Bug#777849: Patch to fix FTBFS error for gcc5
On Thu, Jul 02 2015, Martin Michlmayr wrote: > * Walter Franzini [2015-07-02 08:00]: >> Just yesterday I've uploaded to mentors.debian.net a fix really >> similar to yours and I'm waiting for a sponsor. > > Your wish is my command. Thanks for the uploads! ciao -- walter franzini -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#777849: Patch to fix FTBFS error for gcc5
On Thu, Jul 02 2015, Potter, Tim (Cloud Services) wrote: > Hi there. Hi Tim, > I’ve made a small patch to debian/rules for the fhist package to build > correctly using gcc5. I’ve also tested with gcc 4.9.2 and there are > noadverse effects. Just yesterday I've uploaded to mentors.debian.net a fix really similar to yours and I'm waiting for a sponsor. thanks -- walter franzini -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#774436: 3.2.65-1+deb7u1~test NOT working
Hi, after a suspend to ram my laptop (fujitsu lifebook S761) started to misbehave as others has already reported. However even with a different kernel, 3.2.63 or 3.2.65-1+dev7u1~test the system halt after ~9min. Going back to the old 2.6.32 does not fix the problem. Even booting under Windows 7 has the same problem. After the first suspend to ram the fan is making noise as it's spinning at maximum speed. It "seems" something that persist across reboot and operating systems has been changed. There is something I should try to reset the system state? thank you -- walter franzini
Bug#147832: catching fclose errors.
tags 147832 + patch
bye
Hi,
I offer the attached patch to fix this bug. It uses a temporary file
and overwrite the servlist_.conf file only if fclose returns without
errors.
I've intentionally ignored the file permissions issue mentioned above.
HTH
--
Walter Franzini
http://aegis.stepbuild.org/
Index: xchat-2.8.8/src/common/servlist.c
===
--- xchat-2.8.8.orig/src/common/servlist.c 2012-08-20 15:34:54.0 +0200
+++ xchat-2.8.8/src/common/servlist.c 2012-08-20 15:59:53.607555834 +0200
@@ -1077,17 +1077,24 @@
ircserver *serv;
GSList *list;
GSList *hlist;
+ char *servlist = "servlist_.conf";
+ char *new_servlist;
+
#ifndef WIN32
int first = FALSE;
- snprintf (buf, sizeof (buf), "%s/servlist_.conf", get_xdir_fs ());
+ snprintf (buf, sizeof (buf), "%s/servlist_.conf.new", get_xdir_fs ());
if (access (buf, F_OK) != 0)
first = TRUE;
#endif
- fp = xchat_fopen_file ("servlist_.conf", "w", 0);
+ new_servlist = malloc (strlen(servlist) + 5);
+ strcpy (new_servlist, servlist);
+ strcat (new_servlist, ".new");
+
+ fp = xchat_fopen_file (new_servlist, "w", 0);
if (!fp)
- return FALSE;
+ goto failure;
#ifndef WIN32
if (first)
@@ -1140,17 +1147,25 @@
hlist = hlist->next;
}
- if (fprintf (fp, "\n") < 1)
- {
- fclose (fp);
- return FALSE;
- }
+ fprintf (fp, "\n");
list = list->next;
}
- fclose (fp);
- return TRUE;
+ if (0 == fclose (fp))
+ {
+#ifdef WIN32
+ unlink (servlist); /* win32 can't rename to an existing file */
+#endif
+ xchat_rename_file (new_servlist, servlist, 0);
+
+ free (new_servlist);
+ return TRUE;
+ }
+
+failure:
+ free(new_servlist);
+ return FALSE;
}
static void
Bug#643347: aegis: FTBFS: libaegis/http.cc:407:43: error: format not a string literal and no format arguments [-Werror=format-security]
peter green writes:
> Patching the actual compile failures was easy enough. I've attatched a
> patch for that. I'm not absoloutely sure of it's correctness though
> (apparently there are some weird ways printf can be used and I don't
> know if aegis is using any of them).
Thanks for the patch, I'll apply.
>
> Unfortunately after fixing the compile failures the build failed for
> me with test failures. The last screenful of output is below
>
> aegis: warning: chmod("/var/lib", 00755): Operation not permitted [user
>"nobody", group "nogroup"]
> NO RESULT when testing the aedist functionality (new project 154)
> CXX="g++" /bin/sh etc/test.sh -shell /bin/sh -run \
>test/02/t0266a-walt.sh test/02/t0266a-walt.ES
> su: Authentication failure
> (Ignored)
You are probably running the tests as root and it not supported.
--
Walter Franzini
http://aegis.stepbuild.org/
--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#553212: fhist: FTBFS on armel
Patrick Matthäi writes: [...] > But btw, the buildds are almost switched to gcc-4.4 now. > > Maybe uploading 1.18 and give it a try? > Just send me the new dsc link. ok. The package can be found on mentors.debian.net: - URL: http://mentors.debian.net/debian/pool/main/f/fhist - Source repository: deb-src http://mentors.debian.net/debian unstable main contrib non-free - dget http://mentors.debian.net/debian/pool/main/f/fhist/fhist_1.18-1.dsc ciao -- Walter Franzini http://aegis.stepbuild.org/ pgpvyIMSsTkem.pgp Description: PGP signature
Bug#553212: fhist: FTBFS on armel
Patrick Matthäi writes: [...] >> Is it ok to build-depends on gcc-4.4 on all platforms? >> >> ciao > > No, this has to be done by the buildd admins. Where can I find the right procedure? I've read the policy a bit, is Build-Depends: gcc-4.4 [armel], still the wrong thing to do? -- Walter Franzini http://aegis.stepbuild.org/ pgpwZeSBhqiQ1.pgp Description: PGP signature
Bug#553212: fhist: FTBFS on armel
Patrick Matthäi writes: > Hello, > > see: > https://buildd.debian.org/fetch.cgi?pkg=fhist;ver=1.17-1;arch=armel;stamp=1255548238 > > You package fails the tests on armel: [...] > make[1]: Entering directory > `/build/buildd-fhist_1.17-1-armel-k5lGfy/fhist-1.17' > /bin/sh test/00/t0001a.sh > Segmentation fault > FAILED test of basic fcomp functionality The problem is due to gcc-4.3, using gcc-4.4 everything works ok. Is it ok to build-depends on gcc-4.4 on all platforms? ciao -- Walter Franzini http://aegis.stepbuild.org/ pgp3HOPfUhe7i.pgp Description: PGP signature
Bug#553212: fhist - FTBFS
Bastian Blank writes: > There was an error while trying to autobuild your package: Thanks. Is it possible, I'm not a DD nor a DM, to have access on an s390 system to debug the problem? ciao -- Walter Franzini http://aegis.stepbuild.org/ pgpIbDob7MYN3.pgp Description: PGP signature
Bug#528807: FTBFS: applying patch 41_lesstif_cpp to ./ ... failed.
According to #458763, patches 40_lesstif_copy.dpatch and 41_lesstif_cpp.dpatch are no longer needed. It also has a patch attached to update debian/*. HTH -- Walter Franzini http://aegis.stepbuild.org/ pgpC7TVR8JBvm.pgp Description: PGP signature
Bug#549231: aegis: diff for NMU version 4.24-5.2
tags 549231 + patch thanks Hi, Attached is the diff for my aegis 4.24-5.2 NMU. -- walter franzini http://aegis.stepbuild.org/ diff -u aegis-4.24/debian/aegis.install aegis-4.24/debian/aegis.install --- aegis-4.24/debian/aegis.install +++ aegis-4.24/debian/aegis.install @@ -128,3 +128 @@ -usr/share/man/man1/aexver.1 usr/share/man/man5 - diff -u aegis-4.24/debian/changelog aegis-4.24/debian/changelog --- aegis-4.24/debian/changelog +++ aegis-4.24/debian/changelog @@ -1,3 +1,9 @@ +aegis (4.24-5.2) unstable; urgency=low + + * avoid installing aexver man page as part of the aegis package (Closes: #549231). + + -- Walter Franzini Thu, 01 Oct 2009 21:26:01 +0200 + aegis (4.24-5.1) unstable; urgency=low * Non-maintainer upload.
Bug#549231: aegis and aegis-tk: error when trying to install together
owner 549231 ! reassign 549231 aegis 4.24-5.1 retitle 549231 Bug#549231: aegis must not install aexver(1) man page stop [...] > automatic installation tests of packages that share a file and at the > same time do not conflict by their package dependency relationships has > detected the following problem: Thanks for the report, I'll take care of it. ciao -- Walter Franzini http://aegis.stepbuild.org/ pgpBbOWWGLIwJ.pgp Description: PGP signature
Bug#522336: aegis: diff for NMU version 4.24-5.1
+ removed cake from dependencies as it's not available. ++ added fhist, cook, cvs, subversion, cssc to BuildDepends in order to have + more test run. + * debian/postinst: aelock is now installed suid root. + * debian/aegis.install: install aesvt man page. + * debian/aegis-web.install: install aexver man page. + * added lintian-overrides for man pages warning. + * fix sccs test test/01/t0127a.sh. + * added patch to remove thigs left over by make clean. + + -- Walter Franzini Sat, 26 Sep 2009 14:30:43 +0200 + aegis (4.24-5) unstable; urgency=low * Fix pending l10n issues, thanks to Christian Perrier. @@ -418 +441,0 @@ - diff -u aegis-4.24/debian/po/es.po aegis-4.24/debian/po/es.po --- aegis-4.24/debian/po/es.po +++ aegis-4.24/debian/po/es.po @@ -2,117 +2,83 @@ -# Copyright (C) 2005 Software in the Public Interest +# Copyright (C) 2005, 2008 Software in the Public Interest # This file is distributed under the same license as the aegis package. # # Changes: -# - Initial translation -# César Gómez Martín +# - Initial translation +# César Gómez Martín , 2005 # +# - Updates +# Francisco Javier Cuadrado , 2008 +# +# +# Traductores, si no conoce el formato PO, merece la pena leer la +# documentación de gettext, especialmente las secciones dedicadas a este +# formato, por ejemplo ejecutando: +# info -n '(gettext)PO Files' +# info -n '(gettext)Header Entry' # -# Traductores, si no conoce el formato PO, merece la pena leer la -# documentación de gettext, especialmente las secciones dedicadas a este -# formato, por ejemplo ejecutando: -# info -n '(gettext)PO Files' -# info -n '(gettext)Header Entry' # Equipo de traducción al español, por favor, lean antes de traducir # los siguientes documentos: # -# - El proyecto de traducción de Debian al español -#http://www.debian.org/intl/spanish/ -#especialmente las notas de traducción en -#http://www.debian.org/intl/spanish/notas -# -# - La guía de traducción de po's de debconf: -#/usr/share/doc/po-debconf/README-trans -#o http://www.debian.org/intl/l10n/po-debconf/README-trans +# - El proyecto de traducción de Debian al español +# http://www.debian.org/intl/spanish/ +# especialmente las notas de traducción en +# http://www.debian.org/intl/spanish/notas +# +# - La guía de traducción de po's de debconf: +# /usr/share/doc/po-debconf/README-trans +# o http://www.debian.org/intl/l10n/po-debconf/README-trans # msgid "" msgstr "" -"Project-Id-Version: aegis\n" +"Project-Id-Version: aegis 4.24-5\n" "Report-Msgid-Bugs-To: ae...@packages.debian.org\n" "POT-Creation-Date: 2008-06-14 22:02+0200\n" -"PO-Revision-Date: 2005-10-03 18:28+0100\n" -"Last-Translator: César Gómez Martín \n" +"PO-Revision-Date: 2008-12-18 19:59+0100\n" +"Last-Translator: Francisco Javier Cuadrado \n" "Language-Team: Debian l10n spanish \n" "MIME-Version: 1.0\n" "Content-Type: text/plain; charset=utf-8\n" "Content-Transfer-Encoding: 8bit\n" -"X-Poedit-Language: Spanish\n" -"X-Poedit-Country: SPAIN\n" -"X-Poedit-SourceCharset: utf-8\n" #. Type: note #. Description #: ../aegis.templates:1001 -#, fuzzy -#| msgid "/var/lib/aegis is not on a local drive." msgid "/var/lib/aegis not on a local drive" -msgstr "/var/lib/aegis no está en una unidad de disco local." +msgstr "«/var/lib/aegis» no está en una unidad de disco local" #. Type: note #. Description #. Type: note #. Description -#: ../aegis.templates:1001 ../aegis.templates:2001 -msgid "" -"Aegis requires that /var/lib/aegis be owned by sys.sys and be set-gid, and " -"group writable." -msgstr "" -"Para que Aegis funcione correctamente es necesario que el directorio /var/" -"lib/aegis pertenezca al usuario «sys» y grupo «sys», tenga el bit «setgid» y " -"pueda ser escrito por los miembros del grupo." +#: ../aegis.templates:1001 +#: ../aegis.templates:2001 +msgid "Aegis requires that /var/lib/aegis be owned by sys.sys and be set-gid, and group writable." +msgstr "Aegis necesita que el propietario del directorio «/var/lib/aegis» sea «sys.sys» y tenga activado el bit «set-gid», y el grupo tenga permisos de escritura." #. Type: note #. Description #: ../aegis.templates:1001 -#, fuzzy -#| msgid "" -#| "Your /var/lib/aegis is on a remote partition, most likely an NFS server. " -#| "This could cause the postinst to fail, since we may not have permission " -#| "to change the owner/permissions of the directory on the remote server." -msgid "" -"Your /var/lib/aegis is on a remote partition, most likely an NFS server. "
Bug#522336: Intention to NMU
Luk Claes writes: [...] > Ok, I was just pointing out that you did not really mention the changes, > but only the bugs that got fixed and for this one I was interested in > how you fixed it... Ok. Changed the description. -- Walter Franzini http://aegis.stepbuild.org/ pgpSbwtw4KeGf.pgp Description: PGP signature
Bug#522336: Intention to NMU
Luk Claes writes: > Walter Franzini wrote: >> In a week I'll issue a (sposored) NMU with the following changelog: >> >> * FTBFS: test failure (Closes: #533921) >> * Linked with OpenSSL, seems to be a GPL violation (Closes: #522336) > > How did you fix these? BuildDepends: libcurl4-gnutls-dev Peter Miller (Aegis author) has expressed no objection permitting the link with openssl, however the work needed to add the proper clause to each file (~3000) is not yet started. ciao -- Walter Franzini http://aegis.stepbuild.org/ pgpplc1fnx8PG.pgp Description: PGP signature
Bug#522336: Intention to NMU
In a week I'll issue a (sposored) NMU with the following changelog: * FTBFS: test failure (Closes: #533921) * Linked with OpenSSL, seems to be a GPL violation (Closes: #522336) * FTBFS under pbuilder even building as non-root (Closes: #419472) * dependency does not exists (Closes: #540343) * bashism in /bin/sh script (Closes: #530042) * [INTL:es] Spanish debconf template translation for aegis (Closes: #509994) * removed cake from dependencies as it's not available. * added fhist, cook, cvs, subversion to BuildDepends in order to have more test run. -- Walter Franzini http://aegis.stepbuild.org/ pgpsXUS8GhsII.pgp Description: PGP signature
Bug#533921: aegis: FTBFS: test failure
Lucas Nussbaum writes: > Package: aegis > Version: 4.24-5 > Severity: serious > User: debian...@lists.debian.org > Usertags: qa-ftbfs-20090620 qa-ftbfs > Justification: FTBFS on amd64 > > Hi, > > During a rebuild of all packages in sid, your package failed to build on > amd64. > > Relevant part: >> CXX="g++" /bin/sh etc/test.sh -shell /bin/sh -run \ >> test/02/t0228a-matt.sh test/02/t0228a-matt.ES >> rss.expected.gif rss.served.gif differ: char 24, line 1 >> FAILED test of the aeget rssicon functionality (Compare gifs) > [...] >> Passed 276 of 277 tests. >> Failed 1 of 277 tests. >> make[1]: *** [sure] Error 1 Thank's for the report. The patch below should fix the problem. # # This change-set make t0228a-matt more robust wrt file type # identification. # Index: test/02/t0228a-matt.sh --- test/02/t0228a-matt.sh +++ test/02/t0228a-matt.sh @@ -135,7 +135,11 @@ $bin/test_base64 -nh -uu -i rss.gif.uue icon/rss.gif if test $? -ne 0 ; then fail; fi -echo "Content-Type: image/gif" > rss.expected.gif +# +# We need to invoke file here because the behaviour of libmagic change +# with versions and we need to replicate it exactly to have the test pass. +# +echo "Content-Type:" `file -ib icon/rss.gif` > rss.expected.gif echo "Content-Length: 360" >> rss.expected.gif echo >> rss.expected.gif cat icon/rss.gif >> rss.expected.gif -- Walter Franzini http://aegis.stepbuild.org/ pgpFBMlNfGIS3.pgp Description: PGP signature pgpTlyR7TLKw0.pgp Description: PGP signature
Bug#522336: aegis: Linked with OpenSSL, seems to be a GPL violation
Adrian Bunk writes: [...] > I haven't checked aegis thorougly, but in the best case it might be > enough to just switch to libcurl3-gnutls-dev. You are right, since Aegis does not call OpenSSL directly, however changing the license will fix the problem for all other distributions for free. -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgp5TZK6n9cjt.pgp Description: PGP signature
Bug#522336: aegis: Linked with OpenSSL, seems to be a GPL violation
Adrian Bunk writes: [...] > I didn't find any statement that all copyright holders of GPL'ed code > have given extra permission to link with OpenSSL. Thanks for pointing out the problem, we (aegis developers) do not notice it probably because Aegis does not require to link directly with openssl. I'll contact other members of the development team to decide actions to be taken. ciao -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgpnpywJSDkor.pgp Description: PGP signature
Bug#461034: Fixed?
Matt Kraai <[EMAIL PROTECTED]> writes: > Howdy, > > I believe that this problem is fixed. In particular, it appears to > have been fixed upstream during the 4.23 development series by change > 440: > > http://aegis.sourceforge.net/cgi-bin/aeget/aegis.4.23.C440/?menu > > I was unable to reproduce the problem using a pbuilder on an i386 > system. It has been reopened when 4.24-3 arrived in testing, but, yes, it is fixed. -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgp5A0qeQPzuK.pgp Description: PGP signature
Bug#486852: aegis - FTBFS: No result for 21 of 277 tests.
Riku Voipio <[EMAIL PROTECTED]> writes: [...] >> Is it possible to see the build logs for 4.24-1 to verify on which >> platform it failed? > > http://buildd.debian.org/build.cgi?pkg=aegis > > As you see, most archs now build, and the testsuite fails anymore on > arm, s390 and mips. Thanks. -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgpWL0ZSBp5iI.pgp Description: PGP signature
Bug#486852: aegis - FTBFS: No result for 21 of 277 tests.
Walter Franzini <[EMAIL PROTECTED]> writes: > Riku Voipio <[EMAIL PROTECTED]> writes: > >> | aefinish: cd /srv/buildd >> >> Other buildd: >> >> | aefinish: cd /nonexistent >> >> In some buildd's user doesn't have a home dir - not being able to >> write there is is not a reason to fail build. > > You are obviously right and the 4.24-2 package is supposed to fix that > problem. Maybe the problem is more subtle. [...] > Is it possible to know on which buildd server the buildd user does not > have the home dir? I've verified that if $HOME is placed in a directory not accessible aefinish fails. Can someone with access to the failing buildd servers (arm, mips and s390) verify if that is the case? ciao -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgpBjGutiA1yW.pgp Description: PGP signature
Bug#486852: aegis - FTBFS: No result for 21 of 277 tests.
Riku Voipio <[EMAIL PROTECTED]> writes: > | aefinish: cd /srv/buildd > > Other buildd: > > | aefinish: cd /nonexistent > > In some buildd's user doesn't have a home dir - not being able to > write there is is not a reason to fail build. You are obviously right and the 4.24-2 package is supposed to fix that problem. Maybe the problem is more subtle. Is it possible to see the build logs for 4.24-1 to verify on which platform it failed? Is it possible to know on which buildd server the buildd user does not have the home dir? -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgpaGSgiLnuqo.pgp Description: PGP signature
Bug#477231: Patch does not fix the build problem
Christian Perrier <[EMAIL PROTECTED]> writes: [...] > So, at least as a temporary solution, building without the testsuite > seems to be the best way to get the package in Debian lenny... As a *temporary* solution it's ok for me. In the meantime I will look at a change-set in the development branch [1] that should fix some debian related problem. [1] http://aegis.sourceforge.net/cgi-bin/aeget/aegis.4.25.C1001/?menu -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgp4PmK2jaCwZ.pgp Description: PGP signature
Bug#477231: Patch does not fix the build problem
Christian Perrier <[EMAIL PROTECTED]> writes: [...] > Well, what I wonder is the point of running the testsuite when > building the package. I understand that, as upstream author, you > introduce a testsuitethis is even a very good practice. > > However, I don't really see any point in running that testsuite during > builds. Moreover, from my tests, this is a very time-consuming (and > verbose) process. It's a procedure advised in the Aegis doc. I also assume that the package maintainer want to be sure that locally applied patch (not debian specific) did not change the aegis behavior and/or that aegis works correctly on the various platform debian supports. As an example of a problem catched by running the testsuite on hppa: http://permalink.gmane.org/gmane.linux.debian.ports.hppa/3767 [...] > I think it would be better to not run the testsuite. I'm not sure it will be the right choice. > The maintainer scripts seem to build with "make sure", which I guess > is responsible for the testsuite being used. > > Can you confirm this? Yes. -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgpxZrJlhn6ns.pgp Description: PGP signature
Bug#477231: Patch does not fix the build problem
Christian Perrier <[EMAIL PROTECTED]> writes: > Quoting Walter Franzini ([EMAIL PROTECTED]): >> Christian Perrier <[EMAIL PROTECTED]> writes: >> >> > As far as I have tested the proposed patch does not fix the build >> > problem... >> >> Strange, the test script in the patch was used to verify the >> correcteness of the fix. >> >> I will look at it again. >> >> Can you send me a sample of the errors? thanks. > > > This is actually more weird than I expected. > > The package properly builds on my system. However, it fails to build > in pbuilder (and will therefore fail to build on autobuilders). Then it's a different (and known) problem. I'm not sure what autobuilders are, however on http://buildd.debian.org/pkg.cgi?pkg=aegis I can see some succesfull build (ia64 and m68k). Is it the right page? If I'm not wrong pbuilder need to be run as root, because it need to populate /dev/. On the other hand Aegis is not supposed to be invoked by root. > So, trying to reproduce the failure means that you *have* to build the > package with pbuilder. I've looked in the past at fixing the problem of running the testsuite under pbuilder. *If* there is interest I can look at it again, however I may need help From someone with a good knowledge of pbuilder and/or fakeroot. A change-set related to this issue has been integrated the (upstream) development branch, maybe it can be backported. I'll investigate. -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgpsYPfWct6z2.pgp Description: PGP signature
Bug#477231: Patch does not fix the build problem
Christian Perrier <[EMAIL PROTECTED]> writes: > As far as I have tested the proposed patch does not fix the build > problem... Strange, the test script in the patch was used to verify the correcteness of the fix. I will look at it again. Can you send me a sample of the errors? thanks. -- Walter Franzini http://aegis.stepbuild.org/ PGP Public key ID: 1024D/CB3FEB43 Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43 Key server : http://www.keyserver.net pgpW5QwAXqLzG.pgp Description: PGP signature
Bug#477231: aegis - FTBFS: aefinish: chdir /srv/buildd: No such file or directory
Bastian Blank <[EMAIL PROTECTED]> writes:
> Package: aegis
> Version: 4.24-1
> Severity: serious
The attacched patch should fix the problem.
HTH
--
Walter Franzini
http://aegis.stepbuild.org/
PGP Public key ID: 1024D/CB3FEB43
Key fingerprint : FA26 C33B CAFF 7848 EFEB 7327 96AA 2D57 CB3F EB43
Key server : http://www.keyserver.net
MIME-Version: 1.0
Content-Type: application/aegis-patch
Subject: aegis.4.25 - aefinish vs. non-existent directory
Content-Name: aegis.4.25.C510.patch
Content-Disposition: attachment; filename=aegis.4.25.C510.patch
X-Aegis-Project-Name: aegis.4.25
X-Aegis-Change-Number: 510
#
# The aefinish command is now able to ipass a change even if the home
# directory of the user does not exists.
# My thanks to Ralph Smith for diagnosing the problem.
#
# Signed-off-by: Walter Franzini <[EMAIL PROTECTED]>
#
# Aegis-Change-Set-Begin
# QlpoMzFBWSZTWXvbKakAALJfgAAQUAf//0ECHI6/79/6QAH9tZiNYaJDRo0mm0m0BpRp6NNG
# oAYnqYagEbQiaPUZNAAABoADDQ0AAAGgBKaQoeJqEeUZTPKYp6R6IMmmTT0juOoUjAb4
# ZMxugnaTB9vpyOR5DumderO5+/Ty3evux/JiJVEhCEgKSROXS4fJFBHWw61gqjWZ9rYhkMko
# N+JXDkGplz5QzvtTrKqjI+093a8TPo59FVBHSB7SA9NGWG1IdBVJ6qnEq/eVmPMJ9T05VTuh
# B56BaOxb5w2sQobMCeSJUlopppGS3ajD2c6Ej5yeBJSUODDGbB/wkpJzg2x0ieXZOkRGfRLq
# NcJiqby2/609KtD2XxJs3E7y/HnhgzfRxdwwwPu9op6tO3FcvFElmMm/yjYWAnCxMVqBkuld
# Aj3GTdbyRsxDn0IDjxXMLoiVYxQpUYS+y6lpJzh84+ZIloOAw3MKeBIqrKhA2Q7baWJK6LKK
# 8QjccLl1OYisH+wPRcevrQYvzTWp0rbbSylDnbaZWr/iIRgEdEyY9Q7jKsL3v5QE+xRAXHPB
# uzZe92dnBL3Zat5jPTch0MpwRMjEvNFunPt7+2YYLZtyuJHyWbNfXVpaRrkjndTGaQRi2etg
# dfWCS39pa8LB7Zm0JlEThzrlqQjGQwNNGWszFRQXUHkZcC12KtAFYvICOila4WbXUlOUGmOc
# PpTbqpIJ3+clqwsF4F3JFOFCQe9spqQ=
# Aegis-Change-Set-End
#
Index: aefinish/finish.cc
--- aefinish/finish.cc
+++ aefinish/finish.cc
@@ -1,7 +1,7 @@
//
// aegis - project change supervisor
// Copyright (C) 2006-2008 Peter Miller
-// Copyright (C) 2006 Walter Franzini
+// Copyright (C) 2006, 2008 Walter Franzini
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
@@ -665,8 +665,12 @@
home()
{
const char *cp = getenv("HOME");
-if (!cp)
+
+os_become_orig();
+if (!cp || !os_isa_directory(nstring(cp).get_ref()))
cp = "/";
+os_become_undo();
+
return cp;
}
Index: test/02/t0269a-walt.sh
--- test/02/t0269a-walt.sh
+++ test/02/t0269a-walt.sh
@@ -0,0 +1,149 @@
+#!/bin/sh
+#
+# aegis - The "aegis" program.
+# Copyright (C) 2008 Walter Franzini
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or (at
+# your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along
+# with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+TEST_SUBJECT="aefinish vs. non-existent home"
+
+# load up standard prelude and test functions
+. test_funcs
+
+HOME=/path/not/existing
+export HOME
+
+#
+# test the aeipass functionality
+#
+workproj=$work/proj
+workchan=$work/chan
+
+AEGIS_PROJECT=example ; export AEGIS_PROJECT
+
+#
+# make a new project
+#
+activity="new project 39"
+aegis -npr $AEGIS_PROJECT -vers "" -dir $workproj > log 2>&1
+if test $? -ne 0 ; then cat log; no_result; fi
+
+#
+# change project attributes
+#
+activity="project attributes 46"
+cat > tmp << 'end'
+description = "A bogus project created to test the aefinish functionality.";
+developer_may_review = true;
+developer_may_integrate = true;
+reviewer_may_integrate = true;
+default_test_exemption = true;
+develop_end_action = goto_awaiting_integration;
+end
+if test $? -ne 0 ; then no_result; fi
+aegis -pa -f tmp > log 2>&1
+if test $? -ne 0 ; then cat log; no_result; fi
+
+#
+# add the staff
+#
+activity="staff 62"
+aegis -nd $USER > log 2>&1
+if test $? -ne 0 ; then cat log; no_result; fi
+aegis -nrv $USER > log 2>&1
+if test $? -ne 0 ; then cat log; no_result; fi
+aegis -ni $USER > log 2>&1
+if test $? -ne 0 ; then cat log; no_result; fi
+
+#
+# create a new change
+#
+activity="new change 73"
+cat > tmp << 'end'
+brief_description = "The first change";
+cause = internal_bug;
+end
+if test $? -ne 0 ; then no_result; fi
+aegis -nc 1 -f tmp -p $AEGIS_PROJECT > log 2>&1
+if test $? -ne 0 ; then cat log; no_result; fi
+
+#
+# begin development of a change
+#
+aegis -db 1 -dir $workchan > log 2>&1
+if test $? -ne 0 ; then cat log; no_resu
Bug#393624: aegis: FTBFS: test failed
On 10/30/06, Julien Danjou <[EMAIL PROTECTED]> wrote: reopen 393624 severity 393624 serious thanks Hello, There was a problem while autobuilding your package, and I don't think that simply a wishlist bug since it fails to build on a standard autobuilder. > Automatic build of aegis_4.22-2 on nasya by sbuild/sparc 0.50 > Build started at 20061030-0314 > ** > PASSED > CXX="g++" /bin/sh etc/test.sh -shell /bin/sh -run \ > test/00/t0002a-jerr.sh test/00/t0002a-jerr.ES > aegis: user root too privileged > FAILED test of multi-architecture test (new developer 236) This kind of problem should be solved with a previously sent patch. I will look also at the other problems. ciao walter -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#351888: pcmcia-cs: pcmcia-cs does not depend from kernel >= 2.6.13
Per Olofsson <[EMAIL PROTECTED]> writes: > tag 351888 moreinfo > thanks > > Walter Franzini: >> The kernel for testing is 2.6.12-10 and /etc/init.d/pcmcia refuse to >> start because it require at least 2.6.13 > > pcmcia-cs should always start _except_ if the kernel is 2.6.13 or > higher and udev is enabled. Are you sure that you haven't mixed it up > with pcmciautils? Could you try "/etc/init.d/pcmcia start" and send me > the output? I think you are right. I've seen an error message related to pcmcia and I've failed to notice it's from pcmciautils. I'm sorry. Walter Franzini Responsabile Servizi Professionali SysNet s.n.c. Via Dossi, 8 - 27100 Pavia - ITALIA http://www.sys-net.it -- Office: +39.0382.573859 (110) Mobile: +39.348.8979532 Email:[EMAIL PROTECTED] -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#351888: pcmcia-cs: pcmcia-cs does not depend from kernel >= 2.6.13
Package: pcmcia-cs Version: 3.2.8-6 Severity: grave The kernel for testing is 2.6.12-10 and /etc/init.d/pcmcia refuse to start because it require at least 2.6.13 -- System Information: Debian Release: testing/unstable APT prefers testing APT policy: (990, 'testing'), (500, 'unstable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.12-1-686 Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1) Versions of packages pcmcia-cs depends on: ii debconf [debconf-2.0] 1.4.67 Debian configuration management sy ii dmidecode 2.7-2 Dump Desktop Management Interface ii libc6 2.3.5-8GNU C Library: Shared libraries an ii lsb-base 3.0-14 Linux Standard Base 3.0 init scrip ii module-init-tools 3.2.2-1tools for managing Linux kernel mo ii modutils 2.4.27.0-3 Linux module utilities ii pcmciautils 012-1 PCMCIA utilities for Linux 2.6 ii psmisc22.1-1 Utilities that use the proc filesy Versions of packages pcmcia-cs recommends: ii udev [hotplug]0.081-1/dev/ and hotplug management daemo -- debconf information: pcmcia-cs/abort_upgrade: true pcmcia-cs/error_stopping: pcmcia-cs/abort_msg: pcmcia-cs/not_stopped: pcmcia-cs/fix_pcic: false pcmcia-cs/upgrade_note: pcmcia-cs/start_later: * pcmcia-cs/upgrade_action: always * pcmcia-cs/start_pcmcia: true pcmcia-cs/stop_pcmcia: true pcmcia-cs/run_probe: true Walter Franzini Responsabile Servizi Professionali SysNet s.n.c. Via Dossi, 8 - 27100 Pavia - ITALIA http://www.sys-net.it -- Office: +39.0382.573859 (110) Mobile: +39.348.8979532 Email:[EMAIL PROTECTED] -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#306184: aegis: FTBFS: failed test
Lars Wirzenius <[EMAIL PROTECTED]> writes: > I can build aegis from source on i386 in a sid chroot using > "dpkg-buildpackage -us -uc -rfakeroot", but not with "fakeroot > dpkg-buildpackage -us -uc" or "sudo pbuilder build". When the builds > fail, I see either of these kinds of errors. Because I'm new to debian, can you kindly explain me what differentiate the various commands? Thanks. [...] > I can make build logs if there is interest. Yes thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#306770: aegis: creates ~/integrate.log while building
In order to fix the problem, Jerry Pendergraft, the aeintegratq
author, has prepared the patch in the attached message.
--- Begin Message ---
It would be aeintegratq being run by some tests that is creating the log.
Easy fix. I will create the log in the aegis test directory if/when being
run by tests. Here is the patch and I will send Peter an aedist.
***
*** 152,158
hash_option("-c", $Option{change}, \%Ready)if
defined($Option{change});
# define some locations for logging and such
! if ( -d "$ComDir/integrations" ) {
$LogFile = "$ComDir/integrations/integrate.log";
}else{
$LogFile = "$ENV{'HOME'}/integrate.log";
--- 152,160
hash_option("-c", $Option{change}, \%Ready)if
defined($Option{change});
# define some locations for logging and such
! if ( defined( $ENV{'AEGIS_TEST_DIR'} ) ) {
! $LogFile = "$ENV{'AEGIS_TEST_DIR'}/integrate.log";
! }elsif ( -d "$ComDir/integrations" ) {
$LogFile = "$ComDir/integrations/integrate.log";
}else{
$LogFile = "$ENV{'HOME'}/integrate.log";
--
Jerry Pendergraft[EMAIL PROTECTED]
St. Jude Medical voice: 651-523-6935
1350 Energy Lane, Suite 110 fax: 651-644-7897
St Paul, MN 55108-5254 mobile: 651-491-0163
On Thu, 28 Apr 2005, Walter Franzini wrote:
> I'm now working on debian and monitoring aegis related bugs.
>
>
>
> SysNet - via Dossi,8 27100 Pavia Tel: +390382573859 Fax: +390382476497
>
>
--- End Message ---
Bug#306184: aegis: FTBFS: failed test
Roland Stigge <[EMAIL PROTECTED]> writes: > Package: aegis > Version: 4.20-3 > Severity: serious [...] > CXX="g++" /bin/sh etc/test.sh -shell /bin/sh -run \ > test/01/t0142a.sh test/01/t0142a.ES > aegis: project "foo": change 1: duplicate review > FAILED test of the review policy functionality (review pass 642) Under which login have you run the testsuite? Test 142 uses some fake username to simulate multiple review actions and I think one may clash with the one you used to run the test. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
