Freeze exception for quagga
Hello Please allow quagga 0.99.17-1 into testing. It fixes two security flaws which are hard to backport as they are quite big and deep in the core of the BGP code. We try, together with redhat, to convince upstream to provide patches for the last stable releasee but that might take a while and for testing it's just one minor revision although it does contain other stuff like bugfixes and enhancements apart from the mentioned security flaws. The package built fine so far and is 2 days old. Changelog is: quagga (0.99.17-1) unstable; urgency=high . * SECURITY: "This release provides two important bugfixes, which address remote crash possibility in bgpd discovered by CROSS team.": 1. Stack buffer overflow by processing certain Route-Refresh messages CVE-2010-2948 2. DoS (crash) while processing certain BGP update AS path messages CVE-2010-2949 Closes: #594262 bye, -christian- -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/8200372f16e0cd82a28b1b1d0f71f65a.squir...@mail3.westend.com
Quagga 0.99.10-1lenny1 acceptable?
Hallo May I upload quagga_0.99.10-1lenny1 to testing-proposed-updates to get it into lenny? Date: Sun, 04 Jan 2009 20:08:28 +0100 Source: quagga Binary: quagga quagga-doc Architecture: source all amd64 Version: 0.99.10-1lenny1 Distribution: testing-proposed-updates Urgency: low Maintainer: Christian Hammers Changed-By: Christian Hammers Description: quagga - BGP/OSPF/RIP routing daemon quagga-doc - documentation files for quagga Closes: 495232 Changes: quagga (0.99.10-1lenny1) testing-proposed-updates; urgency=low . * Fixed bug that caused routes which were added externally, e.g. by "ip route add", to be ignored by Quagga (thanks to Hannes Schulz). Closes: #495232 The bug was verified to be present in 0.99.10-1 and to be fixed in 0.99.10-1lenny1. The patch is just a one liner and has been reviewed on the quagga-dev mailing list. bye, -christian- -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Quagga 0.99.10-1lenny1 acceptable?
Hello Am Fri, 09 Jan 2009 00:57:39 +0100 schrieb Luk Claes : > Christian Hammers wrote: > > Hallo > > > > May I upload quagga_0.99.10-1lenny1 to testing-proposed-updates to > > get it into lenny? > > > > Date: Sun, 04 Jan 2009 20:08:28 +0100 > > Source: quagga > > Binary: quagga quagga-doc > > Architecture: source all amd64 > > Version: 0.99.10-1lenny1 > > Distribution: testing-proposed-updates > > Urgency: low > > Maintainer: Christian Hammers > > Changed-By: Christian Hammers > > Description: > > quagga - BGP/OSPF/RIP routing daemon > > quagga-doc - documentation files for quagga > > Closes: 495232 > > Changes: > > quagga (0.99.10-1lenny1) testing-proposed-updates; urgency=low > > . > >* Fixed bug that caused routes which were added externally, e.g. > > by "ip route add", to be ignored by Quagga (thanks to Hannes > > Schulz). Closes: #495232 > > > > The bug was verified to be present in 0.99.10-1 and to be > > fixed in 0.99.10-1lenny1. The patch is just a one liner and has been > > reviewed on the quagga-dev mailing list. > > Yes, please upload. It has been uploaded and built on all archs but is still in testing-proposed-updates. Please unblock! bye, -christian- -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Please unblock libphp-jpgraph_1.5.2-11+u1
Hello
After some days without reaction to a patch that fixes an RC bug in
libphp-jpgraph I now uploaded a NMU to unstable. Please unblock!
bye,
-christian-
$ debdiff libphp-jpgraph_1.5.2-11.dsc libphp-jpgraph_1.5.2-11+u1.dsc
diff -u libphp-jpgraph-1.5.2/src/jpgraph.php
libphp-jpgraph-1.5.2/src/jpgraph.php
--- libphp-jpgraph-1.5.2/src/jpgraph.php
+++ libphp-jpgraph-1.5.2/src/jpgraph.php
@@ -246,7 +246,9 @@
//user
//
if( isset($GLOBALS['php_errormsg']) ) {
-JpGraphError::Raise("General PHP
error:".$GLOBALS['php_errormsg']);
+// Disabled by Debian to prevent showing PHP5 warnings
+// ("Non-static method called statically")
+// JpGraphError::Raise("General PHP
error:".$GLOBALS['php_errormsg']);
}
//
diff -u libphp-jpgraph-1.5.2/debian/changelog
libphp-jpgraph-1.5.2/debian/changelog
--- libphp-jpgraph-1.5.2/debian/changelog
+++ libphp-jpgraph-1.5.2/debian/changelog
@@ -1,3 +1,14 @@
+libphp-jpgraph (1.5.2-11+u1) unstable; urgency=low
+
+ * Non-maintainer upload.
+ * Applied patch that prevented PHP5 compile time generated warning
+"Non-static method called statically" to appear on screen and corrupts
+the output of graphics. The usual error_reporting and display_errors
+had no effect as a custom error handler was installed by the library.
+Closes: #436289
+
+ -- Christian Hammers Sat, 10 Jan 2009 00:10:30 +0100
+
libphp-jpgraph (1.5.2-11) unstable; urgency=low
* Added support for php5(-gd), reviewed all example that seems to work
--
To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Transition plans for MySQL-5.1
Hello Release-Team We packaged MySQL-5.1 in experimental and reached a state where the packages could enter unstable. Along with the new upstream version comes a change from libmysqlclient15 to libmysqlclient16 which is a build-dep for a lot of packages. When would you want us to do the upload to unstable? Are there any other things we should consider except those on Wiki:TransitionBestPractices? BTW, in #450535 we were asked to think about symlinking libmysqlclient.so.16 to libmysqlclient_r.so.16 because it would mean less problems for e.g. Apache which uses other libraries who sometimes link against one and sometimes the other. But as libmysqlclient_r is lacking the global my_errno variable they are not binary compatible so we stick with two different libraries even if both have the same shared library symbol "libmysqlclient_16". Do you agree on that? bye, -christian- -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Transition plans for MySQL-5.1
Hello Nobody answered so I'm going to upload MySQL this evening and then file bug report with priority serious against the 142 packages currently depending on the old libmysqlclient15off. I tag them with user pkg-mysql-ma...@lists.alioth.debian.org usertag nn + libmysqlclient16 bye, -christian- Am Wed, 15 Jul 2009 22:09:33 +0200 schrieb Christian Hammers : > Hello Release-Team > > We packaged MySQL-5.1 in experimental and reached a state where the > packages could enter unstable. Along with the new upstream version > comes a change from libmysqlclient15 to libmysqlclient16 which is > a build-dep for a lot of packages. > > When would you want us to do the upload to unstable? > > Are there any other things we should consider except those on > Wiki:TransitionBestPractices? > > BTW, in #450535 we were asked to think about symlinking > libmysqlclient.so.16 to libmysqlclient_r.so.16 because it would mean > less problems for e.g. Apache which uses other libraries who sometimes > link against one and sometimes the other. But as libmysqlclient_r is > lacking the global my_errno variable they are not binary compatible so > we stick with two different libraries even if both have the same > shared library symbol "libmysqlclient_16". Do you agree on that? > > bye, > > -christian- > > > > > > -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Transition plans for MySQL-5.1
Am Sun, 26 Jul 2009 17:00:36 +0200 schrieb Cyril Brulebois : > Christian Hammers (25/07/2009): > > Nobody answered so I'm going to upload MySQL this evening and then > > file bug report with priority serious against the 142 packages > > currently depending on the old libmysqlclient15off. I tag them with > > user pkg-mysql-ma...@lists.alioth.debian.org > > usertag nn + libmysqlclient16 > > Given you're still building libmysqlclient15-dev, packages currently > needing transitioning will get built/rebuilt against the new mysql > version, and won't be able to transition until mysql is ready > (independently of the RC bugs about that mysql transition). Sounds > like a bad plan. :( Meanwhile I learned that binNMUs are enough as the package indeed ships with an empty libmysqlclient15-dev package so I closed the bugs. bye, -christian- signature.asc Description: PGP signature
Re: Transition plans for MySQL-5.1
Hello Am Thu, 30 Jul 2009 12:35:09 +0200 schrieb Michael Biebl : > Christian Hammers wrote: > > Am Sun, 26 Jul 2009 17:00:36 +0200 > > schrieb Cyril Brulebois : > > > >> Christian Hammers (25/07/2009): > >>> Nobody answered so I'm going to upload MySQL this evening and then > >>> file bug report with priority serious against the 142 packages > >>> currently depending on the old libmysqlclient15off. I tag them > >>> with user pkg-mysql-ma...@lists.alioth.debian.org > >>> usertag nn + libmysqlclient16 > >> Given you're still building libmysqlclient15-dev, packages > >> currently needing transitioning will get built/rebuilt against the > >> new mysql version, and won't be able to transition until mysql is > >> ready (independently of the RC bugs about that mysql transition). > >> Sounds like a bad plan. :( > > > > > > Meanwhile I learned that binNMUs are enough as the package indeed > > ships with an empty libmysqlclient15-dev package so I closed the > > bugs. > > > > Hi Christian, > > I maintain one of the packages (rsyslog) which build-depends on > libmysqlclient15-dev. > I plan to make a sourceful upload soonish, so I was wondering if I > should also change the build dependency from libmysqlclient15-dev to > libmysqlclient-dev. > > What's your recommendation as mysql maintainer? > Will the libmysqlclient15-dev transition package go away and if so, > when? You don't need to do a sourceful upload just for mysql because we will ask for Binary NMUs once MySQL has build on all archs. But if you do one anyway we recommend to use libmysqlclient-dev so you don't have to change once we upload MySQL-5.4 in two years or so :) bye, -christian- signature.asc Description: PGP signature
Please unblock mysql-dfsg-5.0 5.0.32-3etch1
Hello Please allow mysql-dfsg-5.0 version 5.0.32-3etch1 into testing. Changes: mysql-dfsg-5.0 (5.0.32-3etch1) testing-proposed-updates; urgency=high . * Backported upstream patch for a bug that crashed the server when using certain join/group/limit combinations. Users of the Joomla CMS seemed to be affected by this. Closes: #403721 * The debian-start script that runs on every server start now first upgrades the system tables (if neccessary) and then check them as it sometimes did not work the other way around (e.g. for MediaWiki). The script now uses mysql_update instead of mysql_update_script as recommended. Closes: 409780 * The old_passwords setting that is set according to a Debconf question is now written to /etc/mysql/conf.d/old_passwords.cnf instead directly to the conffile /etc/mysql/my.cnf which would be fobidden by policy (thanks to Robert Bihlmeyer). Closes: #409750 * Added bison to build dependencies. * Synced Debconf translations with 5.0.32-7. (the last point is a template change with translations so that there is no regression) thanks, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Please unblock mysql-dfsg-5.0 5.0.32-3etch1
On 2007-02-24 Steve Langasek wrote: > On Sat, Feb 24, 2007 at 12:48:19PM +0100, Christian Hammers wrote: > > Please allow mysql-dfsg-5.0 version 5.0.32-3etch1 into testing. > > No, 5.0.32-7 has already been accepted into testing. Sorry, I'm not sure > why you uploaded this to t-p-u; I think I said the only blocker for -6 was > the translation regression, which is precisely what was fixed in -7? I understood your concerns regarding the NDB fix from Monty Taylor for a "probably bug" as a blocker for a migration to testing, but 5.0.32-7 in testing is fine - thanks! :) bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please unblock mysql-dfsg-5.0 (CVE-2007-1420)
Hello @Security-Team: The only affected package is mysql-server-5.0 from testing and unstable as prior MySQL versions did not support the "information schema" meta database. @RM: Please unblock the following upload: mysql-dfsg-5.0 (5.0.32-7etch1) testing-proposed-updates; urgency=high . * SECURITY: CVE-2007-1420: Single Row Subselect DoS. Specially crafted subselect queries could crash the mysql server. Patch backported from upstream changeset 19685 (46_CVE-2007-1420_subselect_dos.dpatch). Closes: #414790. interdiff says: debian/patches/46_CVE-2007-1420_subselect_dos.dpatch|148+++ mysql-dfsg-5.0-5.0.32/debian/changelog | 9 + mysql-dfsg-5.0-5.0.32/debian/patches/00list | 1 3 files changed, 158 insertions(+) thanks, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Patch for libdbi-perl dataloss problem into stable-proposed-updates?
Hello stable release team
May I upload libdbi-perl_1.53-1etch1 to "stable-proposed-updates"?
It contains a backported 2-line patch for a "potential" dataloss
problem (confirmed in the changelog of 1.57) which turned out to be a
real problem as written by a user in the below quoted mail. I applied
only the second part of the below patch as suggested by the DBI author.
bye,
-christian-
On 2007-06-19 Tim Bunce wrote:
> Looks fine, though the DBI.xs change (assert) isn't needed and isn't in
> the DBI code.
>
> Tim.
>
> On Tue, Jun 19, 2007 at 12:43:53AM +0200, Christian Hammers wrote:
> > Hello
> >
> > As reported in the Debian bug tracking system. Can you comment if this
> > patch is OK?
> >
> > bye,
> >
> > -christian-
> >
> >
> >
> > From: Tim Cutts <[EMAIL PROTECTED]>
> > To: [EMAIL PROTECTED]
> > Subject: libdbi-perl 1.55
> > Date: Mon, 14 May 2007 11:48:26 +0100
> > X-Mailer: Apple Mail (2.752.2)
> >
> > Hi Christian,
> >
> > One of the changes introduced in DBI 1.55 is a very small fix which
> > fixes a bug which can really cause data loss (a problem with
> > selectall_arrayref). The bug was spotted by one of my colleagues
> > here (Rob Davies, whom Tim Bunce credits in the changelog). We have
> > actually seen it cause real data loss here, so it's not just a
> > theoretical problem.
> >
> > What I was wondering was whether there is any way we can get that fix
> > into Debian on etch's next point release? The patch to fix that
> > particular issue is as follows:
> >
> > diff -Naur DBI-1.54/DBI.xs DBI-1.54.patched/DBI.xs
> > --- DBI-1.54/DBI.xs 2007-02-22 01:36:15.0 +
> > +++ DBI-1.54.patched/DBI.xs 2007-04-26 09:38:00.66700 +0100
> > @@ -9,7 +9,7 @@
> >
> > #define IN_DBI_XS 1 /* see DBIXS.h */
> > #define PERL_NO_GET_CONTEXT
> > -
> > +#include
> > #include "DBIXS.h"/* DBI public interface for DBD's written in
> > C */
> >
> > # if (defined(_WIN32) && (! defined(HAS_GETTIMEOFDAY)))
> > @@ -3104,6 +3104,9 @@
> > outitems = perl_call_sv(isGV(imp_msv) ?
> > (SV*)GvCV(imp_msv) : imp_msv,
> > (is_DESTROY ? gimme | G_EVAL | G_KEEPERR : gimme) );
> > }
> > +
> > + assert(outitems >= 0);
> > +
> > SPAGAIN;
> >
> > /* XXX restore local vars so ST(n) works below */
> > diff -Naur DBI-1.54/Driver.xst DBI-1.54.patched/Driver.xst
> > --- DBI-1.54/Driver.xst 2007-02-22 01:36:15.0 +
> > +++ DBI-1.54.patched/Driver.xst 2007-04-26 15:15:47.08400
> > +0100 @@ -121,6 +121,7 @@
> > }
> > else {
> > sth = dbixst_bounce_method("prepare", 3);
> > +SPAGAIN; SP -= items; /* because stack might have been
> > realloc'd */ if (!SvROK(sth))
> > XSRETURN_UNDEF;
> > }
> > @@ -157,6 +158,7 @@
> > else {
> > /* --- prepare --- */
> > sth = dbixst_bounce_method("prepare", 3);
> > +SPAGAIN; SP -= items; /* because stack might have been
> > realloc'd */ if (!SvROK(sth)) {
> > if (is_selectrow_array) { XSRETURN_EMPTY; } else
> > { XSRETURN_UNDEF; } }
signature.asc
Description: PGP signature
MySQL transition 5.0->5.1 with new libmysqlclient16
Hello release team While preparing 5.1.21beta I noticed that the client libray soname version has been raised which was not planned when they first started the 5.1 branch. .plan: * During the next 1-2 weeks I will upload the new packages to experimental to ensure that they at least build on all architectures. * I build PHP, Python and Perl against it to check if there are any obvious incompatibilities. * After your "go" I upload to unstable and file bugs against all build-depending packages. * If you want a deadline I NMU remaining build-depending packages afterwards. bye, -christian- signature.asc Description: PGP signature
Re: MySQL transition 5.0->5.1 with new libmysqlclient16
On 2007-08-27 Steve Langasek wrote: > Hi Christian, > > On Tue, Aug 28, 2007 at 12:53:10AM +0200, Christian Hammers wrote: > > Hello release team > > > While preparing 5.1.21beta I noticed that the client libray soname > > version has been raised which was not planned when they first started the > > 5.1 branch. > > > .plan: > > * During the next 1-2 weeks I will upload the new packages to > > experimental to ensure that they at least build on all architectures. > > * I build PHP, Python and Perl against it to check if there are any > > obvious incompatibilities. > > * After your "go" I upload to unstable and file bugs against all > > build-depending packages. > > * If you want a deadline I NMU remaining build-depending packages > > afterwards. > > My only request concerning your proposal is that the -dev package name be > left alone, so that existing packages which build-depend on > libmysqlclient15-dev can be fixed with a simple rebuild on the autobuilders > instead of forcing everyone to do sourceful uploads of their packages to > change build-dependencies. The greater the number of reverse-dependencies a > library has, the more important this is for making a smooth transition > possible. > > Unless you were planning on keeping libmysqlclient15off around in unstable > in parallel? I never want to support two versions in parallel (as with 12+14) again. As putting headers from libmysqlclient.so.16 into libmysqlclient15-dev sounds ugly, can I change the current Package: libmysqlclient15-dev Provides: libmysqlclient-dev to Package: libmysqlclient-dev Provides: libmysqlclient15-dev Replaces: libmysqlclient15-dev Conflicts: libmysqlclient15-dev Or does this not work for Build-Depends? bye, -christian- signature.asc Description: PGP signature
Re: Patch for libdbi-perl dataloss problem into stable-proposed-updates?
Hello Please accept libdbi-perl_1.53-1etch1 which I've uploaded to stable. http://lists.debian.org/debian-release/2007/06/msg00175.html > It contains a backported 2-line patch for a "potential" dataloss > problem (confirmed in the changelog of 1.57) which turned out to be a > real problem as written by a user in the below quoted mail. I applied > only the second part of the below patch as suggested by the DBI author bye, -christian- signature.asc Description: PGP signature
Re: Do not make gratuitous source uploads just to provoke the buildds!
Hello Wouter On 2005-03-16 Wouter Verhelst wrote: > That's not to say that a request to prioritize a package is to be > ignored; however, the power of deciding which packages get built first > should be with those that actually build the packages, rather than with > those who want their packages to be built. The former are expected to be > following the larger picture; the latter are not. Given that the only really relevant thing is when the package enters testing, which already can be changed by the maintainers, the only thing a maintainer can wish is to have his package build within the 2 or 5 days, or? As only 2 days might be a problem, wouldn't just prefering high+security uploads be enough to make everybody happy without disrupting the buildd order too much? friendly, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
RM: sarge-ignore requests for two woody-only bug reports
Hello If it eases the work of the RM or the people who check sarge's security via bugscan mails, you can tag #296674 and #285276 (merged) as "sarge-ignore". It's a low risk mysql vulnerability issue (CAN-2004-0956 and CAN-2004-0957) that is already fixed in sarge/sid, has a patch and waits for the security team's comments. (btw: this one is not related to the MySQL bugs published this month) bye, -christian- pgpiqmZ4c8QuW.pgp Description: PGP signature
t-p-u: quagga_0.98.3-6
Hello As suggested by todays freeze announcement (juhuu!) I hereby tell you of an uploaded a bugfix for Quagga to t-p-u as unstable will get the new upstream release 0.99.1 tomorrow which may or may not be suiteable but is at least not important for Sarge. bye, -christian- quagga (0.98.3-6) testing-proposed-updates; urgency=high * Removed "Recommends kernel-image-2.4" as aptitude then installes a kernel-image for an arbitrary architecture as long as it fullfill that recommendation which can obviously fatal at the next reboot :) Also it is a violation of the policy which mandates a reference to real packages (thanks to Holger Levsen). Closes: #307281 $ interdiff -z quagga_0.98.3-5.diff.gz quagga_0.98.3-6.diff.gz | diffstat changelog | 11 +++ control |1 - 2 files changed, 11 insertions(+), 1 deletion(-) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please allow mysql (4.0.24-8, 4.1.11-3) into Sarge!
Hello
Please let mysql-dfsg (4.0.24-8) and mysql-dfsg-4.1 (4.1.11-3) into
testing.
Both were blocked during the free announcement by bug #307473 which
could lead to the accidental remove of all mysql databases if one of
the packages is removed but not purged, the other one is installed and
then the first package completely purged as both use /var/lib/mysql
as storage path and ask "do you want to remove everything" when purging.
Even worse for the old Woody packages which asked this question at
install-time and saved the answer in Debconf.
For mysql-dfsg-4.1 there was also a problem with a cronscript that was not
beeing installed as dh_installcron did not liked the name (#302712).
Both uploads were made to unstable and built fine on all architectures.
thanks,
-christian-
The changelogs for both versions (also including the parts that probably
would have entered testing if not beeing blocked) are:
mysql-dfsg-4.1 (4.1.11-3) unstable; urgency=high
* The "do you want to remove /var/lib/mysql when purging the package" flag
from old versions is removed once this package is beeing installed so
that purging an old Woody mysql-server package while having a
mysql-server-4.1 package installed can no longer lead to the removal of
all databases. Additionaly clarified the wording of this versions Debconf
template and added a check that skips this purge in the postrm script
if another mysql-server* package has /usr/sbin/mysqld installed.
(thanks to Adrian Bunk for spotting that problem) Closes: #307473
* Cronfile was not beeing installed as the filename was not in the
correct format for "dh_installcron --name" (thanks to Tomislav
Gountchev). Closes: #302712
-- Christian Hammers <[EMAIL PROTECTED]> Sat, 23 Apr 2005 22:55:15 +0200
mysql-dfsg-4.1 (4.1.11-2) unstable; urgency=low
* Sean Finney:
- don't freak out if we can't remove /etc/mysql during purge.
- debian/rules clean works again.
* Christian Hammers:
- Fixed typo in README.Debian (thanks to Joerg Rieger). Closes: #304897
- Completely removed the passwordless test user as it was not only
insecure but also lead to irritations as MySQL checks first the
permissions of this user and then those of a password having one.
See bug report from Hilko Bengen for details. Closes: #301741
-- Christian Hammers <[EMAIL PROTECTED]> Sat, 16 Apr 2005 15:55:00 +0200
mysql-dfsg (4.0.24-8) unstable; urgency=high
* The "do you want to remove /var/lib/mysql when purging the package" flag
from old versions is removed once this package is beeing installed so
that purging an old Woody mysql-server package while having a
mysql-server-4.1 package installed can no longer lead to the removal of
all databases. Additionaly clarified the wording of this versions Debconf
template and added a check that skips this purge part if another
mysql-server* package is in status 'installed' or 'hold'.
(thanks to Adrian Bunk for spotting that problem) Closes: #307473
-- Christian Hammers <[EMAIL PROTECTED]> Thu, 5 May 2005 04:12:00 +0200
mysql-dfsg (4.0.24-7) unstable; urgency=low
* Sean Finney:
- don't freak out if we can't remove /etc/mysql during purge. thanks
to Michael Ablassmeier <[EMAIL PROTECTED]> for reporting this
(closes: #305332).
- dpatch fix so debian/rules clean works again. thanks to
Falk Siemonsmeier <[EMAIL PROTECTED]> for
reporting this (closes: #304303).
* Christian Hammers
- Fixed typo in README.Debian (thanks to Joerg Rieger). Closes: #304897
- Completely removed the passwordless test user as it was not only
insecure but also lead to irritations as MySQL checks first the
permissions of this user and then those of a password having one.
See bug report from Hilko Bengen for details. Closes: #301741
-- Christian Hammers <[EMAIL PROTECTED]> Sat, 23 Apr 2005 17:45:40 +0200
mysql-dfsg (4.0.24-6) unstable; urgency=low
* Corrected minor issue in Debconf template regarding skip-networking
(thanks to Isaac Clerencia). Closes: #303417
* Made dependency to gawk unnecessary (thanks to Zoran Dzelajlija).
Closes: #302284
-- Christian Hammers <[EMAIL PROTECTED]> Fri, 8 Apr 2005 00:52:28 +0200
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RFC on mysql 4.1 in sarge
Hello [1st issue - dpkg symlink bug workaround] On 2005-05-18 Roberto C. Sanchez wrote: > Quoting sean finney <[EMAIL PROTECTED]>: > > > so at this point, we're not sure what to do to cover this last problem, > > as we have no guarantee the preinst of mysql-server-4.1 will even run > > before mysql-server/woody is removed. the only fix we can think of is > > to remove the two directories from the files.list of the woody package. > > > > so we've come up with three options, none of which are great: > > > > > I may be misunderstanding what you are saying. But, I think that if you > create a package called mysql-server-4.1-upgrage (or something else > suitable) and then you make you make mysql-server-4.1 predepend on it, then > mysql-server-4.1-upgrade can check for the existence of the symlinks. No, predepends can be used to ensure that some other package has already been completely installed i.e. postinst had been executed before the depending package even gets unpackaged. But that won't help here. apt-get calculates that mysql-server (3.23) has to be removed in order to get mysql-server-4.1 installed so it would first completely remove mysql-server (3.23) and *then* maybe install your temporary package which would be as useless as our current preinst hack. As a direct upgrade from mysql-server (3.23) to mysql-server-4.1 without a prior upgrade to mysql-server (4.0) is quite unlikely, I would like to see our last upload, which in addition has some notes in README.Debian and the Debconf installation notes regarding this, to be accepted for Sarge. Relevant versions are 4.1.11a-2 and 4.0.24-10. [2nd issue - statically linked db3] The new bug #308966 explains that mysql-server and libnss-db produce segfault crashes as mysql-server until now still had the obsolete BDB engine enabled which uses a statically linked local db3 version. Thus BDB support could now complete been removed as luckily support for BDB was not present on most architectures, disabled by default and being warned at startup for a while now and will most likely be removed in 5.0 upstream anyway. I have a package ready (4.1.11a-3), do you want it for Sarge, Steve, or do we rather live with this bug? bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RFC on mysql 4.1 in sarge
Hello On 2005-05-19 Steve Langasek wrote: ... > > so we've come up with three options, none of which are great: > > > 1 the most recenty woody security update caused problems for some > > people, and there's a package already waiting to go in to fix this > > problem. we could put a fix into the woody mysql-server package into > > this package before the security team handles it. > > 2 if there's going to be a final woody point release, we could put a > > fixed version in there > > 3 give up on trying to fix it, assume that symlinks might get lost, and > > put something in a README file telling users what they have to do > > in order to fix up their database after restoring the symlinks. ... > I see the same three options. Joey has said he is working on a final woody > point release for the last weekend in May; you'll probably need to > coordinate with him and get something uploaded soon if you want to try for > this option. > > 3 does not sound so bad to me; it's arguably user error anyway to replace a > package-provided directory with a symlink in this manner, so having a corner > case of "partially upgraded woody system and installing mysql-server-4.1 and > messed with a package directory" is not the end of the world... I guess you missed my response to this thread so I here the relevant parts: As a direct upgrade from mysql-server (3.23) to mysql-server-4.1 without a prior upgrade to mysql-server (4.0) is quite unlikely, I would like to see our last upload, which in addition has some notes in README.Debian and the Debconf installation notes regarding this, to be accepted for Sarge. Relevant versions are 4.1.11a-2 and 4.0.24-10. [new: 2nd issue - statically linked db3] The new bug #308966 complains that mysql-server and libnss-db produce segfault crashes as mysql-server until now still had the obsolete BDB (aka BerkeleyDB) engine enabled which uses a statically linked local db3 version. Thus BDB support could now complete been removed as luckily support for BDB was not present on most architectures, disabled by default and being warned at startup for a while now and will most likely be removed in 5.0 upstream anyway. Or we could try a patch that Piotr Roszatycki is currently evaluating which would add versioned symbols to the bdb functions in MySQL. He already suspected though, that the patch would not be a oneliner.. So do you want 1. just stay with 4.1.11a-2 and the bug 2. an upload without BDB support as 4.1.11a-3 3. wait to decide upon the forthcoming "versioned symbols patch" later bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RFC on mysql 4.1 in sarge
Hello [1st RC issue - dpkg removes symlinks when upgrading from 3.23] As discussed before in some corner cases we can do nothing except for showing the user an explanation what happened which has been done in 4.1.11a-2 and 4.0.24-10. [2nd RC issue - statically linked db3] > The new bug #308966 complains that mysql-server and libnss-db produce > segfault crashes as mysql-server-4.1 until now still had the obsolete BDB (aka > BerkeleyDB) engine enabled which uses a statically linked local db3 version. This affects only mysql-server-4.1, not the 4.0 branch. Option 3 changed: 1. just stay with 4.1.11a-2 and the bug 2. an upload without BDB support as 4.1.11a-3 3. an upload with Piotr Roszatyckis 8 line patch that mainly only adds --with-uniquename=_mysql to the configure options and runs sed over one header file (the actual diff is ~50 lines because the patch is been saved as dpatch file like all other patches, too) I verified that the patched package - runs in those cases where the old segfaultet - contains the unique-fied symbols only in /usr/sbin/mysqld and not in the libraries which would be problematic - succeeds the mysql benchmark and some basic tests I did The patch itself can be reviewed at http://www.lathspell.de/linux/debian/mysql/mysql-dfsg-4.1-4.1.11a.debian.diff So I'm in favour of 3. but could live with the other choices, too. Steve? bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RFC on mysql 4.1 in sarge
Hello Steve On 2005-05-19 Steve Langasek wrote: > > Thus BDB support could now complete been removed as luckily support for > > BDB was not present on most architectures, disabled by default and > > being warned at startup for a while now and will most likely be removed in > > 5.0 upstream anyway. > > > I have a package ready (4.1.11a-3), do you want it for Sarge, Steve, or > > do we rather live with this bug? > > I would like to see bug #308966 fixed for sarge, yes. That has been fixed using the compile option suggested by Piotrs. > While you're at it, could you please move your CACHEDIR in the maintainer > scripts out of /var/cache and into /var/lib -- perhaps to Ok, done. 4.1.11a-3 has been uploaded to unstable and is now finally waiting for your approval to go into Sarge. thanks, -christian- P.S.: The complete list of changes up to the current version in Sarge: mysql-dfsg-4.1 (4.1.11a-3) unstable; urgency=high * Added patch from Piotr Roszatycki to compile the bundled db3 library that is needed for the BerkeleyDB support with versioned symbols so that mysqld no longer crashes when it gets linked together with the Debian db3 version which happens when e.g. using libnss-db. Closes: #308966 -- Christian Hammers <[EMAIL PROTECTED]> Thu, 19 May 2005 01:41:14 +0200 mysql-dfsg-4.1 (4.1.11a-2) unstable; urgency=high * Okay, the hackery with /var/lib/dpkg/info/mysql-server.list will not stand and is removed from the preinst of mysql-server. * New workaround for the symlink problem that does not involve mucking with dpkg's file lists is storing the symlinks in a temporary location across upgrades. As this sometimes fails since apt-get does not always call new.preinst before old.postrm, some remarks were added to README.Debian and the Debconf installation notes to minimize the inconvinience this causes. -- sean finney <[EMAIL PROTECTED]> Sun, 15 May 2005 10:25:31 -0400 mysql-dfsg-4.1 (4.1.11a-1) unstable; urgency=high * Added the "a" to the version number to be able to upload a new .orig.tar.gz file which now has the non-free Docs/ directory removed as this has been forgotten in the 4.1.11 release (thanks to Goeran Weinholt). Closes: #308691 * The Woody package listed /var/lib/mysql and /var/log/mysql in its /var/lib/dpkg/info/mysql-server.list. These directories are often replaced by symlinks to data partitions which triggers a dpkg bug that causes these symlinks to be removed on upgrades. The new preinst prevents this by removing the two lines from the .list file (thanks to Andreas Barth and Jamin W. Collins). See dpkg bug #287978. * Updated French Debconf translation (thanks to Christian Perrier). Closes: #308353 -- Christian Hammers <[EMAIL PROTECTED]> Thu, 12 May 2005 21:52:46 +0200 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: final report on all woody DSAs and sarge
Hello On Wed, Aug 11, 2004 at 10:42:03PM -0300, Joey Hess wrote: > We have now finished checking all the DSAs since woody's release, except > for a few that we didn't reach any conclusions on. That the following > DSAs seem to still be unfixed in sarge: > > mysql-dfsg 4.0.18-6 needed, have 4.0.18-5 for DSA-483 Thanks for checking all those packages. As this one is mine, I can estimate that it should be no problem as mysql-dfsg-4.0.20-10 will enter Sarge in 5 days except it will be further delayed by gcc. (to my understanding for optional packages uploading to "unstable" and not "proposed-updates" was still ok, tell me if not) bye, -christian-
Re: libiodbc2 arm binaries are missing
Hello On 2004-08-21 Domenico Andreoli wrote: > it looks that libiodbc2 arm binaries need to be uploaded. they are > required for the qt-x11-free build to happen (qt-x11-freeis is Dep-Wait > on arm because of this). There was a problem: although the build logs were "maybe-successful" and I saw no obvious error in them, someone on IRC told me that libtool/autoconf misdetected some shared library detection setting and therefore the package was hold back. Sadly nobody obened a bug report for this or similar so I waited a week or so before I got suspicious enough to ask why it's still not installed :-/ (BTW: who could I propose to file bugs in such situations? [EMAIL PROTECTED] Or did I get an email but missed it?) I uploaded version 3.51.2-5 today which has hopefully correct cdbs variables to update libtool and rerun autoconf/automake etc before each build. bye, -christian- pgpMAUOUZlt36.pgp Description: PGP signature
Re: FTFBS in sarge
On 2004-09-02 Bastian Blank wrote: > > This seems a problem due to you buildd environment. > > The chroots have no proc mounted. Nothing describes them as necesary. You are referring to the buildd chroots, right? From my experience I would say that all buildd chroots do have /proc mountet. One of my packages also needs /proc to build because ./configure needs it to figure out how to call "ps" among other things and mounting /proc is by far easier than patching a lot of similar packages using GNU autoconf. bye, -christian- pgplDwdImyq7y.pgp Description: PGP signature
Re: wanna-build only knows about older versions?
On 2005-07-06 Marc Haber wrote: > On Tue, Jul 05, 2005 at 03:02:53PM -0700, Steve Langasek wrote: > > Can we please stop playing this little game of implying that the people > > involved think it's somehow acceptable to not have security support for > > stable? > > If that's the case, why do the people involved tolerate no security > support for stable being available for more than a month now? Security is like health. Debian currently is ill. But not yet dead. Since Linuxtag is over Debian has started healing although these weeks surely leaves an ugly scar. bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Drop the minor release number
On 2005-07-08 Eduard Bloch wrote: > Debian 4.0 for etch, 4.1 for etch stable release 1, 4.2 for etch stable > release 2, 4.2a for etch stable release 2 with a minor CD mastering fix > (for example), etc.pp. I hate letters in version strings, what about: 4.0etch release 4.1etch minor release, i.e. first etch update by Joey 4.1.1 etch patchlevel release, i.e. fix to broken cd image bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Drop the minor release number
People, On 2005-07-08 Wouter Verhelst wrote: > > Counting numbers start at one. The first update would be the second > > release of etch. So really it should be 4.1 for the first release of > > etch and 4.2 for the second release and so on. > > Except that we're computer people, and we start counting at 0. Heck, we > even had a DebConf0 back in 2000. ... start getting used to the version schema of TeX and Latex :-) (TeX has an idiosyncratic version numbering system. Since version 3, updates have been indicated by adding an extra digit at the end, so that the version number asymptotically approaches . The current version is 3.141592. http://en.wikipedia.org/wiki/Version) bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Transition from libmysqlclient12 to libmysqlclient14
Hello As today the FTP maintainers have accepted the packages for MySQL-5.0, we now have three branches of MySQL: 4.0, 4.1 and 5.0beta. My goal would be to only have one version, 5.0 in Etch but as it would probably be a bad idea to rely on MySQL-5.0 maturing fast enough (and database guys are conservative guys anyways), I would like to proposed to use MySQL-4.1 as standard version for now and at least drop MySQL-4.0 completely. This means that "libmysqlclient12" will be removed and thus about 75 binary packages have to be rebuild. The RM's should give their ok here as it will AFAIK lead to problems if e.g. apache-auth-mysql uses a different library than php4-mysql or libnss-mysql etc... Luckily there should be no code change involved for these packages but it allowes them to connect to newer versions of the MySQL server which uses an enhanced password format during the connects. Is it reasonable and ok for the release team if I fill "normal" bugs against all packages using libmysqlclient12, upgrade those bugs to a higher severity in about two weeks and ask for permission to NMU two weeks after that? Regarding the transition of the server packages, it seems that I have to create an empty transition package called "mysql-server" which depends on "mysql-server-4.1" to get apt-get to force the user to upgrade their servers as just providing "mysql-server" in the mysql-server-4.1 package is not enough, right? bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bug#319042: mysql-dfsg still FTBFS for me
Hello On 2005-10-07 Daniel Schepler wrote: > Le Mercredi 05 Octobre 2005 23:43, vous avez écrit : > > On 2005-10-04 Daniel Schepler wrote: > > > According to the bug log, you thought you'd fixed the FTBFS in > > > mysql-dfsg, but I can still reproduce it in version 4.0.24-10 using a > > > pbuilder chroot created just today: > > > > There is no mysql-dfsg 4.0 in Debian unstable anymore, it has been > > superceeded by mysql-dfsg-4.1 (the existing packages are only left for the > > shared library, the server/client packages exist with 4.1 version numbers > > from the new package, too). > > The results of running "apt-cache showsrc mysql-dfsg" disagree with you here. > For the users' point of view it has been "superceeded", from the technical one it is still there as it provides libmysqlclient12. Once mysql-dfsg-4.1 has been installed in testing (needs to be built on sparc) and debian-release has not too much to do with other transitions I will ask the other maintainers to switch to libmysqlclient14 (from 4.1). > Anyway, I'm not going to play BTS tug-of-war here, but I still disagree with > the bug being closed. I'm CC'ing debian-release to get a more official > opinion on this. You could have got that from me, too :) Anyway, as I hopefully have made the situation a bit clearer now, let's just agree that the bug should not be closed now but that I will not spend my time on fixing it, too, as it'll be obsolete soon :-) bye, -christian-
libmysqlclient12 to libmysqlclient14 transition
Hello In Debian Sarge we had libmysqlclient12 from MySQL 4.0.x as main mysql library in use. In Etch I already have MySQL 4.1 (libmysqlclient14) and MySQL 5.0 (libmysqlclient15) which is still only release-candidate, so I'd like to drop support for the 4.0 branch completely. Note though, that libmysqlclient10-dev from mysql (3.23.x), maintained by Steve Langasek, which is also still present is left untouched because it was released under LGPL instead of "GPL+FLOSS Exceptions" and thus be preferred by some people. As we have recently introduced versioned symbols to libmysqlclient14 and -15 the transition "should be easy"(tm) and done just by mass filing a bug report and maybe after 2 weeks NMU the leftover packages. Maintainers: Recompiling with build-deps set to libmysqlclient14-dev instead of libmysqlclient12-dev should be enough. The last published release schedule speaks of 2006 as release date. It might be that MySQL 5.0 will be stable enough next summer to be shipped - maybe at the cost that of some more necessary bugfix updates but the benefit to only have one MySQL for the security support afterwards. Release Team, please comment :-) bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: libmysqlclient12 to libmysqlclient14 transition
Hello On 2005-10-18 Nathanael Nerode wrote: > > Release Team, please comment :-) > > Can you wait on this until the KDE/JACK transition goes through, new libpng > gets into etch, and new openssl does too? But I asked first back in july, JACK cheated the queue! :-))) Apropos, does the release team have a list of waiting transitions or should I just bounce my last mail every couple of weeks when I see that the list is getting quiet? I found this nice page on the web, maybe you want to use it if you don't already:http://wiki.debian.org/OngoingTransitions bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
libmysqlclient15 (easy-)transition ok now?
Hello May I submit wishlist bug reports to ca. 60 packages now that build-depend on an old version of libmysqlclientNN-dev? The proposed new libmysqlclient15 uses versioned symbols so it's not a problematic transition and can be done by the maintainers whenever they will make their upload anyway. (it's already registered at http://wiki.debian.org/OngoingTransitions) bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: should packages begin using /srv ?
Hello On 2005-12-21 Holger Levsen wrote: > On Monday 19 December 2005 12:05, Steve Langasek wrote: > > On Mon, Dec 19, 2005 at 01:23:24AM +0100, Holger Levsen wrote: > > > > is the release teams opinion and decission on this ? (Steve commented > > > > on IRC that it would be good to have this decission made by the team > > > > and in an archived media.) > > If no one else has any comments, I guess you can take that for silent > > agreement with my position. :) > > I'm not so sure, that "no comment" is widely and forever seen as consent. I'm against simple "me too" mails, but if more comments are what is needed to get this into our policy then you have mine. Having this kind of data in a different location (/opt, /home, /var, /usr/local, /data) on every customer server has bugged me since long :) > http://www.pathname.com/fhs/pub/fhs-2.3.html#SRVDATAFORSERVICESPROVIDEDBYSYSTEM > > like Isaac does, that is, webservers should use /srv/www, not /var/www. ack bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
All OngoingTransitions tags are not working any more?
Hello I've noticed that all "bugs" urls on wiki.debian.net -> OngoingTransitions show no open bug reports anymore. What's wrong here? Has anything changed with the usertags? bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#699047: unblock: quagga/0.99.21-4
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Please unblock package quagga (see changelog below) bye, -christian- -=(~/debian/quagga)$ debdiff quagga_0.99.21-3.dsc quagga_0.99.21-4.dsc diff -Nru quagga-0.99.21/debian/changelog quagga-0.99.21/debian/changelog --- quagga-0.99.21/debian/changelog 2012-06-08 01:29:47.0 +0200 +++ quagga-0.99.21/debian/changelog 2013-01-06 15:58:00.0 +0100 @@ -1,3 +1,12 @@ +quagga (0.99.21-4) unstable; urgency=medium + + * Fixed regression bug that caused OSPF "distribute-list" statements to be +silently ignored. The patch has already been applied upstream but there +has been no new Quagga release since then. +Thanks to Hans van Kranenburg for reporting. Closes: #697240 + + -- Christian Hammers Sun, 06 Jan 2013 15:50:32 +0100 + quagga (0.99.21-3) unstable; urgency=high * SECURITY: diff -Nru quagga-0.99.21/debian/patches/99_distribute_list.diff quagga-0.99.21/debian/patches/99_distribute_list.diff --- quagga-0.99.21/debian/patches/99_distribute_list.diff 1970-01-01 01:00:00.0 +0100 +++ quagga-0.99.21/debian/patches/99_distribute_list.diff 2013-01-06 15:30:27.0 +0100 @@ -0,0 +1,35 @@ +From 1e47fb668d0125cbb6d53f73469093e744164a47 Mon Sep 17 00:00:00 2001 +From: Christian Franke +Date: Tue, 04 Dec 2012 19:31:16 + +Subject: ospfd: fix argument parsing for distribute-list + +Use the correct argument for the protocol lookup in +ospf distribute-list commands. + +Signed-off-by: Christian Franke +Signed-off-by: Scott Feldman +--- +diff --git a/ospfd/ospf_vty.c b/ospfd/ospf_vty.c +index 3655cfe..862ef95 100644 +--- a/ospfd/ospf_vty.c b/ospfd/ospf_vty.c +@@ -6042,7 +6042,7 @@ DEFUN (ospf_distribute_list_out, + int source; + + /* Get distribute source. */ +- source = proto_redistnum(AFI_IP, argv[0]); ++ source = proto_redistnum(AFI_IP, argv[1]); + if (source < 0 || source == ZEBRA_ROUTE_OSPF) + return CMD_WARNING; + +@@ -6061,7 +6061,7 @@ DEFUN (no_ospf_distribute_list_out, + struct ospf *ospf = vty->index; + int source; + +- source = proto_redistnum(AFI_IP, argv[0]); ++ source = proto_redistnum(AFI_IP, argv[1]); + if (source < 0 || source == ZEBRA_ROUTE_OSPF) + return CMD_WARNING; + +-- +cgit v0.9.0.2 diff -Nru quagga-0.99.21/debian/patches/series quagga-0.99.21/debian/patches/series --- quagga-0.99.21/debian/patches/series
Bug#700250: unblock: fofix-dfsg/3.121-3
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Please unblock package fofix-dfsg * Removed Debian specific no_numeric_module patch as it introduced bugs and the program runs fine without it. Thanks to Alessio Gaeta. Closes: #696641 (include/attach the debdiff against the package in testing) -=(~/debian/fofix)$ debdiff fofix-dfsg_3.121-2.dsc fofix-dfsg_3.121-3.dsc diff -Nru fofix-dfsg-3.121/debian/changelog fofix-dfsg-3.121/debian/changelog --- fofix-dfsg-3.121/debian/changelog 2010-05-24 13:47:41.0 +0200 +++ fofix-dfsg-3.121/debian/changelog 2013-01-09 00:24:53.0 +0100 @@ -1,3 +1,11 @@ +fofix-dfsg (3.121-3) unstable; urgency=low + + * Removed Debian specific no_numeric_module patch as it introduced bugs +and the program runs fine without it. +Thanks to Alessio Gaeta. Closes: #696641 + + -- Christian Hammers Wed, 09 Jan 2013 00:24:42 +0100 + fofix-dfsg (3.121-2) unstable; urgency=low * Removed architecture tag [amd64] from python-pygame dependency as this diff -Nru fofix-dfsg-3.121/debian/patches/no_numeric_module fofix-dfsg-3.121/debian/patches/no_numeric_module --- fofix-dfsg-3.121/debian/patches/no_numeric_module 2010-02-21 12:34:45.0 +0100 +++ fofix-dfsg-3.121/debian/patches/no_numeric_module 1970-01-01 01:00:00.0 +0100 @@ -1,27 +0,0 @@ -# Import Numpy instead
Bug#597327: unblock: quagga/0.99.17-2
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Please unblock package quagga Only change: * Added Danisch Debconf translation (thanks to Joe Dalton). Closes: #596259 unblock quagga/0.99.17-2 -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.32-5-amd64 (SMP w/2 CPU cores) -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100918172541.4240.21880.report...@james.intern
New libmysqlclient transition necessary
Hello There is another libmysqlclient transition coming so I like to inform the release team about it. In the last month there already was one where I tried to get rid of libmysqlclient10/12/14 in favour of libmysqlclient15. Thanks to the versioned symbols this was no big deal and was almost finished. Then came 5.0.19... On 2006-03-13 Steve Langasek wrote: > On Tue, Mar 14, 2006 at 01:25:21AM +0100, Christian Hammers wrote: > > > During the last month I have build my libmysqlclient15 with > > shared symbols that looked in "objdump -T" like: > > 0013a154 gDO .bss 0004 MYSQL_5.0 my_dont_interrupt > > 00026d70 gDF .text 02fa MYSQL_5.0 my_strntoll_8bit > > 00015730 gDF .text 0025 MYSQL_5.0 my_no_flags_free > > > Now MySQL finally closed my bug report to them and provides symbols > > in their upstream source. Sadly they look like: > > f280 gDF .text 000b libmysqlclient_15 mysql_row_tell > > f4d0 gDF .text 0043 libmysqlclient_15 mysql_escape_string > > da30 gDF .text 00e1 libmysqlclient_15 > > mysql_slave_send_query ... > Yes, this is a backwards-incompatible ABI change. If libmysqlclient15 had > been present in sarge, such a change without a rename of the library > package would be a release-critical bug for etch; since it wasn't, it's > only severity: important, but either way all packages built against the > previous symbol versions would have a release-critical bug requiring a > rebuild. > > And changing the package name is actually the easiest way to make sure that > no RC-buggy reverse-dependencies are overlooked. Here are my plans, please comment :) 1. Package changes: * Change package names (without changing the library filename, of course): libmysqlclient15-> libmysqlclient15off libmysqlclient15-dev-> libmysqlclient15off-dev * libmysqlclient15off package gets: Conflict: libmysqlclient15 Replaces: libmysqlclient15 * modifying rules: dh_makeshlibs -plibmysqlclient15off -V'libmysqlclient15off (>= 5.0.19-1)' * Add a short NEWS.Debian and long README.Debian warning that software which was locally build against libmysqlclient15 has to be rebuild. 2. Upload now to experimental to check if all buildds eat it 3. Wait for the release-teams of to start the transition (or may I start ASAP?) 4. File bug reports against all packages with severity=important and user-tag "transition-libmysqlclient15-dev" (already present on wiki.debian.org/OngoingTransitions). 5. Upload to unstable with prio=medium 6. Immediately NMU packages that were already linked against libmysqlclient15 with prio=high (like libdbd-mysql-perl which mysql-server-5.0 depends on) 7. Keep an eye on the rest and NMU at will. bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: New libmysqlclient transition necessary
Hi On 2006-03-17 Bastian Blank wrote: > On Fri, Mar 17, 2006 at 01:50:41AM +0100, Christian Hammers wrote: > > 1. Package changes: > >* Change package names (without changing the library filename, of > > course): > > libmysqlclient15-dev-> libmysqlclient15off-dev > > The release team will kill you for that change. There are no source > interface changes and therefor no reason to change the package name. I didn't know about the possibility to do automated binary rebuilds. Else it would have been for cosmetic reasons and all maintainers would have had to change their Build-Depends anyway. But now no need for NMU and name changes, of course :) bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [Pkg-octave-devel] Re: hdf5 transition
On 2006-03-20 Rafael Laboissiere wrote: > > Do you mind if I ask why you feel such notification is important? > > Because I, for instance, was not aware of the bin-NMU until Thomas sent Steve, you probably should create a little "transitions FAQ" entry in the developers reference to not always have to answer to the same wrong proposals of developers who only do a transition once every couple of years :) bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: New libmysqlclient transition necessary
Hello On 2006-03-16 Steve Langasek wrote: > > > > During the last month I have build my libmysqlclient15 with > > > > shared symbols that looked in "objdump -T" like: > > > > 0013a154 gDO .bss 0004 MYSQL_5.0 my_dont_interrupt > > > > 00026d70 gDF .text 02fa MYSQL_5.0 my_strntoll_8bit > > > > 00015730 gDF .text 0025 MYSQL_5.0 my_no_flags_free > > > > > Now MySQL finally closed my bug report to them and provides symbols > > > > in their upstream source. Sadly they look like: > > > > f280 gDF .text 000b libmysqlclient_15 mysql_row_tell > > > > f4d0 gDF .text 0043 libmysqlclient_15 > > > > mysql_escape_string da30 gDF .text 00e1 > > > > libmysqlclient_15 mysql_slave_send_query > > ... > > > Yes, this is a backwards-incompatible ABI change. If libmysqlclient15 > > > had been present in sarge, such a change without a rename of the > > > library package would be a release-critical bug for etch; since it > > > wasn't, it's only severity: important, but either way all packages > > > built against the previous symbol versions would have a > > > release-critical bug requiring a rebuild. The source package mysql-dfsg-5.0 5.0.19-2 is now in unstable and build on all archs except m68k. It's now your turn to schedule the binNMU mechanism, right? The resulting binaries should end up linked against "libmysqlclient15off". bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Proposing sarge update: MySQL replication crash with multiupdate and subselects?
Hello Stable-Release-Team Is bug #383165 worth to be fixed by an upload to stable-proposed-updates? "mysql-server-4.1 in sarge has a bug that lets the server crash when a specially crafted query is executed on a host that replicates data to another mysql host. Getting replication to work again needs manual intervention and know how." The upstream patch[1] is very small, just the move of some lines a bit up inside the function. bye, -christian- 1: http://lists.mysql.com/internals/26123 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
ABI breakage in libmysqlclient15off_5.0.24-1 - binNMU possible?
Hello MySQL introduced a ABI incompatibility in libmysqlclient15off_5.0.24-1 which already caused a lot of segfaults and crashes esp. with the Perl bindings. A patch is available at least. Is it possible to schedule a binNMU rebuild for all packages depending on libmysqlclient15off without a package name change and just change the line "dh_makeshlibs -V"? I guess not as it was necessary back in 2006-03-15 at the last ABI breakage? bye, -christian- P.S.: I should start with calling it libmysqlclient15abi3 and so on :) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: ABI breakage in libmysqlclient15off_5.0.24-1 - binNMU possible?
On 2006-08-25 Steve Langasek wrote: > On Fri, Aug 25, 2006 at 10:25:25PM +0200, Christian Hammers wrote: > > > MySQL introduced a ABI incompatibility in libmysqlclient15off_5.0.24-1 > > which already caused a lot of segfaults and crashes esp. with the Perl > > bindings. A patch is available at least. > > > Is it possible to schedule a binNMU rebuild for all packages depending on > > libmysqlclient15off without a package name change and just change > > the line "dh_makeshlibs -V"? > > I guess not as it was necessary back in 2006-03-15 at the last ABI > > breakage? > > What do you mean, 'change the line "dh_makeshlibs -V"'? A binNMU can't > include any changes to the source, but this line shouldn't appear in the > packages that *depend* on libmysqlclient15off anyway (or at least it's not > relevant to this change). Sorry for being unclear, the dh_makeshlibs command is of course in my source package which I will have to upload anyway to revert the accidental ABI change. > Please fix the library first -- ideally with a package name change or by > restoring the ABI, Hm, yes, that "or" was exactly my question, in cases where an unintended ABI change has been around only for a week in unstable, is it enough to upload a fixed version and rebuild everything? > but at least with a bumped shlibs -- and then we can binNMU the packages > depending on it as needed. I upload 5.0.24-2 once it has been built here. It contains *no* package name nor soname change but just a fixed libmysqlclient.so.15 and an updated shlibs file. And a NEWS entry for apt-listchanges to warn people to rebuild binaries that were compiled locally last week. bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please accept new but transitional package from mysql-dfsg-5.0
Hello Two (merged) RC bugs would be closed if the latest upload of mysql-dfsg-5.0 which introduced a new transitional package, would be accepted... bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please unblock mysql-dfsg-5.0_5.0.30-3
Hello mysql-dfsg-5.0_5.0.30-1 is currently in testing. I would like to see version -2 there, too, as it contains an important bugfix and preferable now -3, too, which just adds some more Debconf translations. (the -2 version was not requested earlier as I was waiting for another upstream bugfix. It has not yet come so maybe there will be a request for -4 but in the meantime I didn't want to wait any longer with -2) The changelogs are below. If you have any objections or questions, please let me know. thanks, -christian- mysql-dfsg-5.0 (5.0.30-3) unstable; urgency=low * Updated Brazilian Debconf translation (thanks to Andre Luis Lopes). Closes: #403821 * Added Romanian Debconf translation (thanks to Stan Ioan-Eugen). Closes: #403943 * Updated Spanish Debconf translation (thanks to Javier Fernandez-Sanguino Pena). Closes: #404084 * Updated Galician Debconf translation (thanks to Jacobo Tarrio). Closes: #404318 * Updated Dutch Debconf translation (thanks to Vincent Zweije). Closes: #404566 * Updated Danish Debconf translation (thanks to Claus Hindsgaul). Closes: #405018 -- Christian Hammers <[EMAIL PROTECTED]> Thu, 21 Dec 2006 21:35:09 +0100 mysql-dfsg-5.0 (5.0.30-2) unstable; urgency=high * Fixed upstream regression in header files that lead to FTBFS for mysql-admin, mysql-query-browser and probably other pacakges. (thanks to Andreas Henriksson). Closes: #403081, #403082 * Fixed some upstream scripts by replacing /etc by /etc/mysql (thanks to Julien Antony). Closes: #401083 * Updated French Debconf translation (thanks to Christian Perrier). Closes: #401434 * Added Spanish Debconf translation (thanks to Javier Fernandez-Sanguino Pena). Closes: #401953 * Marked a Debconf question that is just a dummy and only internally used as not-needing-translation. Closes: #403163 * Fixed mysqlslowdump patch to not remove the usage() function (thanks to Monty Tailor). -- Christian Hammers <[EMAIL PROTECTED]> Sun, 3 Dec 2006 19:20:10 +0100 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please unblock mysql-dfsg-5.0.32-2
Hello
MySQL release another upstream version which fixes a lot of "can cause the
server to crash" type bugs which of course could potentially allow people to
cause data loss to other users and thus could be considered as DoS.
Also there was a bug that let "mysql_upgrade" crash which would probably a
nasty thing on sarge->etch upgrades.
I had to make a -2 upload as I inadvertently deleted a necessary .dpatch
file. The new "upstream" is now in unstable for 15 days without any reported
regressions. Alpha builds are still missing but this seems to be due
to a lack of redundancy in our buildd environment (*g*), the last couple of
upstream versions all worked fine on alpha.
bye & TIA,
-christian-
mysql-dfsg-5.0 (5.0.32-2) unstable; urgency=high
.
* The last upload suffered from a regression that made NDB totally
unusable and caused a dependency to libmysqlclient15-dev in the
mysql-server-5.0 package. The relevant 85_* patch was re-added again.
Closes: #406435
* Added lintian-overrides for an error that does not affect our packages.
There are now only warnings and not errors left.
mysql-dfsg-5.0 (5.0.32-1) unstable; urgency=high
.
* New upstream version.
* SECURITY: mysql_fix_privilege_tables.sql altered the
table_privs.table_priv column to contain too few privileges, causing
loss of the CREATE VIEW and SHOW VIEW privileges. (MySQL Bug#20589)
* SECURITY (DoS): ALTER TABLE statements that performed both RENAME TO
and {ENABLE|DISABLE} KEYS operations caused a server crash. (MySQL
Bug#24089)
* SECURITY (DoS): LAST_DAY('-00-00') could cause a server crash.
(MySQL Bug#23653)
* SECURITY (DoS): Using EXPLAIN caused a server crash for queries that
selected from INFORMATION_SCHEMA in a subquery in the FROM clause.
(MySQL Bug#22413)
* SECURITY (DoS): Invalidating the query cache (e.g. when using stored
procedures) caused a server crash for INSERT INTO ... SELECT statements that
selected from a view. (MySQL Bug#20045)
* Using mysql_upgrade with a password crashed the server. Closes:
#406229
* yaSSL crashed on pre-Pentium Intel and Cyrix CPUs. (MySQL Bug#21765)
Closes: #383759
* Lots of small fixes to the NDB cluster storage engine.
* Updated Japanese Debconf template (thanks to Hideki Yamane).
Closes: #405793
* Fixed comment regarding "mycheck" in debian-start (thanks to
Enrico Zini). Closes: #405787
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: why is alpha a release candidate?
Hello On 2007-01-17 Steve Langasek wrote: > On Wed, Jan 17, 2007 at 09:21:07AM -0800, Thomas Bushnell BSG wrote: > > So the release criteria require buildd redundancy. And yet, half the > > release candidate archs still don't have it. It gets marked in yellow > > on http://release.debian.org/etch_arch_qualify.html. > > Yes, it gets marked in yellow because this requirement has in practice been > waived as a hard requirement for release qualification Do we really have so few donations that we cannot afford to get at least two buildd per architecture (or four if security-buildd are separate) and at least one with ssh access for developers? ... > The point of the arch requirements is to facilitate a release, not to > penalize architectures. On the list I count seven (including m68k) architectures that are listed with incomplete buildd redundancy. This is not only about RC bugs while we're getting closer to the release but also important for DSAs once we have released. As we probably do have enough manpower, money, hardware and housing offerings but they just don't get used, would this be some issue where our project leader could coordinate a bit? Or appoint a infrastructure delegate? bye, -christian- P.S.: It's funny that we're open about all and everything but debian-admin is still a closed list and do they don't announce very much neither :-( -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Please unblock mysql-dfsg-5.0.32-2
Hello Luk On 2007-01-19 Luk Claes wrote: > Christian Hammers wrote: > > Hello > > > > MySQL release another upstream version which fixes a lot of "can cause > > the server to crash" type bugs which of course could potentially allow > > people to cause data loss to other users and thus could be considered as > > DoS. Also there was a bug that let "mysql_upgrade" crash which would > > probably a nasty thing on sarge->etch upgrades. > > > > I had to make a -2 upload as I inadvertently deleted a necessary .dpatch > > file. The new "upstream" is now in unstable for 15 days without any > > reported regressions. Alpha builds are still missing but this seems to > > be due to a lack of redundancy in our buildd environment (*g*), the last > > couple of upstream versions all worked fine on alpha. > > Unblocked, setting waiting time to 20 days. The piuparts team found that my package should pre-depend on adduser and filed an RC-bug. In addition two new debconf translations appeared. Please extend the current exception to 5.0.32-3! thanks, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please melt quagga-0.99.5-5etch1
Hello Please unfreeze quagga-0.99.5etch1 which just adds a new Debconf translation. It was uploaded to t-p-u as unstable already contains 0.99.6. bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please unblock mysql-dfsg-5.0 5.0.32-6 <- ABI problems!
Hello
Please allow mysql-dfsg-5.0 to be updated from 5.0.32-3 to -6.
The main issue is probably the one from 5.0.32-4:
After the accidental upstream ABI change in libmysqlclient15.so we forgot
to raise the minimum required version in the .shlibs file. Now some
packages like libdbd-mysql-perl crash with strange error messages as they
receive communication packages from MySQL that are slightly longer than
they expect and thus overwrites unrelated variables :-(
I guess binNMUs have to be made here, as you did after the other upstream
ABI problem in June 2006. Sorry for this.
The early -4 changes contains some stuff like more comments in my.cnf which
are not RC of course but completely harmless on the other hand so it would be
nice if you would safe me the time to create a 5.0.32-3etch1 branch.
The changes from Monty Taylor address issues that would probably have caused
RC bugs in the future (he's mysql employee and sent patches instead of filing
a bug report :))
thanks,
-christian-
mysql-dfsg-5.0 (5.0.32-6) unstable; urgency=low
* Changed wording in Debconf templates to better fit to the graphical
interface (thanks to Frank Kuester). Closes: #411165
* Lintian suggested style changes to some other Debconf questions.
* Removed accidently stdout output from init script.
-- Christian Hammers <[EMAIL PROTECTED]> Fri, 16 Feb 2007 20:29:18 +0100
mysql-dfsg-5.0 (5.0.32-5) unstable; urgency=medium
* Backported upstream patch for a bug that crashed the server when using
certain join/group/limit combinations.
Users of the Joomla CMS seemed to be affected by this. Closes: #403721
* The debian-start script that runs on every server start now first upgrades
the system tables (if neccessary) and then check them as it sometimes did
not work the other way around (e.g. for MediaWiki). The script now uses
mysql_update instead of mysql_update_script as recommended. Closes: 409780
* Remove the Debconf generated config file in postrm.
-- Christian Hammers <[EMAIL PROTECTED]> Thu, 15 Feb 2007 04:47:04 +0100
mysql-dfsg-5.0 (5.0.32-4) unstable; urgency=high
[Christian Hammers]
* Changed minimum required version in dh_makeshlibs to 5.0.27-1 as
5.0.26 had an ABI breakage in it!
This is the cause for Perl programs crashing with the following error:
"Transactions not supported by database at /usr/lib/perl5/DBI.pm line 672"
* The old_passwords setting that is set according to a Debconf question is
now written to /etc/mysql/conf.d/old_passwords.cnf instead directly to the
conffile /etc/mysql/my.cnf which would be fobidden by policy (thanks to
Robert Bihlmeyer). Closes: #409750
* Added some more comments to the default my.cnf.
[Monty Taylor]
* Added bison to build dependencies.
* Added a "start-initial" option to the Data Node init script to support
initial node starts.
* Changed NDB Data and Management node startup seqence. Prevented both from
restarting on upgrade to address rolling upgrade issues.
* Updated build-depends to depend on automake1.9 instead of automake1.8
to match what upstream uses.
-- Christian Hammers <[EMAIL PROTECTED]> Wed, 31 Jan 2007 01:14:09 +0100
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Please unblock mysql-dfsg-5.0 5.0.32-6 <- ABI problems!
Hello On 2007-02-17 Steve Langasek wrote: > On Sat, Feb 17, 2007 at 01:05:48PM +0100, Christian Hammers wrote: > > > Please allow mysql-dfsg-5.0 to be updated from 5.0.32-3 to -6. > > > The main issue is probably the one from 5.0.32-4: > > After the accidental upstream ABI change in libmysqlclient15.so we forgot > > to raise the minimum required version in the .shlibs file. > > If this is the only reason for a freeze exception, then it's a non-issue. > There is no old version of libmysqlclient15off around anywhere in the > archive for packages to get built or installed against, so even if the > shlibs are off, that's not something we need to unblock the package for. Ah, OK. > > Now some packages like libdbd-mysql-perl crash with strange error > > messages as they receive communication packages from MySQL that are > > slightly longer than they expect and thus overwrites unrelated > > variables :-( > > This, OTOH, sounds like an incompatible ABI break that should have had a > *package* name change when it happened, not a shlibs bump. The ABI was broken and then reverted i.e. fixed shortly after. In such cases no package name changes are needed, or? > I'm sure it's far too entrenched now to be worth trying to change the > library name, but we do need info about what packages will need binNMUed. > Is it just any binaries in the archive, depending on libmysqlclient15off, > that were uploaded before 5.0.27-1 was available (7 Nov 2006)? Precisely only the ones built between 5.0.26-1 (Oct 2006) and 5.0.27-1 but to be sure and if it's automated, better upload everything built before 7 Nov 2006. > > The changes from Monty Taylor address issues that would probably have > > caused RC bugs in the future (he's mysql employee and sent patches > > instead of filing a bug report :)) > > Um, that's a vague rationale; I'm "probably" not going to make a freeze > exception for a "probably" RC bug. :) I ask him if he can give me a scenario where data loss is involved and then file a bug :) (or tell you if the patch is not that important) > I need to review this for consideration anyway, because of the my.cnf > bugfix; so I'll let you know if I think there are any problems. OK bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Please unblock mysql-dfsg-5.0 5.0.32-6 <- ABI problems!
Hi On 2007-02-18 Christian Perrier wrote: > > The version in testing has an RC bug, #409750. But an uncoordinated > > template change is enough of a reason for me to reject this update and > > require a targetted fix via t-p-u. I know Christian will have this > > covered no matter what, but the l10n update policy is intended for the > > benefit of users of all languages, not just French. :) > > Another possibility is also handling a call for translation updates > lasting for 3 days (the templates changes are very minor and can > easily be updated very quickly by translators) and then upload a new > version of the package in unstable. It seems that I have to prepare a t-p-u version with only the important bits from the changelogs of from my initial mail anyway so if it's OK for the RMs I will add the template change alongside with all translations to that, too. bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Please unblock mysql-dfsg-5.0 5.0.32-6 <- ABI problems!
Hello On 2007-02-17 Steve Langasek wrote: > > The changes from Monty Taylor address issues that would probably have > > caused RC bugs in the future (he's mysql employee and sent patches > > instead of filing a bug report :)) > > Um, that's a vague rationale; I'm "probably" not going to make a freeze > exception for a "probably" RC bug. :) After asking Monty about this he told me that at least no data corruption can appear. So that can to wait. So I would prepare the a t-p-u upload with the following changes if that's ok with you: * Backported upstream patch for a bug that crashed the server when using certain join/group/limit combinations. Users of the Joomla CMS seemed to be affected by this. Closes: #403721 * The debian-start script that runs on every server start now first upgrades the system tables (if neccessary) and then check them as it sometimes did not work the other way around (e.g. for MediaWiki). The script now uses mysql_update instead of mysql_update_script as recommended. Closes: 409780 * The old_passwords setting that is set according to a Debconf question is now written to /etc/mysql/conf.d/old_passwords.cnf instead directly to the conffile /etc/mysql/my.cnf which would be fobidden by policy (thanks to Robert Bihlmeyer). Closes: #409750 * Added bison to build dependencies. The l10n Debconf changes will be made in unstable only. Do you allow t-p-u uploads that consists of Debconf changes with *complete* translations? Then I will do that once I have them all (thanks to Christian Perriers mail 8/10 are already there). bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Please unblock mysql-dfsg-5.0 5.0.32-6 <- ABI problems!
On 2007-02-19 Steve Langasek wrote: > On Sun, Feb 18, 2007 at 03:28:37PM +0100, Christian Hammers wrote: > > It seems that I have to prepare a t-p-u version with only the > > important bits from the changelogs of from my initial mail anyway > > so if it's OK for the RMs I will add the template change alongside > > with all translations to that, too. > > No, the reason for insisting on a t-p-u upload was that the *template* > change was inappropriate because it was a regression for users of several I meant the "all" before translations literally :) At the current speed they will be completed in 1-2 days. So I wait a bit before doing my upload to t-p-u and else do it without template change, of course. Regarding the ABI break, after going mad about not being able to reproduce it, I figured that it only happens because I had libmysqlclient.so.15 from the experimental mysql-5.1 installed. So it's probably really the same bug we had in unstable last year but this time it only concerns me and mysql and no binNMUs are needed! bye, -christian- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
frozen-proposed-updates?
Hello I'd like to upload a package that fixes a bug in testing's quagga_0.99.21-4. I can't upload it via unstable as we have already 0.99.22 there and neither can I upload it to stable-proposed-updates as this is still for squeeze, right? So I have to wait a week for the release and then use s-p-u? bye, -christian- -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130426024405.63faf44d@james.intern
Re: frozen-proposed-updates?
Hello
Am Fri, 26 Apr 2013 09:47:27 +0200
schrieb Julien Cristau :
> On Fri, Apr 26, 2013 at 02:44:05 +0200, Christian Hammers wrote:
>
> > Hello
> >
> > I'd like to upload a package that fixes a bug in testing's
> > quagga_0.99.21-4. I can't upload it via unstable as we have
> > already 0.99.22 there and neither can I upload it to
> > stable-proposed-updates as this is still for squeeze, right?
> >
> > So I have to wait a week for the release and then use s-p-u?
> >
> Please describe the fixes and include the debdiff, we'll figure things
> out from there.
Here is the debdiff with changelog entry and diff:
-=(~/debian/quagga/wheezy)$ debdiff quagga_0.99.21-4.dsc
quagga_0.99.21-4+wheezy1.dsc
diff -Nru quagga-0.99.21/debian/changelog quagga-0.99.21/debian/changelog
--- quagga-0.99.21/debian/changelog 2013-01-06 15:58:00.0 +0100
+++ quagga-0.99.21/debian/changelog 2013-04-25 02:09:06.0 +0200
@@ -1,3 +1,10 @@
+quagga (0.99.21-4+wheezy1) testing-proposed-updates; urgency=medium
+
+ * Applied a patch to prevent ospfd to crash randomly when a neighbor
+goes down. Thanks to David Lamparter for pointing out. Closes: #706120
+
+ -- Christian Hammers Mon, 22 Apr 2013 23:00:01 +0200
+
quagga (0.99.21-4) unstable; urgency=medium
* Fixed regression bug that caused OSPF "distribute-list" statements to be
diff -Nru quagga-0.99.21/debian/patches/99_fix_thread_cancel_event.diff
quagga-0.99.21/debian/patches/99_fix_thread_cancel_event.diff
--- quagga-0.99.21/debian/patches/99_fix_thread_cancel_event.diff
1970-01-01 01:00:00.0 +0100
+++ quagga-0.99.21/debian/patches/99_fix_thread_cancel_event.diff
2013-04-22 22:09:13.0 +0200
@@ -0,0 +1,51 @@
+From aaa964a8f861d5cd68723adc27236548f3f05713 Mon Sep 17 00:00:00 2001
+From: Jorge Boncompte [DTI2]
+Date: Mon, 07 May 2012 15:17:31 +
+Subject: lib: fix thread_cancel_event()
+
+ ospfd was crashing some times on neighbour going down. The cause was that
+ospf_nsm_event() was accessing already freed memory in ospf_nbr_delete()
+call from ospf_nsm_event().
+
+ What happens is that since commit b5043aab (lib: fix incorrect thread
+list...) now a thread can be on the event and ready lists but
+thread_cancel_event() doesn't account for that.
+
+* thread.c: (thread_cancel_event) loop on the ready list too to cancel
+ pending events.
+
+Signed-off-by: Jorge Boncompte [DTI2]
+Signed-off-by: David Lamparter
+---
+diff --git a/lib/thread.c b/lib/thread.c
+index b36c43a..dd0413b 100644
+--- a/lib/thread.c
b/lib/thread.c
+@@ -916,6 +916,24 @@ thread_cancel_event (struct thread_master *m, void *arg)
+ thread_add_unuse (m, t);
+ }
+ }
++
++ /* thread can be on the ready list too */
++ thread = m->ready.head;
++ while (thread)
++{
++ struct thread *t;
++
++ t = thread;
++ thread = t->next;
++
++ if (t->arg == arg)
++{
++ ret++;
++ thread_list_delete (&m->ready, t);
++ t->type = THREAD_UNUSED;
++ thread_add_unuse (m, t);
++}
++}
+ return ret;
+ }
+
+--
+cgit v0.9.0.2
+curl: try 'curl --help' or 'curl --manual' for more information
diff -Nru quagga-0.99.21/debian/patches/series
quagga-0.99.21/debian/patches/series
--- quagga-0.99.21/debian/patches/series2013-01-06 15:45:54.0
+0100
+++ quagga-0.99.21/debian/patches/series2013-04-25 02:33:43.0
+0200
@@ -7,3 +7,4 @@
10_doc__Makefiles__makeinfo-force.diff
99_CVE-2012-1820_bgp_capability_orf.diff
99_distribute_list.diff
+99_fix_thread_cancel_event.diff
bye,
-christian-
signature.asc
Description: PGP signature
Quagga 0.99.21 update (was Re: frozen-proposed-updates?)
Am Sun, 28 Apr 2013 17:00:59 +0100 schrieb "Adam D. Barratt" : > On Sat, 2013-04-27 at 13:12 +0200, Christian Hammers wrote: > > +quagga (0.99.21-4+wheezy1) testing-proposed-updates; urgency=medium > > > > + > > + * Applied a patch to prevent ospfd to crash randomly when a > > neighbor > > +goes down. Thanks to David Lamparter for pointing out. Closes: > > #706120 > > That bug is filed as severity "important", whereas the text says > "completely unusable"; please could you clarify which is the case? Quagga consists of several independant daemons for different routing protocols (BGP/OSPF/RIP) i.e. the package as a whole is not completely unusable. Just of its daemons, and an important one, is unstable. > The version information also indicates that the problem applies to the > version of quagga in unstable. If that's correct, please fix it in > unstable; if not, please add an appropriate fixed version. We've 0.99.22 in unstable (and 0.99.22.1 with a fix coming soon) and 0.99.21 in testing. At this stage of the freeze, I'd rather prefer a fixed 0.99.21 than the 0.99.22 to be released. So may I upload 0.99.21-4+wheezy1 to t-p-u? bye, -christian- -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130428181802.5b029c74@james.intern
Re: Quagga 0.99.21 update (was Re: frozen-proposed-updates?)
Am Sun, 28 Apr 2013 17:24:26 +0100 schrieb "Adam D. Barratt" : > On Sun, 2013-04-28 at 18:18 +0200, Christian Hammers wrote: > > Am Sun, 28 Apr 2013 17:00:59 +0100 > > schrieb "Adam D. Barratt" : > > > The version information also indicates that the problem applies > > > to the version of quagga in unstable. If that's correct, please > > > fix it in unstable; if not, please add an appropriate fixed > > > version. > > > > We've 0.99.22 in unstable (and 0.99.22.1 with a fix coming soon) and > > 0.99.21 in testing. At this stage of the freeze, I'd rather prefer a > > fixed 0.99.21 than the 0.99.22 to be released. > > I wasn't suggesting that we ship 0.99.22 in wheezy, I was asking > whether the package in unstable contains the fix. Looking at the > code, the answer is "yes, it does", which makes your answer that it's > "coming soon" slightly confusing. :-( Sorry, came to my mind as well and I've already send a mail to control@bugs to mark the 0.99.22 version as fixed. > Please go ahead with the upload. Please also fix the version > information for the bug to indicate that it doesn't apply to the > package in unstable. OK, thanks! bye, -christian- -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130428183035.2e8b4d61@james.intern
