :-)
Created attachment 80788 [details] ls -al output Output after these commands: # ebuild /usr/portage/dev-lang/perl/perl-5.8.7-r3.ebuild clean fetch unpack # cd /var/tmp/portage/perl-5.8.7-r3/work/perl-5.8.7/
Created attachment 80789 [details] config_h.SH output After ebuild ... compile stops waiting for CR, this is the output of: # sh -x ./config_h.SH > output 2>&1
So far, my guess is that that something gets b0rked during the creation of the Configure script, thus making configure think it should use hints/3b1.sh (wtf???). Right now I'm trying to figure out what goes wrong... Could you send the updated output of ebuild [...] compile? Also, I could be wrong :)
Created attachment 80790 [details] output of ebuild [...] compile
sed --version and your config.sh, please? Sorry for all that, I am nearly clueless right now :-(
Created attachment 80791 [details] config.sh
Do not worry, I appreciate your help with this. I am glad to do whatever is needed to help you diagnose the problem. In fact, if I can help you by running something more complex (e.g. debug something), please let me know. # sed --version GNU sed version 4.1.4 Copyright (C) 2003 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, to the extent permitted by law.
cat /etc/mtab please - is your tmp space mounted seperately? This really looks like a read/write issue to the mount point (but then, i'm just about as clueless as yuval other than to say this really doesn't look like a perl bug per se)
Created attachment 80792 [details] sh Configure -des output Configure output after: # cd ~ # ebuild /usr/portage/dev-lang/perl/perl-5.8.7-r3.ebuild clean fetch unpack # cd /var/tmp/portage/perl-5.8.7-r3/work/perl-5.8.7/ # sh Configure -des
Before we mark this upstream (since it was filed on rt.perl.org after the last, non-ebuild/portage test) a few questions popped in my head: What's root's shell set to? Have you "messed" with /etc/profile or the default shell's rc? My line of thought being that if you tweaked this (like, say, setting the shell to sash for example) then when the Configure script breaks out it might not know where sed was being set. On the command line, try: echo "Hi there" | sed -e "s/Hi/Hola/" and let us know if it says Hola there like it should. Can we see your /etc/hosts file? (been a bug in the past) Can you post the Policy.sh and myconfig that were generated?
I had the case insensitive flag enabled for grep. It was in /et/env.d/ which is later compiled into /etc/profile on a Gentoo system. A mistake I made long time ago... Now everything works just fine. Many thanks for the help and sorry for the stupid problem.
BAH!!!! And another thing to watch out for - thanks :)
