Bug#871062: collectd: FTBFS: client.c:621:23: error: '%s' directive output may be truncated writing up to 1023 bytes into a region of size 1010 [-Werror=format-truncation=]

[Marc Fournier]

Hello Michael,

On Mon, Aug 21, 2017 at 06:11:02PM +0200, Michael Stapelberg wrote:
> Hi,
> 
> Steve Langasek  writes:
> > The attached one-liner patch corrects this build failure by simply ignoring
> > the (IMHO uninteresting) new gcc-7 warning.  I think this is a reasonable
> > way to handle this until it gets fixed upstream.
> 
> Marc, Sebastian, does either of you have time to upload this patch, or
> would you prefer an NMU?
> 
> This is somewhat time-critical, because this RC bug will cause
> freeradius to be removed from testing.

Thanks for the heads-up ! Sorry for having let these FTBFS issues pile up :-(

I'm getting back on track and will take care of uploading 5.7.2 + the misc
fixes needed so the package builds again.

For this specific issue, I feel this fix would be better suited:
https://github.com/collectd/collectd/issues/2200

Cheers,
Marc

Bug#852924: collectd: FTBFS: ld: cannot find -lssl

[Marc Fournier]

On Sat, Jan 28, 2017 at 09:30:20AM +0100, Lucas Nussbaum wrote:
> During a rebuild of all packages in sid, your package failed to build on
> amd64.
> 
> Relevant part (hopefully):
> > /bin/bash ../libtool  --tag=CC   --mode=link x86_64-linux-gnu-gcc -Wall 
> > -Werror -g -O2 -fdebug-prefix-map=/<>=. 
> > -fstack-protector-strong -Wformat -Werror=format-security -Wall 
> > -Wno-error=deprecated-declarations -module -avoid-version 
> > -export-symbols-regex '\' -Wl,-z,relro -o notify_email.la 
> > -rpath /usr/lib/collectd notify_email.lo -lesmtp -lssl -lcrypto 
> > libtool: link: /usr/bin/x86_64-linux-gnu-nm -B  .libs/notify_email.o   | 
> > sed -n -e 's/^.*[   ]\([ABCDGIRSTW][ABCDGIRSTW]*\)[ ][  
> > ]*\([_A-Za-z][_A-Za-z0-9]*\)$/\1 \2 \2/p' | sed '/ __gnu_lto/d' | /bin/sed 
> > 's/.* //' | sort | uniq > .libs/notify_email.exp
> > libtool: link: /bin/grep -E -e "\" 
> > ".libs/notify_email.exp" > ".libs/notify_email.expT"
> > libtool: link: mv -f ".libs/notify_email.expT" ".libs/notify_email.exp"
> > libtool: link: echo "{ global:" > .libs/notify_email.ver
> > libtool: link:  cat .libs/notify_email.exp | sed -e "s/\(.*\)/\1;/" >> 
> > .libs/notify_email.ver
> > libtool: link:  echo "local: *; };" >> .libs/notify_email.ver
> > libtool: link:  x86_64-linux-gnu-gcc -shared  -fPIC -DPIC  
> > .libs/notify_email.o   -lesmtp -lssl -lcrypto  -g -O2 
> > -fstack-protector-strong -Wl,-z -Wl,relro   -Wl,-soname -Wl,notify_email.so 
> > -Wl,-version-script -Wl,.libs/notify_email.ver -o .libs/notify_email.so
> > /usr/bin/ld: cannot find -lssl
> > /usr/bin/ld: cannot find -lcrypto
> > collect2: error: ld returned 1 exit status
> 
> The full build log is available from:
>http://aws-logs.debian.net/2017/01/28/collectd_5.7.1-1_unstable.log

Nice catch, thanks !

One thing that puzzles me, is why didn't this pop up earlier. collectd does
*not* build-depend on libssl-dev. But according to the latest build log[1]
libssl1.0-dev gets installed nevertheless (making this mistake in
collectd's Makefile work by accident). This is not the case on the
AWS-based builder.

There seems to be some discrepancy between the official builders and the
AWS one (the latter being correct/stricter).

Cheers,
Marc

[1] 
https://buildd.debian.org/status/fetch.php?pkg=collectd=amd64=5.7.1-1=1485207190=0

Bug#852658: collectd-core: Won't start without non-required packages libsensors4 liboping0

[Marc Fournier]

On Thu, Jan 26, 2017 at 08:05:06AM +, Jan Huijsmans wrote:
> Package: collectd-core
> Version: 5.7.0-3
> Severity: grave
> Justification: renders package unusable
> 
> Dear Maintainer,
> 
> Installation of collectd fails on the start of the package.
> The package misses files from libsensors4 liboping0, but these
> packages are only recommended or even suggested.

The configuration file shipped with the collectd package only loads these
plugins by default:

marc@lonquimay:~/src/pkg-collectd/debian$ grep ^LoadPlugin collectd.conf
LoadPlugin syslog
LoadPlugin battery
LoadPlugin cpu
LoadPlugin df
LoadPlugin disk
LoadPlugin entropy
LoadPlugin interface
LoadPlugin irq
LoadPlugin load
LoadPlugin memory
LoadPlugin processes
LoadPlugin rrdtool
LoadPlugin swap
LoadPlugin users

None of them depend on libsensors4 or liboping0 (the sensors and ping
plugins do, but they aren't enabled by default).

So my guess is that this system previously had a non-default configuration
(maybe some config snippets in /etc/collectd/collectd.conf.d/ ?) in place,
and installing/upgrading collectd-core made the missing runtime
dependencies strike out.

Are you able to confirm ?

NB: I agree such a failure is undesirable. The collectd plugin loading
mechanism could maybe be changed to not abort startup in this case (just
skip loading the plugin and emit an error message).

> Manual installation of libsensors4 liboping0 solved the issue.
> 
> When a package won't complete it's installation without a
> package, it should be a requirement, not anything less.
> 
> Probably an issue with the debian packages as well.

Are you aware of /usr/share/doc/collectd-core/README.Debian.plugins.gz ?

Cheers,
Marc

Bug#851163: Update

[Marc Fournier]

Thanks for reporting this issue !

According to your config file, you don't have the WriteQueueLimit{Low,High}
global options enabled. In the case where this issue would be caused by one
of the write plugins not being able to keep up with the data collection
rate, collectd would indeed buffer up values forever. So one lead would be
to set both of these options to the same value (for ex 5000) and observe
evolution of memory consumption over the same time range as without.

As you suggested, selectively disabling plugins could help narrowing down
which one is causing trouble *assuming the problem is in a plugin, not in
the core engine*. While doing this, try starting with the write plugins
(network, graphite and rrdtool) and consider setting a much lower global
Interval (such as 0.1 instead of the default 10). This should make the
memory allocation functions run more often, and the leak to happen faster.

The fact that this issue appears after upgrading from 5.6 is clearly
something that warrants investigation, thanks for your help !

Marc

Bug#840349: intel-cmt-cat: please enable building of shared library

[Marc Fournier]

Package: intel-cmt-cat
Version: 0.1.5-1
Severity: important

Dear Maintainer,

Current version of intel-cmt-cat doesn't provide the shared library
(/usr/lib/libpqos.so) which is required to link against it.

For example, running "gcc -lpqos ..." fails with:
  relocation R_X86_64_PC32 against undefined symbol `stderr@@GLIBC_2.2.5' can 
not be used when making a shared object; recompile with -fPIC

When rebuilding the intel-cmt-cat package after setting:
  export SHARED=y
in debian/rules, linking against libpqos works just fine, and the
resulting binary runs as expected.

Thanks !

-- System Information:
Debian Release: stretch/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'stable'), (500, 'oldstable'), (300, 
'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.7.0-rc7-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages intel-cmt-cat depends on:
ii  libc6  2.24-3

intel-cmt-cat recommends no packages.

intel-cmt-cat suggests no packages.

-- no debconf information

Bug#839771: collectd-core: collectd graphite plugin is missing libyajl2 dependency

[Marc Fournier]

On Tue, Oct 04, 2016 at 08:02:43PM +0100, Shish wrote:
> 
> I ran "apt upgrade", collectd failed to restart, complaining that it
> didn't understand the graphite section of the config file, because the
> write_graphite plugin wasn't loaded. It recommended that I tried running
> lld on the .so file, which I did, and it said that libyajl.so.2 was not
> found. I ran apt install libyajl2, then it worked. I expect libyajl2
> should be marked as recommends rather than suggests, since not having it
> will break currently working installations?

write_graphite should not be linked against libyajl ! Congrats, you found a
genuine bug :-)

I just submitted a patch upstream to fix this:
https://github.com/collectd/collectd/pull/1976

In the meantime, installing libyajl2 should be harmless, and is probably
the only possible workaround.

Thanks for reporting this problem !

Marc

Bug#803163: collectd: Please run autoreconf

[Marc Fournier]

Hello,

On Mon, May 30, 2016 at 11:06:32AM +0200, Marc Fournier wrote:
> On Sat, May 28, 2016 at 12:43:21PM +0200, Guillem Jover wrote:
> > On Tue, 2015-10-27 at 16:27:24 +0100, Benjamin Drung wrote:
> > 
> > > To make it easier to patch Makefile.am after applying patches [1],
> > > please run autoreconf when building collectd.
> > 
> > It seems part of this got fixed when fixing #823012, but it still
> > needs some minor changes, otherwise this fails when doing a second
> > build in a row. Attached the needed change to solves this.
> 
> I must admit I was initially a bit reluctant to run autoreconf: I ran into
> numerous problems when doing this while setting up the CI pipeline at
> https://ci.collectd.org/. Admittedly mostly when combining ancient autoconf
> versions from legacy distros with the latest collectd code.

So unfortunately your patch introduces one of these issues: dh_autoreconf
is now run only if "configure" is not found, but as we're patching
"configure.ac", running autoreconf is mandatory.

What happens then is that it triggers automake by itself, which fails if
there's a version mismatch with the version used to generate the orig
tarball (collectd 5.5.1 was released using 1.14, sid currently comes with
1.15).

I have only little time for this this week, so I'm going to upload 5.5.1-4
without your patch (to fix a lintian error I introduced in -3) in a couple
of minutes.

Let me know if you come up with a patch solving this problem, else I'll
take a look at this again in a few days.

Thanks !

Marc



signature.asc
Description: PGP signature

Bug#803163: collectd: Please run autoreconf

[Marc Fournier]

Thanks Guillem & Benjamin for your input on this !

On Sat, May 28, 2016 at 12:43:21PM +0200, Guillem Jover wrote:
> On Tue, 2015-10-27 at 16:27:24 +0100, Benjamin Drung wrote:
> 
> > To make it easier to patch Makefile.am after applying patches [1],
> > please run autoreconf when building collectd.
> 
> It seems part of this got fixed when fixing #823012, but it still
> needs some minor changes, otherwise this fails when doing a second
> build in a row. Attached the needed change to solves this.

I must admit I was initially a bit reluctant to run autoreconf: I ran into
numerous problems when doing this while setting up the CI pipeline at
https://ci.collectd.org/. Admittedly mostly when combining ancient autoconf
versions from legacy distros with the latest collectd code.

But as there doesn't seem to be any reason to hold this back for the
combination of recent autoconf & collectd versions, I suggest running
autoreconf for the stable/testing/unstable lines, but disabling it for the
backports (+ other non-debian lines such as the nightly builds).

Unless Tokkee has any comments on this, I'll prepare a 5.5.1-4 release
shortly including Guillem's patch.

Thanks !

Marc



signature.asc
Description: PGP signature

Bug#823034: transition: rrdtool

[Marc Fournier]

Hello,

On Sat, May 21, 2016 at 05:05:42PM +, Jean-Michel Vourgère wrote:
> Hi
> 
> Last package using librrd4 is collectd.

Thanks for the heads up !

I just uploaded collectd 5.5.1-3, which includes the patch you submitted in
#823012. Thanks so much for taking care of all this Jean-Michel !

Cheers,
Marc



signature.asc
Description: PGP signature

Bug#822478: libopenipmi-dev: 2.0.21 re-introduces pkg-config issue reported in #474087

[Marc Fournier]

Package: libopenipmi-dev
Version: 2.0.21-1.2
Severity: normal

Dear Maintainer,

I have a package failing to build, because its configure script relies
on pkg-config to extract build flags from libopenipmi:

$ pkg-config --libs OpenIPMIpthread
Package pthread was not found in the pkg-config search path.
Perhaps you should add the directory containing `pthread.pc'
to the PKG_CONFIG_PATH environment variable
Package 'pthread', required by 'OpenIPMIpthread', not found

I won't describe the issue more in details, as it's exactly the same as
#474087.

I'm not sure exactly which of the 2.0.21 uploads introduced this
regression, but 2.0.16-1.4 shipped with jessie definetely doesn't suffer
from this problem.

Thanks for taking care of this package in Debian !

Marc

-- System Information:
Debian Release: stretch/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (500, 'oldstable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.4.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages libopenipmi-dev depends on:
ii  libgdbm-dev   1.8.3-13.1
ii  libopenipmi0  2.0.21-1.2

libopenipmi-dev recommends no packages.

libopenipmi-dev suggests no packages.

-- no debconf information

Bug#811580: FTBFS with GCC 6: statement indented as if it were guarded by

[Marc Fournier]

Thanks for reporting this issue !

It is now fixed upstream: https://github.com/collectd/collectd/commit/376667a

... which means it will be part of 5.5.1.

FWIW, I just built collectd with gcc-6 and make -k, and this was the only
issue reported.

Bug#806196: collectd: Missing several plugin stanzas in default conffile

[Marc Fournier]

Guillem,

Part of these are plugins which don't build on debian (typically missing
build-dependencies). So they have been removed from the package's
collectd.conf on purpose.

But you're right about onewire and write_redis. To which at least conntrack
should be added.

Thanks for pointing this out !

Marc

Bug#797687: [collectd] 'Spammy debug messages' at normal verbosity

[Marc Fournier]

On Tue, Sep 01, 2015 at 08:42:13AM -0700, Antonio Russo wrote:
> 
> This bug[1] results in many lines in syslog reading
> 
>"0 Success: 1 value has been dispatched."
> 
> and has already been resolved upstream:
> 
> https://github.com/collectd/collectd/pull/714/commits
> 
> [1] https://github.com/collectd/collectd/issues/664

NB: this issue *hasn't* been fixed upstream. The suggested patch wasn't
merged because it would have broken collectd's plain text protocol.

Just my 2¢...

Bug#785216: pgbadger: New upstream releases available

[Marc Fournier]

Hello,

I wasn't able to find pgbadger 7.0 in the archive. Am I missing something ?

FWIW, I rebuilt the package myself using the 3.3-2 source package as a
starting point and simply updating the sources code to version 7.0. It
worked out of the box and runs fine on a jessie box.

Cheers,
Marc


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#588153: collectd disk plugin needs indirection ability

[Marc Fournier]

I believe this feature has been implemented in
https://github.com/collectd/collectd/pull/537 (which will be part of
upcoming 5.5 release).

Cheers,
Marc


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#780365: rsyslog: please build omkafka module

[Marc Fournier]

On Thu, Mar 12, 2015 at 10:06:29PM +0100, Michael Biebl wrote:
 Am 12.03.2015 um 18:12 schrieb Marc Fournier:
  Package: rsyslog
  Version: 8.8.0-3
  Severity: wishlist
  Tags: patch
  
  Dear Maintainer,
  
  Starting with version 8.7.0, rsyslog comes with a new module allowing to
  publish log messages to Apache Kafka. The support lib needed for this
  (librkafka-dev) is part of Debian since Jessie.
  
  The attached patch adds the bits needed to build this module as a
  separate package (based on how support for omelasticsearch was done).
  
  Thanks taking this request in account, and thanks for maintaining this
  package in Debian !
 
 Thanks for your contribution, Marc.
 
 Patch looks good. Since I don't have a Apache Kafka setup, I don't know,
 what environment I need to have installed and what configuration steps
 need to be done to make the plugin work.
 
 Could you write a README.Debian, similar to
 debian/rsyslog-mongodb.README.Debian and update the patch accordingly?
 
 I think this would be helpful for people installing this plugin.

Here you go!

As Kafka isn't part of Debian and as a production service can be
non-trivial to setup, I'm pointing people to the official documentation and
staying voluntarily vague about this point. I hope this is OK.

Cheers,
Marc
From a849a7c5063f4f1840404cd7038a5732446c8d5e Mon Sep 17 00:00:00 2001
From: Marc Fournier marc.fourn...@camptocamp.com
Date: Mon, 16 Mar 2015 23:11:45 +0100
Subject: [PATCH 2/2] Add README.Debian for omkafka module

---
 debian/rsyslog-kafka.README.Debian | 26 ++
 1 file changed, 26 insertions(+)
 create mode 100644 debian/rsyslog-kafka.README.Debian

diff --git a/debian/rsyslog-kafka.README.Debian b/debian/rsyslog-kafka.README.Debian
new file mode 100644
index 000..0470e37
--- /dev/null
+++ b/debian/rsyslog-kafka.README.Debian
@@ -0,0 +1,26 @@
+How to use rsyslog and Apache Kafka
+===
+
+Starting with version 8.7.0, rsyslog comes with an output module named
+“omkafka”, allowing to publish log messages to an Apache Kafka message broker.
+
+Apache Kafka isn't part of the Debian archive, and documenting how to set it up
+is out of the scope of this document. Fortunately, setting up a single-node
+Kafka service is quite easy. The following instructions should be enough to get
+started: https://kafka.apache.org/documentation.html#quickstart
+
+At this point, you'll have to alter your rsyslog configuration, typically by
+adding the following lines to rsyslog.conf (or to a seperate files such as
+/etc/rsyslog.d/kafka.conf, which will be included by the main config file):
+
+ module(load=omkafka)
+ *.* action(type=omkafka topic=test)
+
+Don't forget to restart the rsyslogd daemon, and make sure the “topic”
+parameter matches one of the topics in Kafka. You should then be able to start
+consuming the logs stored in Kafka with the consumer of your choice. A trivial
+consumer comes with the Kafka server installation.
+
+The full configuration details of this module, are found in the following file
+(which is part of the rsyslog-doc package):
+/usr/share/doc/rsyslog-doc/html/configuration/modules/omkafka.html
-- 
2.1.4

Bug#780365: rsyslog: please build omkafka module

[Marc Fournier]

Package: rsyslog
Version: 8.8.0-3
Severity: wishlist
Tags: patch

Dear Maintainer,

Starting with version 8.7.0, rsyslog comes with a new module allowing to
publish log messages to Apache Kafka. The support lib needed for this
(librkafka-dev) is part of Debian since Jessie.

The attached patch adds the bits needed to build this module as a
separate package (based on how support for omelasticsearch was done).

Thanks taking this request in account, and thanks for maintaining this
package in Debian !

Marc
From a881a68631c98d3b73d0171a6d0e857d16ee3592 Mon Sep 17 00:00:00 2001
From: Marc Fournier marc.fourn...@camptocamp.com
Date: Thu, 12 Mar 2015 18:01:47 +0100
Subject: [PATCH] Build omkafka module

---
 debian/control   | 12 
 debian/rsyslog-kafka.install |  1 +
 debian/rules |  1 +
 3 files changed, 14 insertions(+)
 create mode 100644 debian/rsyslog-kafka.install

diff --git a/debian/control b/debian/control
index 2fcb743..409fbac 100644
--- a/debian/control
+++ b/debian/control
@@ -11,6 +11,7 @@ Build-Depends: debhelper (= 8),
libpq-dev,
libmongo-client-dev (= 0.1.4),
libcurl4-gnutls-dev,
+   librdkafka-dev,
libkrb5-dev,
libgnutls28-dev,
librelp-dev (= 1.2.5),
@@ -110,6 +111,17 @@ Description: Elasticsearch output plugin for rsyslog
  distributed, multitenant-capable full-text search engine with a RESTful web
  interface and schema-free JSON documents.
 
+Package: rsyslog-kafka
+Architecture: any
+Priority: extra
+Depends: ${shlibs:Depends},
+ ${misc:Depends},
+ rsyslog (= ${binary:Version}),
+Description: Kafka output plugin for rsyslog
+ This plugin allows rsyslog to write syslog messages to Apache Kafka, a
+ distributed and scalable message broker offering strong durability and
+ fault-tolerance guarantees.
+
 Package: rsyslog-gssapi
 Architecture: any
 Priority: extra
diff --git a/debian/rsyslog-kafka.install b/debian/rsyslog-kafka.install
new file mode 100644
index 000..179aaf0
--- /dev/null
+++ b/debian/rsyslog-kafka.install
@@ -0,0 +1 @@
+usr/lib/rsyslog/omkafka.so
diff --git a/debian/rules b/debian/rules
index 0cd28c6..cc36bc9 100755
--- a/debian/rules
+++ b/debian/rules
@@ -22,6 +22,7 @@ override_dh_auto_configure:
 		--enable-pgsql \
 		--enable-ommongodb \
 		--enable-elasticsearch \
+		--enable-omkafka \
 		--enable-mail \
 		--enable-imfile \
 		--enable-impstats \
-- 
2.1.4

Bug#770680: collectd: please backport fix for segfault in lvm.c

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

A fix for #747093 is available upstream. I believe it should be included
in the package shipped with jessie, as it prevents collectd from
crashing in a pretty common situation.

This patch will be part of upcoming bugfix release 5.4.2 and can be
retrieved at this address:
https://github.com/collectd/collectd/commit/25d7de930baa8a2

Thanks !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#770681: collectd fails to handle acknowledgements from riemann

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

Collectd doesn't handle the communication with a remote riemann server
properly, which eventually leads the riemann server to stall. A patch
for this issue will be part of upcoming bugfix release 5.4.2 and can be
retrieved at this address:
https://github.com/collectd/collectd/commit/e5d3760160c8f

I believe it should be included in the package shipped with jessie, as
it prevents a lockup and data loss when using collectd together with
riemann with the default settings.

Thanks !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#770683: collectd: please backport fix for threading issue in curl* plugins

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

Collectd incorrectly initialises libgcrypt in the plugins using libcurl,
leading to a segfault due to concurrent threads accessing shared
resources. A patch for this issue will be part of upcoming bugfix
release 5.4.2 and can be retrieved at this address:
https://github.com/collectd/collectd/commit/66b400ab01b

I believe it should be included in the package shipped with jessie, as
it prevents collectd from crashing when several libgcrypt-backed plugins
are used together.

Also, this issue was raised but only partly fixed in #735173. The patch
mentioned above fixes it completely.

Thanks !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#770685: collectd: please backport fix for segfault in configfile.c

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

A fix for #750440 is available upstream. I believe it should be included
in the package shipped with jessie, as it prevents collectd from
crashing in a pretty common situation.

This patch will be part of upcoming bugfix release 5.4.2 and can be
retrieved at this address:
https://github.com/collectd/collectd/commit/6207fce91a09

Thanks !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#770688: collectd leak memory when using snmp plugin

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

collectd leaks memory when monitoring more than one SNMP device. This
quickly leads to resource exhaustion on the machine running collectd.
Two patches for this issue will be part of upcoming bugfix release
5.4.2. I've merged them together and I'm attaching the resulting patch
here for your convenience.

I believe this fix should be included in the package shipped with
jessie, as it prevents collectd from saturating the host's memory in a
pretty common situation.

Thanks !
diff --git a/src/snmp.c b/src/snmp.c
index ad81c89..7d340d1 100644
--- a/src/snmp.c
+++ b/src/snmp.c
@@ -1316,6 +1316,8 @@ static int csnmp_read_table (host_definition_t *host, data_definition_t *data)
 snmp_free_pdu (res);
   res = NULL;
 
+  /* snmp_synch_response already freed our PDU */
+  req = NULL;
   sfree (errstr);
   csnmp_host_close_session (host);
 
@@ -1437,6 +1439,10 @@ static int csnmp_read_table (host_definition_t *host, data_definition_t *data)
 snmp_free_pdu (res);
   res = NULL;
 
+  if (req != NULL)
+snmp_free_pdu (req);
+  req = NULL;
+
   if (status == 0)
 csnmp_dispatch_table (host, data, instance_list_head, value_list_head);
 

Bug#770690: collectd: please backport fix for threading issue in java plugin

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

Collectd incorrectly detaches threads in the java plugin, leading to a
plugin lockup. A patch for this issue will be part of upcoming bugfix
release 5.4.2 and can be retrieved at this address:
https://github.com/collectd/collectd/commit/513a5cafb5c1

I believe it should be included in the package shipped with jessie, as
it prevents collectd from locking up.

Thanks !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#770693: collectd: please backport fixes for notifications types

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

Collectd doesn't use the proper type for timestamps when sending
notifications, leading to erroneous date/times getting emitted. Several
patches for this issue will be part of upcoming bugfix release 5.4.2.
I've merged them together and I'm attaching the resulting patch here for
your convenience.

I believe this fix should be included in the package shipped with
jessie, as leads to collectd emitting bogus data with the default
settings.

Thanks !
diff --git a/src/exec.c b/src/exec.c
index cfd82a3..b9a7365 100644
--- a/src/exec.c
+++ b/src/exec.c
@@ -744,8 +744,8 @@ static void *exec_notification_one (void *arg) /* {{{ */
 
   fprintf (fh,
   Severity: %s\n
-  Time: %.3f\n,
-  severity, CDTIME_T_TO_DOUBLE (n-time));
+  Time: %u\n,
+  severity, (unsigned int)CDTIME_T_TO_TIME_T(n-time));
 
   /* Print the optional fields */
   if (strlen (n-host)  0)
diff --git a/src/pyvalues.c b/src/pyvalues.c
index 307af17..4f5c4ce 100644
--- a/src/pyvalues.c
+++ b/src/pyvalues.c
@@ -767,7 +771,7 @@ static void Values_dealloc(PyObject *self) {
 }
 
 static PyMemberDef Values_members[] = {
-	{interval, T_INT, offsetof(Values, interval), 0, interval_doc},
+	{interval, T_DOUBLE, offsetof(Values, interval), 0, interval_doc},
 	{values, T_OBJECT_EX, offsetof(Values, values), 0, values_doc},
 	{meta, T_OBJECT_EX, offsetof(Values, meta), 0, meta_doc},
 	{NULL}
diff --git a/src/threshold.c b/src/threshold.c
index 7df4d61..887dbca 100644
--- a/src/threshold.c
+++ b/src/threshold.c
@@ -942,6 +942,7 @@ static int ut_missing (const value_list_t *vl,
   cdtime_t missing_time;
   char identifier[6 * DATA_MAX_NAME_LEN];
   notification_t n;
+  cdtime_t now;
 
   if (threshold_tree == NULL)
 return (0);
@@ -951,13 +952,15 @@ static int ut_missing (const value_list_t *vl,
   if ((th == NULL) || ((th-flags  UT_FLAG_INTERESTING) == 0))
 return (0);
 
-  missing_time = cdtime () - vl-time;
+  now = cdtime ();
+  missing_time = now - vl-time;
   FORMAT_VL (identifier, sizeof (identifier), vl);
 
   NOTIFICATION_INIT_VL (n, vl);
   ssnprintf (n.message, sizeof (n.message),
   %s has not been updated for %.3f seconds.,
   identifier, CDTIME_T_TO_DOUBLE (missing_time));
+  n.time = now;
 
   plugin_dispatch_notification (n);
 
diff --git a/src/utils_cmd_putnotif.c b/src/utils_cmd_putnotif.c
index 5a9faff..d3cf383 100644
--- a/src/utils_cmd_putnotif.c
+++ b/src/utils_cmd_putnotif.c
@@ -49,13 +49,18 @@ static int set_option_severity (notification_t *n, const char *value)
 
 static int set_option_time (notification_t *n, const char *value)
 {
-  time_t tmp;
-  
-  tmp = (time_t) atoi (value);
-  if (tmp = 0)
+  char *endptr = NULL;
+  double tmp;
+
+  errno = 0;
+  tmp = strtod (value, endptr);
+  if ((errno != 0) /* Overflow */
+  || (endptr == value) /* Invalid string */
+  || (endptr == NULL)  /* This should not happen */
+  || (*endptr != 0))   /* Trailing chars */
 return (-1);
 
-  n-time = tmp;
+  n-time = DOUBLE_TO_CDTIME_T (tmp);
 
   return (0);
 } /* int set_option_time */

Bug#770694: collectd: please backport fix for muted loglevel

[Marc Fournier]

Package: collectd
Version: 5.4.1-5
Severity: important
Tags: patch

Dear Maintainer,

A fix for #687067 is available upstream. I believe it should be included
in the package shipped with jessie, as it prevents collectd from not
emiting any logs at all, including important notifications about
monitoring data which went out of bounds.

The fix consists of two patches, which I've merged together and I'm
attaching the resulting patch here for your convenience.

Thanks !
diff --git a/src/logfile.c b/src/logfile.c
index 0f20f3c..63448cb 100644
--- a/src/logfile.c
+++ b/src/logfile.c
@@ -54,7 +54,11 @@ static int logfile_config (const char *key, const char *value)
 {
 	if (0 == strcasecmp (key, LogLevel)) {
 		log_level = parse_log_severity(value);
-		if (log_level == -1) return 1; /* to keep previous behaviour */
+		if (log_level  0) {
+			log_level = LOG_INFO;
+			ERROR (logfile: invalid loglevel [%s] defaulting to 'info', value);
+			return (1);
+		}
 	}
 	else if (0 == strcasecmp (key, File)) {
 		sfree (log_file);
diff --git a/src/syslog.c b/src/syslog.c
index 4f5d0c4..834ba79 100644
--- a/src/syslog.c
+++ b/src/syslog.c
@@ -48,7 +48,11 @@ static int sl_config (const char *key, const char *value)
 	{
 		log_level = parse_log_severity (value);
 		if (log_level  0)
+		{
+			log_level = LOG_INFO;
+			ERROR (syslog: invalid loglevel [%s] defaulting to 'info', value);
 			return (1);
+		}
 	}
 	else if (strcasecmp (key, NotifyLevel) == 0)
 	{

Bug#750440: collectd: segfault on Include *.conf after Include *.conf

[Marc Fournier]

A patch for this issue was provided by Wilfried Goesgens (thanks to him)
and will be part of the next collectd-5.4 bugfix release. Meanwhile, you
can fetch it here: https://github.com/dothebart/collectd/commit/e09d935fc

Thanks !
Marc


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#734788: collectd: Too many dependencies

[Marc Fournier]

On Thu, Jan 09, 2014 at 11:59:47PM +0400, Andrey Perminov wrote:
 
 Please, break collectd-core package into smaller size packages on per
 plugin basis. Current collectd-core have far too many dependcies to
 install, most of each aren't needed for most users. Let users install
 only selected plugins like other big packages do.

Did you read /usr/share/doc/collectd-core/README.Debian.plugins ?

I personally disagree with this suggestion, as they already aren't hard
dependencies, merely suggestions/recommendations.

If you need to setup stripped-down systems, it's in any case a good habit
to disable installation of suggestions/recommendations, regardless of
whether collectd is installed or not:

APT::Install-Recommends 0;
APT::Install-Suggests 0;

Just my 2c...


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#719691: RFP: dotdee -- convert a flat file to a file concatenated from a .d-style directory

[Marc Fournier]

Package: wnpp
Severity: wishlist

* Package name: dotdee
  Version : 1.9
  Upstream Author : Dustin Kirkland kirkl...@ubuntu.com
* URL : http://launchpad.net/dotdee
* License : GPL
  Programming Lang: shell-script
  Description : convert a flat file to a file concatenated from a .d-style 
directory

dotdee is a handy utility for converting a single flat text file to a
symlink to a file constructed by concatenating a set of files in a
.d-style directory.

As it's already packaged and maintained in ubuntu, it's probably not
much work to import it into debian.

Thanks !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#711174: base-files: changes in 7.1wheezy1 break lsb-release on stable

[Marc Fournier]

Package: base-files
Version: 7.1wheezy1
Severity: important

Dear Maintainer,

I'm not sure if this bug belongs to lsb or base-files. But here are the
facts:

Upgrading from 7.1 to 7.1wheezy1 (from wheezy-proposed-updates) breaks
the codename returned by lsb_release. n/a is returned instead of
wheezy.

This is a nuisance because everything depending on lsb to identify the
system release by codename instead of version number is also broken.

Downgrading back to 7.1 solves the problem.

Below, the steps I followed to hit this bug:

# lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description:Debian GNU/Linux 7.0 (wheezy)
Release:7.0
Codename:   wheezy

# apt-cache policy base-files
base-files:
  Installed: 7.1
  Candidate: 7.1wheezy1
  Version table:
 7.1wheezy1 0
990 http://mirror.switch.ch/ftp/mirror/debian/ 
wheezy-proposed-updates/main amd64 Packages
 *** 7.1 0
990 http://mirror.switch.ch/ftp/mirror/debian/ wheezy/main amd64 
Packages
100 /var/lib/dpkg/status
 6.0squeeze7 0
 50 http://mirror.switch.ch/ftp/mirror/debian/ squeeze/main amd64 
Packages

# apt-get install base-files
[...]
The following packages will be upgraded:
  base-files
1 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
Need to get 0 B/78.6 kB of archives.
After this operation, 0 B of additional disk space will be used.
(Reading database ... 61598 files and directories currently installed.)
Preparing to replace base-files 7.1 (using .../base-files_7.1wheezy1_amd64.deb) 
...
Unpacking replacement base-files ...
Processing triggers for man-db ...
Processing triggers for install-info ...
Setting up base-files (7.1wheezy1) ...
Installing new version of config file /etc/debian_version ...
Installing new version of config file /etc/issue ...
Installing new version of config file /etc/issue.net ...
Installing new version of config file /etc/os-release ...

# dpkg -l | grep base-files
ii  base-files 7.1wheezy1   amd64   
 Debian base system miscellaneous files

# lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description:Debian GNU/Linux 7.1 (n/a)
Release:7.1
Codename:   n/a


Thanks in advance !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#704996: pgbadger has no dependencies

[Marc Fournier]

Package: pgbadger
Version: 1.0-1
Severity: normal
Tags: patch

Hello,

Thanks for making pgbadger available in debian !

I was surprised the package has no dependency field. I guess it should
at least depend on perl ?

$ apt-cache show pgbadger | grep -c Depen
0

Also, the doc mentions pgbadger can parse CSV logs is Text::CSV_XS is
available. So I guess this module should be at least suggested to users?

The attached patch fixes these 2 details.

Cheers,
Marc

-- System Information:
Debian Release: 7.0
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-4-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

-- no debconf information
diff -r -u pgbadger-1.0.original/debian/control pgbadger-1.0/debian/control
--- pgbadger-1.0.original/debian/control	2013-04-08 17:02:33.0 +0200
+++ pgbadger-1.0/debian/control	2013-04-08 17:16:03.892450383 +0200
@@ -8,7 +8,8 @@
 
 Package: pgbadger
 Architecture: all
-Depends: ${misc:Depends}
+Depends: ${perl:Depends}
+Suggests: libtext-csv-xs-perl
 Description: Fast PostgreSQL log analysis report
  pgBadger is a PostgreSQL log analyzer build for speed with fully detailed
  reports from your PostgreSQL log file. It's a single and small script written

Bug#698214: ruby-diff-lcs: diff-lcs isn't listed by gem list

[Marc Fournier]

Package: ruby-diff-lcs
Version: 1.1.3-1
Severity: normal

Dear Maintainer,

When running gem list, diff-lcs isn't listed, which causes a
Gem::LoadError when running stuff which requires 'diff/lcs'. Example:

$ rake spec
/usr/bin/ruby1.9.1 -S rspec spec/classes/collectd_package_spec.rb
/usr/lib/ruby/1.9.1/rubygems/dependency.rb:247:in `to_specs': Could not find 
diff-lcs (~ 1.1.3) amongst [...long list of gems...] (Gem::LoadError)

Manually creating
/usr/share/rubygems-integration/1.9.1/specifications/diff-lcs-1.1.3.gemspec
and
/usr/share/rubygems-integration/1.8/specifications/diff-lcs-1.1.3.gemspec
solves the problem.

$ apt-cache policy ruby-diff-lcs
ruby-diff-lcs:
  Installed: 1.1.3-1
  Candidate: 1.1.3-1
  Version table:
 *** 1.1.3-1 0
500 http://http.debian.net/debian/ testing/main amd64 Packages
500 http://http.debian.net/debian/ unstable/main amd64 Packages
100 /var/lib/dpkg/status

$ dpkg -L ruby-diff-lcs | grep -c rubygems-integration
0

Curiously, when I apt-get source the package and rebuild it, the
resulting package *does* contain the 2 gemspec files. So the problem
seems to have been fixed at some point. Maybe the package just needs to
be rebuilt and updated on the mirrors ?

FWIW, I seem to have much more ruby-* packages installed on my system
than the total count of gemfiles installed:

$ dpkg -l | grep ruby- | grep ^ii | wc -l
187
$ ls /usr/share/rubygems-integration/1.8/specifications/ | wc -l
96

Thanks !


-- System Information:
Debian Release: 7.0
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-4-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages ruby-diff-lcs depends on:
ii  ruby1.8 [ruby-interpreter]1.8.7.358-6
ii  ruby1.9.1 [ruby-interpreter]  1.9.3.194-5

ruby-diff-lcs recommends no packages.

ruby-diff-lcs suggests no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#694337: lxc: please make initscript work for containers with rootfs on block devices

[Marc Fournier]

Package: lxc
Version: 0.8.0-1
Severity: wishlist
Tags: patch

Dear Maintainer,

Actually when calling, /etc/init.d/lxc stop, lxc-halt will get called
for every containers which have auto-start enabled.

As mentioned in #683443, this doesn't work for containers which have
rootfs on a block device (the host system cannot chroot into a container
in this case).

On the other hand, the newly introduced lxc-shutdown requires the
administrator to modify each container's inittab (or whatever /sbin/init
variant is used).

I feel there's no one size fits all-solution. So my suggestion
(implemented in the attached patch) would be to delegate this choice to
the administrator, via an LXC_SHUTDOWN entry in /etc/default/lxc.

The patch also adds a debconf entry which will ask which shutdown system
should be used.

Thanks !

Cheers,
Marc

-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages lxc depends on:
ii  debconf [debconf-2.0]  1.5.46
ii  libapparmor1   2.7.103-4
ii  libc6  2.13-35
ii  libcap21:2.22-1.2
ii  multiarch-support  2.13-35

Versions of packages lxc recommends:
ii  cdebootstrap  0.5.9
ii  debootstrap   1.0.42
ii  libcap2-bin   1:2.22-1.2

lxc suggests no packages.

-- debconf information excluded
diff -ru lxc-0.8.0.orig/debian/lxc.config lxc-0.8.0/debian/lxc.config
--- lxc-0.8.0.orig/debian/lxc.config	2012-11-12 15:19:19.0 +0100
+++ lxc-0.8.0/debian/lxc.config	2012-11-25 11:05:14.540766013 +0100
@@ -13,6 +13,8 @@
 	db_set lxc/auto ${LXC_AUTO}
 
 	db_set lxc/directory ${LXC_DIRECTORY}
+
+	db_set lxc/shutdown ${LXC_SHUTDOWN}
 fi
 
 db_settitle lxc/title
@@ -23,4 +25,8 @@
 db_input high lxc/directory ${LXC_DIRECTORY} || true
 db_go
 
+db_settitle lxc/title
+db_input high lxc/shutdown ${LXC_SHUTDOWN} || true
+db_go
+
 db_stop
diff -ru lxc-0.8.0.orig/debian/lxc.init lxc-0.8.0/debian/lxc.init
--- lxc-0.8.0.orig/debian/lxc.init	2012-11-12 15:19:19.0 +0100
+++ lxc-0.8.0/debian/lxc.init	2012-11-24 21:21:16.144670238 +0100
@@ -30,6 +30,7 @@
 fi
 
 LXC_AUTO=${LXC_AUTO:-true}
+LXC_SHUTDOWN=${LXC_SHUTDOWN:-halt}
 
 Lxc ()
 {
@@ -99,13 +100,13 @@
 	stop)
 		log_daemon_msg Stopping Linux Containers
 
-		Lxc halt
+		Lxc ${LXC_SHUTDOWN}
 		;;
 
 	restart|force-reload)
 		log_daemon_msg Restarting Linux Containers
 
-		Lxc stop
+		Lxc ${LXC_SHUTDOWN}
 		Lxc start
 		;;
 
diff -ru lxc-0.8.0.orig/debian/lxc.postinst lxc-0.8.0/debian/lxc.postinst
--- lxc-0.8.0.orig/debian/lxc.postinst	2012-11-12 15:19:19.0 +0100
+++ lxc-0.8.0/debian/lxc.postinst	2012-11-25 11:34:50.731098562 +0100
@@ -14,6 +14,9 @@
 		db_get lxc/directory
 		LXC_DIRECTORY=${RET:-/var/lib/lxc} # string (w/o empty)
 
+		db_get lxc/shutdown
+		LXC_SHUTDOWN=${RET:-halt} # string (w/o empty)
+
 		db_stop
 
 		if [ ! -e ${_CONFFILE} ]
@@ -24,6 +27,7 @@
 
 LXC_AUTO=${LXC_AUTO}
 LXC_DIRECTORY=${LXC_DIRECTORY}
+LXC_SHUTDOWN=${LXC_SHUTDOWN}
 EOF
 		fi
 
diff -ru lxc-0.8.0.orig/debian/lxc.templates lxc-0.8.0/debian/lxc.templates
--- lxc-0.8.0.orig/debian/lxc.templates	2012-11-12 15:19:19.0 +0100
+++ lxc-0.8.0/debian/lxc.templates	2012-11-25 11:18:17.765711440 +0100
@@ -20,3 +20,17 @@
  Containers.
  .
  If unsure, use /var/lib/lxc (default).
+
+Template: lxc/shutdown
+Type: string
+Default: halt
+_Description: LXC shutdown method
+ Different lxc subcommands can be used to shutdown Linux Containers. All of
+ them have drawbacks.
+ .
+ lxc-halt(1) won't work for containers which have a dedicated block device as
+ rootfs. lxc-shutdown(1) requires the container's inittab to be configured to
+ respond to SIGINT and SIGPWR signals. lxc-stop(1) kills all the processes
+ inside the container, without allowing them to terminate cleanly.
+ .
+ If unsure, use halt (default).

Bug#683837: please support block devices as rootfs in lxc-debconfig

[Marc Fournier]

Package: lxc
Version: 0.8.0-1
Followup-For: Bug #683837

[re-posting here, sent yesterday to #687767 by mistake]

Hello,

I noticed this bug still occurs with 0.8.0-1.

I would like to submit the attached patch, which solves the problem in
my case (using LVM rootfs).

It's very similar to the 2 other patches, but uses tar instead of cp,
which should work for dotfiles.

Using rsync would be nicer, but this would add a dependency on the
package.

I hope one or another of these patches will find their way in the
package sooner or later :-)

BTW, thanks for making lxc-debconf, it's excellent !

Cheers,
Marc
--- /usr/share/lxc/templates/lxc-debconfig.distrib	2012-11-21 19:50:03.806793345 +0100
+++ /usr/share/lxc/templates/lxc-debconfig	2012-11-21 21:12:22.861169495 +0100
@@ -522,7 +522,8 @@
 	then
 		# system cache in the form of a directory
 		echo -n Copying system cache to ${_ROOTFS}...
-		cp -a /usr/share/lxc/cache/${_MODE}/${_DISTRIBUTION}_${_ARCHITECTURE} ${_ROOTFS} || return 1
+		mkdir -p ${_ROOTFS}
+		(tar -C /usr/share/lxc/cache/${_MODE}/${_DISTRIBUTION}_${_ARCHITECTURE} -c . | tar -C ${_ROOTFS} -x) || return 1
 	elif ls /usr/share/lxc/cache/${_MODE}/${_DISTRIBUTION}_${_ARCHITECTURE}.*  /dev/null 21
 	then
 		# system cache in the form of a tarball
@@ -546,7 +547,8 @@
 
 		# make a local copy of the minimal system
 		echo -n Copying local cache to ${_ROOTFS}...
-		cp -a ${_CACHE}/${_DISTRIBUTION}_${_ARCHITECTURE} ${_ROOTFS} || return 1
+		mkdir -p ${_ROOTFS}
+		(tar -C ${_CACHE}/${_DISTRIBUTION}_${_ARCHITECTURE} -c . | tar -C ${_ROOTFS} -x) || return 1
 	fi
 
 	# Add local repositories configured from preseed file

Bug#687767: please support block devices as rootfs in lxc-debconfig

[Marc Fournier]

Package: lxc
Version: 0.8.0-1
Followup-For: Bug #687767

Hello,

I noticed this bug still occurs with 0.8.0-1.

I would like to submit the attached patch, which solves the problem in
my case (using LVM rootfs).

It's very similar to the 2 other patches, but uses tar instead of cp,
which should work for dotfiles.

Using rsync would be nicer, but this would add a dependency on the
package.

I hope one or another of these patches will find their way in the
package sooner or later :-)

BTW, thanks for making lxc-debconf, it's excellent !

Cheers,
Marc
--- /usr/share/lxc/templates/lxc-debconfig.distrib	2012-11-21 19:50:03.806793345 +0100
+++ /usr/share/lxc/templates/lxc-debconfig	2012-11-21 21:12:22.861169495 +0100
@@ -522,7 +522,8 @@
 	then
 		# system cache in the form of a directory
 		echo -n Copying system cache to ${_ROOTFS}...
-		cp -a /usr/share/lxc/cache/${_MODE}/${_DISTRIBUTION}_${_ARCHITECTURE} ${_ROOTFS} || return 1
+		mkdir -p ${_ROOTFS}
+		(tar -C /usr/share/lxc/cache/${_MODE}/${_DISTRIBUTION}_${_ARCHITECTURE} -c . | tar -C ${_ROOTFS} -x) || return 1
 	elif ls /usr/share/lxc/cache/${_MODE}/${_DISTRIBUTION}_${_ARCHITECTURE}.*  /dev/null 21
 	then
 		# system cache in the form of a tarball
@@ -546,7 +547,8 @@
 
 		# make a local copy of the minimal system
 		echo -n Copying local cache to ${_ROOTFS}...
-		cp -a ${_CACHE}/${_DISTRIBUTION}_${_ARCHITECTURE} ${_ROOTFS} || return 1
+		mkdir -p ${_ROOTFS}
+		(tar -C ${_CACHE}/${_DISTRIBUTION}_${_ARCHITECTURE} -c . | tar -C ${_ROOTFS} -x) || return 1
 	fi
 
 	# Add local repositories configured from preseed file

Bug#636554: jruby: Another new upstream release

[Marc Fournier]

Package: jruby
Version: 1.5.6-4
Followup-For: Bug #636554

Hello,

JRuby 1.7.0 got release a few days ago:
http://jruby.org/2012/10/22/jruby-1-7-0

Would it be an idea to focus on this version instead of 1.6 ?

Thanks !


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#688305: regression: --rsyncable disappeared from manpage

[Marc Fournier]

Package: gzip
Version: 1.5-1.1
Severity: minor
Tags: patch

Dear Maintainer,

A long time ago, #289616 added documentation for the --rsyncable option
in the gzip(1) manpage.

For some reason, the patch is no longer part of
debian/patches/rsyncable.diff and therefore doesn't appear in the manpage.

Appending the attached patch to debian/patches/rsyncable.diff should fix the
problem.

$ apt-cache policy gzip
gzip:
  Installed: 1.5-1.1
  Candidate: 1.5-1.1
  Version table:
 *** 1.5-1.1 0
700 http://http.debian.net/debian/ testing/main amd64 Packages
300 http://http.debian.net/debian/ unstable/main amd64 Packages
100 /var/lib/dpkg/status
 1.3.12-9 0
500 http://http.debian.net/debian/ stable/main amd64 Packages

$ debsums gzip | grep gzip.1.gz
/usr/share/man/man1/gzip.1.gz OK

$ zgrep -c rsync /usr/share/man/man1/gzip.1.gz
0

Thanks !

-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-3-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages gzip depends on:
ii  dpkg  1.16.8
ii  install-info  4.13a.dfsg.1-10
ii  libc6 2.13-35

gzip recommends no packages.

Versions of packages gzip suggests:
ii  less  444-4

-- no debconf information
--- gzip.1.orig	2012-09-21 11:35:29.381116294 +0200
+++ gzip.1	2012-09-21 11:37:06.610287168 +0200
@@ -13,6 +13,7 @@
 .br
 .B gunzip
 .RB [  \-acfhlLnNrtvV  ]
+.RB [ --rsyncable ]
 .RB [ \-S\ suffix ]
 [
 .I name \...
@@ -290,6 +291,16 @@
 .I gunzip
 ).
 .TP
+.B --rsyncable
+While compressing, synchronize the output occasionally based on the input.
+This increases size by less than 1 percent most cases, but means that the
+.BR rsync (1)
+program can take advantage of similarities in the uncompressed input
+when syncronizing two files compressed with this flag.
+.I gunzip
+cannot tell the difference between a compressed file created with this option,
+and one created without it.
+.TP
 .B \-S .suf   --suffix .suf
 When compressing, use suffix .suf instead of .gz.
 Any non-empty suffix can be given, but suffixes

Bug#648176: postgis: pgsql2shp/shp2pgsql no longer in $PATH

[Marc Fournier]

Package: postgis
Version: 1.5.3-1
Severity: normal

Dear Maintainer,

Since 1.5.3, pgsql2shp and shp2pgsql have moved  from /usr/bin to
/usr/lib/postgresql/X.Y/bin and have moved from the postgis package to
the postgresql-X.Y-postgis package.

This is annoying because users expect these tools to be in their default
$PATH, and scripts/cron jobs/etc using them will break after upgrading
from 1.5.2 to 1.5.3.

Furthermore, it was previously possible to install these tools on a
postgresql client machine and use the -h switch to connect to a remote
postgresql server. All that was needed was libpq. Now, if you want to do
this, you are forced by dependencies to install a complete postgresql
server on your client machine.

In any case, thanks for the packaging work !

-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 3.0.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

postgis depends on no packages.

postgis recommends no packages.

Versions of packages postgis suggests:
ii  postgresql-9.1-postgis  1.5.3-1

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#614002: grub-common: [kfreebsd] need a mechanism to pass sysctl parameters to kernel at boot time

[Marc Fournier]

Package: grub-common
Version: 1.98+20100804-14
Severity: wishlist


Some sysctl parameters of FreeBSD kernels can only be set at boot time.

With the traditional FreeBSD bootloader, you would define them in
/boot/loader.conf. With grub, this is done by adding a line such as this one
in /boot/grub/grub.cfg:

  set kFreeBSD.kern.maxswzone=67108864

The problem is obviously that /boot/grub/grub.cfg gets overwritten each time
update-grub2 is run.

The Linux variant of the package has this $GRUB_CMDLINE_LINUX option in
/etc/default/grub which gets included in the right places in
/boot/grub/grub.cfg.
We need the same sort of thing for kFreeBSD.

If it's just a matter of patching /etc/grub.d/10_kfreebsd to handle a
$GRUB_CMDLINE_KFREEBSD option in /etc/default/grub, I would be glad to submit
a patch. But maybe this should be discussed first, or coordinated with upstream
or whatever. Just let me know...

Cheers,
Marc


-- System Information:
Debian Release: 6.0
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: kfreebsd-amd64 (x86_64)

Kernel: kFreeBSD 8.1-1-amd64
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages grub-common depends on:
ii  base-files  6.0  Debian base system miscellaneous f
ii  dpkg1.15.8.10Debian package management system
ii  gettext-base0.18.1.1-3   GNU Internationalization utilities
ii  install-info4.13a.dfsg.1-6   Manage installed documentation in 
ii  libc0.1 2.11.2-10Embedded GNU C Library: Shared lib
ii  libfreetype62.4.2-2.1FreeType 2 font engine, shared lib
ii  libnvpair0  8.1-4OpenSolaris name-value pair librar
ii  libzfs0 8.1-4FreeBSD ZFS library
ii  zlib1g  1:1.2.3.4.dfsg-3 compression library - runtime

Versions of packages grub-common recommends:
ii  os-prober 1.42   utility to detect other OSes on a 

Versions of packages grub-common suggests:
pn  grub-emu  none (no description available)
pn  multiboot-doc none (no description available)
pn  xorriso   none (no description available)

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#613567: processes plugin blocks shutdown and crashes collectd (kFreeBSD only)

[Marc Fournier]

Package: collectd
Version: 4.10.1-1+squeeze2
Severity: important


When running collectd with the processes plugin on a kFreeBSD kernel, I
noticed that it complains about a double free or corruption and then fails
to stop properly after hitting Ctrl-C. I am forced to use kill -9.

root@debian:~# egrep -v '^$|^#' /etc/collectd/collectd.conf
FQDNLookup true
LoadPlugin processes
Include /etc/collectd/filters.conf
Include /etc/collectd/thresholds.conf

root@debian:~# collectd -f
*** glibc detected *** collectd: double free or corruption (top): 
0x006390b0 ***
Initialization complete, entering read-loop.
^CExiting normally.
collectd: Stopping 5 read threads.
^C
Killed

When running in test mode (-T), collectd dumps a core file:

root@debian:~# collectd -T
*** glibc detected *** collectd: double free or corruption (top): 
0x00634be0 ***
Aborted (core dumped)

root@debian:~# gdb /usr/sbin/collectd /var/lib/collectd/collectd.core
GNU gdb (GDB) 7.0.1-debian
Copyright (C) 2009 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type show copying
and show warranty for details.
This GDB was configured as x86_64-kfreebsd-gnu.
For bug reporting instructions, please see:
http://www.gnu.org/software/gdb/bugs/...
Reading symbols from /usr/sbin/collectd...(no debugging symbols found)...done.
Reading symbols from /lib/libpthread.so.0...(no debugging symbols found)...done.
Loaded symbols for /lib/libpthread.so.0
Reading symbols from /usr/lib/libltdl.so.7...(no debugging symbols 
found)...done.
Loaded symbols for /usr/lib/libltdl.so.7
Reading symbols from /lib/libdl.so.2...(no debugging symbols found)...done.
Loaded symbols for /lib/libdl.so.2
Reading symbols from /lib/libc.so.0.1...(no debugging symbols found)...done.
Loaded symbols for /lib/libc.so.0.1
Reading symbols from /lib/ld-kfreebsd-x86-64.so.1...(no debugging symbols 
found)...done.
Loaded symbols for /lib/ld-kfreebsd-x86-64.so.1
Reading symbols from /usr/lib/collectd/processes.so...(no debugging symbols 
found)...done.
Loaded symbols for /usr/lib/collectd/processes.so
Reading symbols from /lib/libkvm.so.0...(no debugging symbols found)...done.
Loaded symbols for /lib/libkvm.so.0
Reading symbols from /lib/libbsd.so.0...(no debugging symbols found)...done.
Loaded symbols for /lib/libbsd.so.0
Reading symbols from /lib/libnss_files.so.2...(no debugging symbols 
found)...done.
Loaded symbols for /lib/libnss_files.so.2
Core was generated by `collectd'.
Program terminated with signal 6, Aborted.
#0  0x000800e9da77 in kill () from /lib/libc.so.0.1
(gdb) bt
#0  0x000800e9da77 in kill () from /lib/libc.so.0.1
#1  0x00080084ab20 in pthread_kill () from /lib/libpthread.so.0
#2  0x00080084ab52 in raise () from /lib/libpthread.so.0
#3  0x000800ea0413 in abort () from /lib/libc.so.0.1
#4  0x000800ed0d5e in ?? () from /lib/libc.so.0.1
#5  0x000800ed97d6 in ?? () from /lib/libc.so.0.1
#6  0x000800ede71a in free () from /lib/libc.so.0.1
#7  0x0008013b4a70 in _kvm_realloc () from /lib/libkvm.so.0
#8  0x0008013b5334 in kvm_getprocs () from /lib/libkvm.so.0
#9  0x0008011b0093 in ?? () from /usr/lib/collectd/processes.so
#10 0x0040baf2 in plugin_read_all_once ()
#11 0x00406085 in main ()
(gdb)

This problem doesn't occur on squeeze's Linux (amd64) kernel. I didn't test
any other architecture.

Thanks !
Marc

-- System Information:
Debian Release: 6.0
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: kfreebsd-amd64 (x86_64)

Kernel: kFreeBSD 8.1-1-amd64
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages collectd depends on:
ii  collectd-core  4.10.1-1+squeeze2 statistics collection and monitori
ii  libc0.12.11.2-10 Embedded GNU C Library: Shared lib
ii  librrd41.4.3-1   time-series data storage and displ

Versions of packages collectd recommends:
ii  libatk1.0-01.30.0-1  The ATK accessibility toolkit
ii  libc0.12.11.2-10 Embedded GNU C Library: Shared lib
ii  libcairo2  1.8.10-6  The Cairo 2D vector graphics libra
ii  libcurl3-gnutls7.21.0-1  Multi-protocol file transfer libra
ii  libdbi00.8.2-3   Database Independent Abstraction L
ii  libdbus-1-31.2.24-4  simple interprocess messaging syst
ii  libdbus-glib-1-2   0.88-2.1  simple interprocess messaging syst
ii  libesmtp5  1.0.4-5   LibESMTP SMTP client library
ii  libfontconfig1 2.8.0-2.1 generic font configuration library
ii  libfreetype6   2.4.2-2.1 FreeType 2 font engine, shared lib
ii  libgcrypt111.4.5-2   LGPL Crypto library - runtime libr
ii  

Bug#609050: include local configuration files in syslog-ng.conf

[Marc Fournier]

Package: syslog-ng
Version: 3.1.3-1
Severity: wishlist
Tags: patch


rsyslog has this convenient $IncludeConfig option, which is configured this way
by default on debian:

$IncludeConfig /etc/rsyslog.d/*.conf

This allows the admin to place rsyslog configuration snipplets into
/etc/rsyslog.d/ without needing to modify the configuration file shipped in the
debian package, or without having to sed/awk/parse it if configuring the system
using an automated process.

As it's possible to do the same thing with syslog-ng 3.x, using the include
directive, I would find nice to have the same sort of facility in syslog-ng as
in rsyslog. Thanks !

I've attached a patch which does this and works for me.

Cheers,
Marc

-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-trunk-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages syslog-ng depends on:
ii  libc6 2.11.2-7   Embedded GNU C Library: Shared lib
ii  libdbi0   0.8.2-3Database Independent Abstraction L
pn  libevtlog0none (no description available)
ii  libglib2.0-0  2.24.2-1   The GLib library of C routines
ii  libpcre3  8.02-1.1   Perl 5 Compatible Regular Expressi
ii  lsb-base  3.2-23.1   Linux Standard Base 3.2 init scrip

Versions of packages syslog-ng recommends:
ii  logrotate 3.7.8-6Log rotation utility

syslog-ng suggests no packages.
diff -ruN orig/debian/rules syslog-ng-3.1.3/debian/rules
--- orig/debian/rules	2010-11-14 12:46:15.0 +0100
+++ syslog-ng-3.1.3/debian/rules	2011-01-05 18:24:46.0 +0100
@@ -93,6 +93,7 @@
 	dh_installlogrotate
 	dh_installlogcheck
 	dh_installchangelogs ChangeLog
+	dh_installdirs
 	dh_strip
 	dh_compress
 	dh_fixperms
diff -ruN orig/debian/syslog-ng.conf syslog-ng-3.1.3/debian/syslog-ng.conf
--- orig/debian/syslog-ng.conf	2010-11-14 12:43:26.0 +0100
+++ syslog-ng-3.1.3/debian/syslog-ng.conf	2011-01-05 18:30:24.0 +0100
@@ -153,3 +153,11 @@
 # All messages send to a remote site
 #
 #log { source(s_src); destination(d_net); };
+
+
+# Include
+
+# Include configuration snippets placed in this directory.
+
+include /etc/syslog-ng/conf.d;
+
diff -ruN orig/debian/syslog-ng.dirs syslog-ng-3.1.3/debian/syslog-ng.dirs
--- orig/debian/syslog-ng.dirs	1970-01-01 01:00:00.0 +0100
+++ syslog-ng-3.1.3/debian/syslog-ng.dirs	2011-01-05 18:26:00.0 +0100
@@ -0,0 +1 @@
+/etc/syslog-ng/conf.d

Bug#605227: apache2.2-common: dissatisfaction about the conf.d/ mechanism

[Marc Fournier]

Package: apache2.2-common
Version: 2.2.16-3
Severity: wishlist


I noticed several bugs (#565738, #553173, #604980) complaining more or less
about the same thing: URL namespace pollution by packages placing configuration
snippets into /etc/apache2/conf.d/.

As stated in /usr/share/doc/apache2.2-common/README.Debian.gz:

Files and Directories in /etc/apache2:
[...]
conf.d/
[...]
If the local administrator is not comfortable with packages
activating their config files by default, it is possible
to change the 'Include /etc/apache2/conf.d/' in apache2.conf
into 'Include /etc/apache2/conf.d.enabled/' and create that
directory. He can then put symlinks to the files in conf.d
which he wants to enable into conf.d.enabled.

So these bugs could basically get closed with a wontfix tag.

As the same bugreports could be filed against most of the packages putting
stuff in /etc/apache2/conf.d/: nagios, gitweb, smb2www, etc, I believe this
problem really belongs to the apache package itself.

Several options would be possible:
 - include /etc/apache2/conf.d.enabled/ by default, instead of
   /etc/apache2/conf.d/
 - make a2enconf and a2disconf utilities
 - including /etc/apache2/conf.d/ only in /etc/apache2/sites-available/default
 - have a policy about what packages are allowed to put into
   /etc/apache2/conf.d/
 - ...

Maybe should this be discussed on debian-devel ?

Cheers,
Marc



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#565738: drupal6: please lower #565738 severity to non-RC

[Marc Fournier]

Excerpt from /usr/share/doc/apache2.2-common/README.Debian.gz:

If the local administrator is not comfortable with packages
activating their config files by default, it is possible
to change the 'Include /etc/apache2/conf.d/' in apache2.conf
into 'Include /etc/apache2/conf.d.enabled/' and create that
directory. He can then put symlinks to the files in conf.d
which he wants to enable into conf.d.enabled.


Given this bug:
 - has a workaround as suggested above
 - is the same as #553173 and #604980
 - could apply to many other packages (nagios, gitweb, doc-central, etc)
 - should probably get fixed in apache itself (#605227)
 - is a release blocker

I suggest its severity should get lowered.

Many thanks !



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#553173: javascript-common: please lower #553173 severity to non-RC

[Marc Fournier]

block 553173 by 605227
thanks

Excerpt from /usr/share/doc/apache2.2-common/README.Debian.gz:

If the local administrator is not comfortable with packages
activating their config files by default, it is possible
to change the 'Include /etc/apache2/conf.d/' in apache2.conf
into 'Include /etc/apache2/conf.d.enabled/' and create that
directory. He can then put symlinks to the files in conf.d
which he wants to enable into conf.d.enabled.


Given this bug:
 - has a workaround as suggested above
 - is the same as #565738 and #604980
 - could apply to many other packages (nagios, gitweb, doc-central, etc)
 - should probably get fixed in apache itself (#605227)
 - is a release blocker

I suggest its severity should get lowered.

Many thanks !



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#474913: javascript-common: please lower #474913 severity to non-RC

[Marc Fournier]

clone 474913 -1
retitle 474913 javascript-common overrides /javascript globally in apache2
retitle -1 javascript-common overrides /javascript globally in lighttpd
notfound -1 5
thanks

Excerpt from /usr/share/doc/apache2.2-common/README.Debian.gz:

If the local administrator is not comfortable with packages
activating their config files by default, it is possible
to change the 'Include /etc/apache2/conf.d/' in apache2.conf
into 'Include /etc/apache2/conf.d.enabled/' and create that
directory. He can then put symlinks to the files in conf.d
which he wants to enable into conf.d.enabled.


Given this bug:
 - has a workaround as suggested above
 - is the same as #565738 and #604980
 - could apply to many other packages (nagios, gitweb, doc-central, etc)
 - should probably get fixed in apache itself (#605227)
 - is a release blocker

I suggest its severity should get lowered.

Furthermore, I'm splitting this bug in:
 - javascript-common overrides /javascript globally in apache2
 - javascript-common overrides /javascript globally in lighttpd

The latter only being in javascript-common/8, it doesn't affect squeeze and
therefore shouldn't be listed in squeeze release blockers.

Many thanks !



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#603789: cdargs should not depend on emacsen-common

[Marc Fournier]

Package: cdargs
Version: 1.35-5
Severity: normal
Tags: patch


I believe this problem can be fixed by removing emacsen-common from the
package dependencies.

This package does not benefit anyhow from having emacsen installed (although
the opposite is true).

Furthermore, when upgrading/installing version 1.35-5 of this package on a
stripped-down machine, it pulls down a whole load of stuff you probably don't
want:

# dpkg -l cdargs | grep cdargs
ii  cdargs  1.35-3  bookmarks and 
browsing for the cd command
# apt-get install cdargs=1.35-5  
Reading package lists... Done
Building dependency tree   
Reading state information... Done
The following extra packages will be installed:
  anthy-common cdargs cpp cpp-4.4 defoma emacs23 emacs23-bin-common 
emacs23-common emacsen-common fontconfig
  fontconfig-config gconf2-common hicolor-icon-theme libanthy0 libasound2 
libatk1.0-0 libatk1.0-data
  libavahi-client3 libavahi-common-data libavahi-common3 libcairo2 libcroco3 
libcups2 libdatrie1 libdbus-glib-1-2
  libfont-freetype-perl libfontconfig1 libfontenc1 libfreetype6 libfribidi0 
libgconf2-4 libgd2-noxpm libgif4
  libgmp3c2 libgsf-1-114 libgsf-1-common libgtk2.0-0 libgtk2.0-bin 
libgtk2.0-common libice6 libidl0 libjasper1
  libjpeg62 libm17n-0 libmpfr4 liborbit2 libotf0 libpango1.0-0 
libpango1.0-common libpixman-1-0 libpng12-0
  librsvg2-2 libsm6 libthai-data libthai0 libtiff4 libxcb-render-util0 
libxcb-render0 libxcomposite1 libxcursor1
  libxdamage1 libxfixes3 libxfont1 libxft2 libxi6 libxinerama1 libxpm4 
libxrandr2 libxrender1 libxt6 m17n-contrib
  m17n-db ttf-dejavu-core x-ttcidfont-conf x11-common xfonts-encodings 
xfonts-utils
Suggested packages:
  cpp-doc gcc-4.4-locales defoma-doc psfontmgr dfontmgr emacs23-common-non-dfsg 
emacs23-el libasound2-plugins
  cups-common libgd-tools librsvg2-common gvfs libjasper-runtime m17n-docs 
ttf-japanese-gothic
  ttf-japanese-mincho ttf-thryomanes ttf-baekmuk ttf-arphic-gbsn00lp 
ttf-arphic-bsmi00lp ttf-arphic-gkai00mp
  ttf-arphic-bkai00mp librsvg2-bin gawk
The following NEW packages will be installed:
  anthy-common cpp cpp-4.4 defoma emacs23 emacs23-bin-common emacs23-common 
emacsen-common fontconfig
  fontconfig-config gconf2-common hicolor-icon-theme libanthy0 libasound2 
libatk1.0-0 libatk1.0-data
  libavahi-client3 libavahi-common-data libavahi-common3 libcairo2 libcroco3 
libcups2 libdatrie1 libdbus-glib-1-2
  libfont-freetype-perl libfontconfig1 libfontenc1 libfreetype6 libfribidi0 
libgconf2-4 libgd2-noxpm libgif4
  libgmp3c2 libgsf-1-114 libgsf-1-common libgtk2.0-0 libgtk2.0-bin 
libgtk2.0-common libice6 libidl0 libjasper1
  libjpeg62 libm17n-0 libmpfr4 liborbit2 libotf0 libpango1.0-0 
libpango1.0-common libpixman-1-0 libpng12-0
  librsvg2-2 libsm6 libthai-data libthai0 libtiff4 libxcb-render-util0 
libxcb-render0 libxcomposite1 libxcursor1
  libxdamage1 libxfixes3 libxfont1 libxft2 libxi6 libxinerama1 libxpm4 
libxrandr2 libxrender1 libxt6 m17n-contrib
  m17n-db ttf-dejavu-core x-ttcidfont-conf x11-common xfonts-encodings 
xfonts-utils
The following packages will be upgraded:
  cdargs
1 upgraded, 76 newly installed, 0 to remove and 0 not upgraded.
[...]


With cdargs not depending on emacsen-common, installation is clean and lean,
and upgrade from 1.35-3 works like a charm:

# dpkg -l cdargs | grep cdargs
ii  cdargs  1.35-3  bookmarks and 
browsing for the cd command
# apt-get install cdargs=1.35-5.1
Reading package lists... Done
Building dependency tree   
Reading state information... Done
Suggested packages:
  emacs23 xemacs21 emacsen
The following packages will be upgraded:
  cdargs
1 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
Need to get 0 B/51.6 kB of archives.
After this operation, 12.3 kB disk space will be freed.
WARNING: The following packages cannot be authenticated!
  cdargs
Install these packages without verification [y/N]? y
(Reading database ... 25886 files and directories currently installed.)
Preparing to replace cdargs 1.35-3 (using ..././cdargs_1.35-5.1_amd64.deb) ...
Unpacking replacement cdargs ...
Processing triggers for man-db ...
Setting up cdargs (1.35-5.1) ...


-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-trunk-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages cdargs depends on:
ii  emacsen-common1.4.19 Common facilities for all emacsen
ii  libc6 2.11.2-7   Embedded GNU C Library: Shared lib
ii  libgcc1   1:4.4.5-6  GCC support library
ii  libncurses5   5.7+20100313-4 shared libraries for terminal hand
ii  libstdc++64.4.5-6  

Bug#596367: nscd doesn't run on GNU/kFreeBSD

[Marc Fournier]

Package: nscd
Version: 2.11.2-2
Severity: important


nscd fails to create it's socket when running on a freebsd kernel:

# nscd --debug
Fri 10 Sep 2010 08:02:42 PM CEST - 32543: /var/run/nscd/socket: Socket 
operation on non-socket

# ls -l /var/run/nscd/
total 0


-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'testing-proposed-updates'), (20, 
'unstable')
Architecture: kfreebsd-amd64 (x86_64)

Kernel: kFreeBSD 8.1-1-amd64
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages nscd depends on:
ii  libc0.1   2.11.2-2   Embedded GNU C Library: Shared lib

nscd recommends no packages.

nscd suggests no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#596373: varnish postinst script fails on GNU/kFreeBSD

[Marc Fournier]

Package: varnish
Version: 2.1.2-1
Severity: important
Tags: patch

*** Please type your report below this line ***

The postinst script copies /proc/sys/kernel/random/uuid to /etc/varnish/secret,
but this proc entry isn't available on GNU/kFreeBSD:

# dpkg --configure varnish
Setting up varnish (2.1.2-1) ...
install: cannot stat `/proc/sys/kernel/random/uuid': No such file or directory
dpkg: error processing varnish (--configure):
 subprocess installed post-installation script returned error exit status 1
Errors were encountered while processing:
 varnish

Here is a suggestion which generates a random secret using tools found in
coreutils, and without relying on /proc entries:


--- /var/lib/dpkg/info/varnish.postinst.orig2010-09-10 21:16:00.0 
+0200
+++ /var/lib/dpkg/info/varnish.postinst 2010-09-10 21:27:41.0 +0200
@@ -38,7 +38,9 @@
 
 varnish_create_secret() {
if ! [ -f ${secret_file} ]; then
-   install -m 0600 /proc/sys/kernel/random/uuid ${secret_file}
+   touch ${secret_file}  \
+   chmod 0600 ${secret_file}  \
+   dd if=/dev/urandom count=1 bs=128 2/dev/null | tr -dc 
A-Za-z0-9  ${secret_file}
fi
 }
 

(tested and works for me).

Cheers,
Marc

-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'testing-proposed-updates'), (20, 
'unstable')
Architecture: kfreebsd-amd64 (x86_64)

Kernel: kFreeBSD 8.1-1-amd64
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages varnish depends on:
ii  adduser   3.112  add and remove users and groups
ii  gcc   4:4.4.4-2  The GNU C compiler
ii  libc0.1   2.11.2-2   Embedded GNU C Library: Shared lib
ii  libc0.1-dev [libc6-dev]   2.11.2-2   Embedded GNU C Library: Developmen
ii  libncurses5   5.7+20100313-2 shared libraries for terminal hand
ii  libvarnish1   2.1.2-1shared libraries for Varnish

varnish recommends no packages.

varnish suggests no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#558121: tomcat6: Extra components missing from package

[Marc Fournier]

 Marc Fournier wrote:
  The package is currently missing the extra components (currently a
  complete commons-logging implementation
 
 That sounds like code duplication. It is supposedly a package-renamed
 version of Commons Logging?

I'm not sure. Maybe there could also be some stuff borrowed from log4j. I
fell into this problem following these instructions:
http://tomcat.apache.org/tomcat-6.0-doc/logging.html#log4j

Apparently the source code shipped in both packages is different:

$ apt-get source tomcat6 libcommons-logging-java
[...]
$ diff -r -q tomcat6-6.0.20/java/org/apache/juli/logging/ 
libcommons-logging-java-1.1.1/src/java/org/apache/commons/logging/ 
Only in tomcat6-6.0.20/java/org/apache/juli/logging/: DirectJDKLog.java
Only in libcommons-logging-java-1.1.1/src/java/org/apache/commons/logging/: impl
Files 
tomcat6-6.0.20/java/org/apache/juli/logging/LogConfigurationException.java and 
libcommons-logging-java-1.1.1/src/java/org/apache/commons/logging/LogConfigurationException.java
 differ
Files tomcat6-6.0.20/java/org/apache/juli/logging/LogFactory.java and 
libcommons-logging-java-1.1.1/src/java/org/apache/commons/logging/LogFactory.java
 differ
Files tomcat6-6.0.20/java/org/apache/juli/logging/Log.java and 
libcommons-logging-java-1.1.1/src/java/org/apache/commons/logging/Log.java 
differ
Only in libcommons-logging-java-1.1.1/src/java/org/apache/commons/logging/: 
LogSource.java
Files tomcat6-6.0.20/java/org/apache/juli/logging/package.html and 
libcommons-logging-java-1.1.1/src/java/org/apache/commons/logging/package.html 
differ

On the other hand, both tomcat6-6.0.20/{build,extras}.xml (and incidently
tomcat6-6.0.20/debian/rules) seem to download quite a few files. So I'm not
really sure what ends up in the package once it's built.


  and support for JSR 109)
  described on this page:
http://tomcat.apache.org/tomcat-6.0-doc/extras.html
 
 That page says it is licensed under CPL 1.0, which might be a problem.

Apparently CPL is approved by OSI as well as FSF, but I'm no licence
expert... Could it be a problem for Debian if it isn't for the Apache
Fondation ?

In any case, as with Commons Logging, the files seem to be downloaded by
ANT, not shipped in the source package.

Cheers,
Marc




-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#558121: tomcat6: Extra components missing from package

[Marc Fournier]

Package: tomcat6
Version: 6.0.20-7
Severity: wishlist


The package is currently missing the extra components (currently a
complete commons-logging implementation and support for JSR 109)
described on this page:
  http://tomcat.apache.org/tomcat-6.0-doc/extras.html

As they can be useful to some users and they aren't really complicated to
build, it could be nice to have them available in tomcat6-common, or
maybe as a separate tomcat6-extras package.


-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (500, 
'stable'), (300, 'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.30-2-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages tomcat6 depends on:
ii  adduser 3.111add and remove users and groups
ii  jsvc1.0.2~svn20061127-10 wrapper to launch Java application
ii  tomcat6-common  6.0.20-7 Servlet and JSP engine -- common f

tomcat6 recommends no packages.

Versions of packages tomcat6 suggests:
pn  tomcat6-admin none (no description available)
pn  tomcat6-docs  none (no description available)
pn  tomcat6-examples  none (no description available)

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#542815: maybe related to #538178 ?

[Marc Fournier]

I had the same error message when using mach and yum today, and found
out that a guy also reported this problem in #538178.

In my case, this was easily solved by removing the files in
/usr/lib/python2.5/site-packages/rpm/ (which didn't belong to any installed
package):

lrwxrwxrwx  1 root root   36 2008-05-22 13:54 _rpmmodule.a - 
/usr/share/pyshared/rpm/_rpmmodule.a
lrwxrwxrwx  1 root root   37 2008-05-22 13:54 _rpmmodule.la - 
/usr/share/pyshared/rpm/_rpmmodule.la
lrwxrwxrwx  1 root root   35 2008-05-22 13:54 __init__.py - 
/usr/share/pyshared/rpm/__init__.py

Maybe it would be an idea to investigate why different people have their
rpm-related python package get broken because of some (apparently)
generated files ?

Let me know if can be of any help !

Marc




-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#511891: iptables: typo in extensions/libxt_multiport.man

[Marc Fournier]

Package: iptables
Version: 1.4.1.1-3
Severity: minor
Tags: patch


I noticed a small typo in iptables manpage. The multiport module accepts
--sports/--dports parameters, not --sport/--dport.

Note: 0405-hyphenation.man.patch already patches the same lines. So the patch
I've attached must be applied after 0405-hyphenation.man.patch.


-- System Information:
Debian Release: 5.0
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (300, 
'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.26-1-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages iptables depends on:
ii  libc6 2.7-18 GNU C Library: Shared libraries

iptables recommends no packages.

iptables suggests no packages.

-- no debconf information
--- extensions/libxt_multiport.man.orig 2009-01-15 12:53:47.0 +0100
+++ extensions/libxt_multiport.man  2009-01-15 12:54:06.0 +0100
@@ -5,7 +5,7 @@
 or
 .BR \-p udp .
 .TP
-[\fB!\fP] \fB\-\-source-ports\fP,\fB\-\-sport\fP 
\fIport\fP[\fB,\fP\fIport\fP|\fB,\fP\fIport\fP\fB:\fP\fIport\fP]...
+[\fB!\fP] \fB\-\-source-ports\fP,\fB\-\-sports\fP 
\fIport\fP[\fB,\fP\fIport\fP|\fB,\fP\fIport\fP\fB:\fP\fIport\fP]...
 Match if the source port is one of the given ports.  The flag
 .B \-\-sports
 is a convenient alias for this option. Multiple ports or port ranges are
@@ -13,7 +13,7 @@
 \fB53,1024:65535\fP would therefore match ports 53 and all from 1024 through
 65535.
 .TP
-[\fB!\fP] \fB\-\-destination-ports\fP,\fB\-\-dport\fP 
\fIport\fP[\fB,\fP\fIport\fP|\fB,\fP\fIport\fP\fB:\fP\fIport\fP]...
+[\fB!\fP] \fB\-\-destination-ports\fP,\fB\-\-dports\fP 
\fIport\fP[\fB,\fP\fIport\fP|\fB,\fP\fIport\fP\fB:\fP\fIport\fP]...
 Match if the destination port is one of the given ports.  The flag
 .B \-\-dports
 is a convenient alias for this option.

Bug#509438: libaugeas-ruby: missing augeas.rb in package

[Marc Fournier]

Package: libaugeas-ruby
Version: 0.2.0-1
Severity: grave
Tags: patch
Justification: renders package unusable


The package ships without the ruby wrapper for augeas which is included in the
upstream ruby gem.

I believe the attached patch solves the problem.

By the way, I suggest decreasing debhelper version from 7 to 5 to allow folks
to backport the package to etch:


diff -urN libaugeas-ruby-0.2.0.orig/debian/compat 
libaugeas-ruby-0.2.0/debian/compat
--- libaugeas-ruby-0.2.0.orig/debian/compat 2008-12-22 12:55:24.0 +0100
+++ libaugeas-ruby-0.2.0/debian/compat  2008-12-22 12:51:36.0 +0100
@@ -1 +1 @@
-7
+5
diff -urN libaugeas-ruby-0.2.0.orig/debian/control 
libaugeas-ruby-0.2.0/debian/control
--- libaugeas-ruby-0.2.0.orig/debian/control  2008-12-22 12:55:24.0 
+0100
+++ libaugeas-ruby-0.2.0/debian/control 2008-12-22 12:51:36.0 +0100
@@ -2,7 +2,8 @@
 Section: libs
 Priority: optional
 Maintainer: Bart Cortooms b...@kumina.nl
-Build-Depends: debhelper (= 7), ruby1.8, ruby1.8-dev, ruby1.9, ruby1.9-dev, 
cdbs, ruby-pkg-tools (= 0.14), libaugeas-dev, pkg-config
+Build-Depends: debhelper (= 5), ruby1.8, ruby1.8-dev, ruby1.9, ruby1.9-dev, 
cdbs, ruby-pkg-tools (= 0.11), libaugeas-dev, pkg-config
 Standards-Version: 3.8.0
 Homepage: http://augeas.net/



-- System Information:
Debian Release: 5.0
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (300, 
'unstable'), (100, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.26-1-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages libaugeas-ruby depends on:
ii  libaugeas-ruby1.8 0.2.0-1Augeaus bindings for the Ruby lang

libaugeas-ruby recommends no packages.

libaugeas-ruby suggests no packages.

-- no debconf information
diff -urN libaugeas-ruby-0.2.0.orig/debian/libaugeas-ruby1.8.dirs libaugeas-ruby-0.2.0/debian/libaugeas-ruby1.8.dirs
--- libaugeas-ruby-0.2.0.orig/debian/libaugeas-ruby1.8.dirs	1970-01-01 01:00:00.0 +0100
+++ libaugeas-ruby-0.2.0/debian/libaugeas-ruby1.8.dirs	2008-12-22 12:51:17.0 +0100
@@ -0,0 +1 @@
+usr/lib/ruby/1.8
diff -urN libaugeas-ruby-0.2.0.orig/debian/libaugeas-ruby1.9.dirs libaugeas-ruby-0.2.0/debian/libaugeas-ruby1.9.dirs
--- libaugeas-ruby-0.2.0.orig/debian/libaugeas-ruby1.9.dirs	1970-01-01 01:00:00.0 +0100
+++ libaugeas-ruby-0.2.0/debian/libaugeas-ruby1.9.dirs	2008-12-22 12:51:17.0 +0100
@@ -0,0 +1 @@
+usr/lib/ruby/1.9
diff -urN libaugeas-ruby-0.2.0.orig/debian/rules libaugeas-ruby-0.2.0/debian/rules
--- libaugeas-ruby-0.2.0.orig/debian/rules	2008-12-22 12:55:24.0 +0100
+++ libaugeas-ruby-0.2.0/debian/rules	2008-12-22 12:51:17.0 +0100
@@ -6,4 +6,10 @@
 include /usr/share/cdbs/1/rules/debhelper.mk
 include /usr/share/ruby-pkg-tools/1/class/ruby-extconf-rb.mk
 
+install/libaugeas-ruby1.8::
+	cp $(CURDIR)/lib/augeas.rb $(CURDIR)/debian/libaugeas-ruby1.8/usr/lib/ruby/1.8/augeas.rb
+
+install/libaugeas-ruby1.9::
+	cp $(CURDIR)/lib/augeas.rb $(CURDIR)/debian/libaugeas-ruby1.9/usr/lib/ruby/1.9/augeas.rb
+
 DEB_RUBY_SETUP_CMD = $(CURDIR)/ext/augeas/extconf.rb

Bug#508939: lyx: Cannot compile Math help - missing xargs.sty

[Pierre-Marc Fournier]

Package: lyx
Version: 1.6.1-1
Severity: normal

Hi,

Select Math from the Help menu. Then select DVI (or another format) from the
View menu. The document does not compile and a Latex error is displayed saying
xargs.sty cannot be found. Latex log below:

**

This is pdfTeXk, Version 3.141592-1.40.3 (Web2C 7.5.6) (format=latex 
2008.10.13)  16 DEC 2008 14:58
entering extended mode
 %-line parsing enabled.
**Math.tex
(./Math.tex
LaTeX2e 2005/12/01
Babel v3.8h and hyphenation patterns for english, usenglishmax, dumylang, noh
yphenation, loaded.

(/usr/share/texmf-texlive/tex/latex/koma-script/scrartcl.cls
Document Class: scrartcl 2006/07/30 v2.95b KOMA-Script document class (article)

(/usr/share/texmf-texlive/tex/latex/koma-script/scrkbase.sty
Package: scrkbase 2006/07/30 v2.95b KOMA-Script package (basics and keyval use)

(/usr/share/texmf-texlive/tex/latex/koma-script/scrlfile.sty
Package: scrlfile 2006/03/28 v2.95 KOMA-Script package (loading files)
Package scrlfile, 2006/03/28 v2.95 KOMA-Script package (loading files)
  Copyright (C) Markus Kohm

) (/usr/share/texmf-texlive/tex/latex/graphics/keyval.sty
Package: keyval 1999/03/16 v1.13 key=value parser (DPC)
\...@toks@=\toks14
))
Package scrkbase Info: You've used the obsolete option `12pt'.
(scrkbase) \KOMAoptions{fontsize=12pt} will be used instead.
(scrkbase) You should do this change too on input line 594.
(/usr/share/texmf-texlive/tex/latex/base/size12.clo
File: size12.clo 2005/09/16 v1.4f Standard LaTeX file (size option)
) (/usr/share/texmf-texlive/tex/latex/koma-script/typearea.sty
Package: typearea 2006/07/30 v2.95b KOMA-Script package (type area)
Package typearea, 2006/07/30 v2.95b KOMA-Script package (type area)
  Copyright (C) Frank Neukam, 1992-1994
  Copyright (C) Markus Kohm, 1994-2002

\...@bcor=\skip41
\...@div=\count79
Package scrkbase Info: You've used the obsolete option `a4paper'.
(scrkbase) \KOMAoptions{paper=a4} will be used instead.
(scrkbase) You should do this change too on input line 419.
\...@hblk=\skip42
\...@vblk=\skip43
\...@temp=\skip44
Package typearea Info: These are the values describing the layout:
(typearea) DIV  = 12
(typearea) BCOR = 21.33955pt
(typearea) \paperwidth  = 597.50793pt
(typearea)  \textwidth  = 432.12631pt
(typearea)  DIV-departure   = -2/100
(typearea)  \evensidemargin = 23.75806pt
(typearea)  \oddsidemargin  = -2.91641pt
(typearea) \paperheight = 845.04694pt
(typearea)  \textheight = 635.5pt
(typearea)  \topmargin  = -41.72441pt
(typearea)  \headheight = 18.125pt
(typearea)  \headsep= 21.75pt
(typearea)  \topskip= 12.0pt
(typearea)  \footskip   = 50.75pt
(typearea)  \baselineskip   = 14.5pt
(typearea)  on input line 832.
)
\...@part=\count80
\...@section=\count81
\...@subsection=\count82
\...@subsubsection=\count83
\...@paragraph=\count84
\...@subparagraph=\count85
\abovecaptionskip=\skip45
\belowcaptionskip=\skip46
\...@pti@n...@sid@b...@x=\box26
\...@figure=\count86
\...@table=\count87
\bibindent=\dimen102
) (/usr/share/texmf-texlive/tex/latex/base/fontenc.sty
Package: fontenc 2005/09/27 v1.99g Standard LaTeX package
(/usr/share/texmf-texlive/tex/latex/base/t1enc.def
File: t1enc.def 2005/09/27 v1.99g Standard LaTeX file
LaTeX Font Info:Redeclaring font encoding T1 on input line 43.
)) (/usr/share/texmf-texlive/tex/latex/base/inputenc.sty
Package: inputenc 2006/05/05 v1.1b Input encoding file
\inp...@prehook=\toks15
\inp...@posthook=\toks16
(/usr/share/texmf-texlive/tex/latex/base/latin9.def
File: latin9.def 2006/05/05 v1.1b Input encoding file
)) (/usr/share/texmf-texlive/tex/latex/graphics/color.sty
Package: color 2005/11/14 v1.0j Standard LaTeX Color (DPC)
(/etc/texmf/tex/latex/config/color.cfg
File: color.cfg 2007/01/18 v1.5 color configuration of teTeX/TeXLive
)
Package color Info: Driver file: dvips.def on input line 130.
(/usr/share/texmf-texlive/tex/latex/graphics/dvips.def
File: dvips.def 1999/02/16 v3.0i Driver-dependant file (DPC,SPQR)
) (/usr/share/texmf-texlive/tex/latex/graphics/dvipsnam.def
File: dvipsnam.def 1999/02/16 v3.0i Driver-dependant file (DPC,SPQR)
)) (/usr/share/texmf-texlive/tex/generic/babel/babel.sty
Package: babel 2005/11/23 v3.8h The Babel package
(/usr/share/texmf-texlive/tex/generic/babel/ngermanb.ldf
Language: ngermanb 2004/02/20 v2.6m new German support from the babel system
(/usr/share/texmf-texlive/tex/generic/babel/babel.def
File: babel.def 2005/11/23 v3.8h Babel common definitions
\ba...@savecnt=\count88
\...@d=\dimen103
)

Package babel Warning: No hyphenation patterns were loaded for
(babel)the language `ngerman'
(babel)I will use the patterns loaded for \language=0 instead.

\...@ngerman = a 

Bug#497940: Augeas and Puppet

[Marc Fournier]

Hello,

 I'd be happy to work together with you on this Marc, if you're  
 interested. Your package looks pretty similar to the one I made so it  
 should be easy to merge the two.

 You can get the source for my package with:

   git clone git://git.kumina.nl/debian/libaugeas-ruby.git

Indeed both the packages are pretty similar.

I've successfully built your package on lenny, sid and ubuntu-intrepid. I
noticed yours doesn't provide augeas.rb (this 3 line file that let's you do
require 'augeas' from ruby). Attached is a patch that corrects this. I'm
not sure if it's the best way to do it but it works for me. This way we
have the best of both packages.

Neither your package nor mine build on etch and older ubuntus so I'm
planning to make another package for these legacy distributions. Nothing
worth being submited to debian, but I'll post to the list when I'm done.
Maybe other people could need them too.


 Perhaps Bart could have a look at your packaging work and see about
 integrating anything. I'm not sure if either of you are interested in
 co-maintainership, or a team-maintainance strategy, but they might be
 worth considering.

Wouldn't it be better to have this package maintained by the Debian Ruby
Extras team ? They already seem to manage most of the ruby libraries in
debian. I'm not sure I've understood well how debian teams work but it
seems to me that people outside debian can also collaborate on packages by
taking part in these teams. Micah, what do you think ?

Anyway I would be glad to help this team or collaborate with Bart if I can
be of any use.

 Also, it seems like you are primarily focused  around Ubuntu, so it
 could be that you take on the Ubuntu packaging, and art the Debian side?

It seems to me this package is so simple, the automated debian to ubuntu
merge will be sufficient. Augeas packages already found their way into
current ubuntu thanks to this procedure. So I think we would better focus
on getting this library into debian.

Marc

diff --git a/debian/changelog b/debian/changelog
index 34c1712..1142bf6 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,10 @@
+libaugeas-ruby (0.2.0-1.1) unstable; urgency=low
+
+  * Non-maintainer upload.
+  * Added missing augeas.rb to package.
+
+ -- Marc Fournier [EMAIL PROTECTED]  Thu, 13 Nov 2008 18:04:59 +0100
+
 libaugeas-ruby (0.2.0-1) unstable; urgency=low
 
   * Initial release. (Closes: #497940)
diff --git a/debian/libaugeas-ruby1.8.dirs b/debian/libaugeas-ruby1.8.dirs
new file mode 100644
index 000..a5e9841
--- /dev/null
+++ b/debian/libaugeas-ruby1.8.dirs
@@ -0,0 +1 @@
+usr/lib/ruby/1.8
diff --git a/debian/libaugeas-ruby1.9.dirs b/debian/libaugeas-ruby1.9.dirs
new file mode 100644
index 000..4bb305a
--- /dev/null
+++ b/debian/libaugeas-ruby1.9.dirs
@@ -0,0 +1 @@
+usr/lib/ruby/1.9
diff --git a/debian/rules b/debian/rules
index 56ed4a5..40944d6 100755
--- a/debian/rules
+++ b/debian/rules
@@ -6,4 +6,10 @@
 include /usr/share/cdbs/1/rules/debhelper.mk
 include /usr/share/ruby-pkg-tools/1/class/ruby-extconf-rb.mk
 
+install/libaugeas-ruby1.8::
+	cp $(CURDIR)/lib/augeas.rb $(CURDIR)/debian/libaugeas-ruby1.8/usr/lib/ruby/1.8/augeas.rb
+
+install/libaugeas-ruby1.9::
+	cp $(CURDIR)/lib/augeas.rb $(CURDIR)/debian/libaugeas-ruby1.9/usr/lib/ruby/1.9/augeas.rb
+
 DEB_RUBY_SETUP_CMD = $(CURDIR)/ext/augeas/extconf.rb


signature.asc
Description: Digital signature

Bug#497940: RFP: libaugeas-ruby1.8 -- Ruby bindings for Augeas

[Marc Fournier]

Package: wnpp
Severity: wishlist


* Package name: libaugeas-ruby1.8
  Version : 0.2.0
  Upstream Author : David Lutterkort [EMAIL PROTECTED]
* URL : http://augeas.net/download.html
* License : LGPL
  Description : Ruby bindings for Augeas

  Augeas is a library and command line tool that focuses on the most
  basic problem in handling Linux configurations programmatically: editing
  actual configuration files in a controlled manner.
  .
  This module provides a ruby interface to the Augeas API.


-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (200, 
'unstable')
Architecture: amd64 (x86_64)



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#483151: sidebar forgets the number of NEW messages, and only shows null entries

[Marc Fournier]

Hello,

  I don't have this problem, but the same mutt-patched version, and use
  local maildir.
 
 I can confirm the above problem, my mails are stored on an IMAP server.


I also confirm the problem with mailboxes hosted on an imap server. But I am
also able to reproduce it with local Maildir mailboxes. The problem can be
triggered by fiddling with the folder/spoolfile parameters.

Here is a minimal muttrc I used for testing:

[EMAIL PROTECTED]:~$ cat muttrc.test 
set spoolfile=~/Maildir/
mailboxes = \
  =.Work.KIS.Sysadmin/ \
  =.Work.C2C.Sysadmin/ \
  =.Perso/ \

bind index \CP sidebar-prev
bind index \CN sidebar-next
bind index \CO sidebar-open


When running mutt with this config file (mutt -F muttrc.test), the
sidebar displays ok (except the messages count):

INBOX 0|
 .Work.KIS.Sysadm 0|
 .Work.C2C.Sysadm 0|
 .Perso   0|

But when changing folder with ctrl-o, mutt replies:

/home/marc/Mail/.Perso is not a mailbox.


If I add the following option to my muttrc.test:

set folder=~/Maildir/

I am able to open my folders fine, the message count is correct, but the
names are lost:

INBOX   759|
null 8034|
null 6225|
null  558|

Let me know if I can provide more info or something.

Cheers,
Marc



signature.asc
Description: Digital signature

Bug#482848: libgphoto2-2: PTP not working with Panasonic DMC-FX30

[Marc Fournier]

Package: libgphoto2-2
Version: 2.4.0-9
Severity: normal


When trying to use this camera in PTP mode, gphoto2 reports:

*** Error ***
PTP I/O error
27.813618 gphoto2-port(2): Closing port...
32.813079 context(0): An error occurred in the io-library ('Unspecified
error'): No error description available

*** Error ***
An error occurred in the io-library ('Unspecified error'): No error
description available
*** Error (-1: 'Unspecified error') ***

This bug is also described on ubuntu's launchpad:
https://bugs.launchpad.net/ubuntu/+source/gphoto2/+bug/173429

And seems to have been fixed upstream:
http://sourceforge.net/tracker/index.php?func=detailaid=1847471group_id=8874atid=108874

As the the problem seems to have been solved, I'm just reporting this
bug to be sure it get's fixed in debian too and to avoid work beeing
done once in ubuntu and once here.

Thanks !
Marc


-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (200, 
'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.24-1-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages libgphoto2-2 depends on:
ii  adduser   3.107  add and remove users and groups
ii  libc6 2.7-10 GNU C Library: Shared libraries
ii  libexif12 0.6.16-2.1 library to parse EXIF files
ii  libgphoto2-port0  2.4.0-9gphoto2 digital camera port librar
ii  libjpeg62 6b-14  The Independent JPEG Group's JPEG 
ii  libltdl3  1.5.26-4   A system independent dlopen wrappe

Versions of packages libgphoto2-2 recommends:
ii  udev  0.114-2/dev/ and hotplug management daemo

-- no debconf information



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#474514: feh: corrupts jpeg images with file - edit in place - rotate

[Marc Fournier]

Package: feh
Version: 1.3.4.dfsg.1-1
Severity: normal


When rotating jpg pictures (90° clockwise), feh cuts a small margin from
the left and puts it on the right. It happens the other way round when
rotating CCW.

Here is an example of an image created with gimp, then rotated 90° with
feh: http://documents.epfl.ch/users/m/ma/mafourni/www/feh-bug.jpg

I noticed the size of the margin depends on the width-height ratio of
the image. With some width-height ratios, I was unable to reproduce the
problem, and with others it only happens when rotating 180°.

The problem doesn't seem to occur with tif, png and gif images. I
haven't tried any other formats.

-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates'), (200, 
'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.24-1-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages feh depends on:
ii  giblib1   1.2.4-5wrapper library for imlib2, and ot
ii  libc6 2.7-6  GNU C Library: Shared libraries
ii  libimlib2 1.4.0-1powerful image loading and renderi
ii  libjpeg62 6b-14  The Independent JPEG Group's JPEG 
ii  libpng12-01.2.15~beta5-3 PNG library - runtime
ii  libx11-6  2:1.0.3-7  X11 client-side library
ii  libxinerama1  2:1.0.3-1  X11 Xinerama extension library

Versions of packages feh recommends:
ii  wget  1.11.1-1   retrieves files from the web

-- no debconf information

Bug#470455: wfrench: encoded in ISO-8859

[Marc Fournier]

Package: wfrench
Version: 1.2.3-2
Severity: wishlist


The french wordlist suffers from the same problem than described in bug
437719.

A quick and easy fix would simply be to call recode l1..u8 in
build-stamp (same fix than in the wspanish package).
Maybe it would be cleaner to modify diacrit7to8.l, but it's a bit more
work.

Thanks !


-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.22-3-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages wfrench depends on:
ii  debconf   1.5.19 Debian configuration management sy
ii  dictionaries-common   0.96.1 Common utilities for spelling dict

wfrench recommends no packages.

-- debconf information:
  shared/packages-wordlist:
  wfrench/languages: francais (French)



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#470459: wngerman: encoded in ISO-8859

[Marc Fournier]

Package: wngerman
Version: 20070829-5
Severity: wishlist


The german (and the swiss one by the way) wordlist suffers from the same
problem than described in bug 437719 for the spanish wordlist.

Would it be possible to recode them to UTF-8 ? Should I also fill a bug
against the wswiss package ?

Thanks !

-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.22-3-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages wngerman depends on:
ii  debconf [debconf-2.0] 1.5.19 Debian configuration management sy
ii  dictionaries-common   0.96.1 Common utilities for spelling dict

wngerman recommends no packages.

-- debconf information:
  wngerman/languages: deutsch (New German)
  shared/packages-wordlist:



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#470461: witalian: encoded in ISO-8859

[Marc Fournier]

Package: witalian
Version: 1.7.3-0.1
Severity: wishlist


The italian wordlist suffers from the same problem than described in bug
437719 for the spanish wordlist.

Would it be possible to recode it to UTF-8 ? The spanish wordlist has
allready fixed the problem. The correction can probably easily be
copied from there.

Thanks !

-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.22-3-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages witalian depends on:
ii  debconf [debconf-2.0] 1.5.19 Debian configuration management sy
ii  dictionaries-common   0.96.1 Common utilities for spelling dict

witalian recommends no packages.

-- debconf information:
  witalian/languages: italiano (Italian)
  shared/packages-wordlist:



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#470463: wamerican: encoded in ISO-8859

[Marc Fournier]

Package: wamerican
Version: 6-2.1
Severity: wishlist


The different english wordlists (not only in wamerican) suffers from the
same problem than described in bug 437719 for the spanish wordlist.

Would it be possible to recode them to UTF-8 ? Maybe the fix in the
wspanish package could help provide inspiration ?

Thanks !

-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.22-3-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages wamerican depends on:
ii  debconf [debconf-2.0] 1.5.19 Debian configuration management sy
ii  dictionaries-common   0.96.1 Common utilities for spelling dict

wamerican recommends no packages.

-- debconf information:
  wamerican/languages: american (American English)
  shared/packages-wordlist:



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#459453: cadaver does not honor /etc/ssl/certs/ca-certificates.crt

[Marc Fournier]

Package: cadaver
Version: 0.23.0-1
Severity: normal


When accessing an https DAV volume, I get the following message even
if the server cert got signed by a trusted CA (ie: provided by the
ca-certificates package).

WARNING: Untrusted server certificate presented for `my.server.tld':
[...]
Do you wish to accept the certificate? (y/n)

I don't have any problem accessing the same server with curl or wget,
which both require an additional command-line option when an https
server presents a certificate which isn't signed by one of the CAs
listed in /etc/ssl/certs/ca-certificates.crt

This prevents using cadaver in a non-interactive way (in combination
with .netrc).

It seems a compile-time option exists, but I was unfortunately unable to
get it work (not enough ):

[EMAIL PROTECTED]:/tmp/cadaver-0.23.0$ ./configure  --help | grep CA
  --with-ca-bundlespecify filename of an SSL CA root bundle

I noticed someone assigned the same bug to libneon26 in ubuntu:
https://bugs.launchpad.net/ubuntu/+source/neon26/+bug/104777

Anyway, thanks for taking care of this useful tool !
Marc


-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (700, 'testing'), (500, 'testing-proposed-updates')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.22-3-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages cadaver depends on:
ii  libc6 2.7-5  GNU C Library: Shared libraries
ii  libncurses5   5.6+20071124-1 Shared libraries for terminal hand
ii  libneon27-gnutls  0.27.2-1   An HTTP and WebDAV client library 
ii  libreadline5  5.2-3  GNU readline and history libraries

cadaver recommends no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#459453: libneon does not honor /etc/ssl/certs/ca-certificates.crt

[Marc Fournier]

On Sun, Jan 06, 2008 at 06:50:44PM +0100, Sebastian Harl wrote:
 [...]
 
 This is a configuration option for libneon. As I'm linking cadaver against the
 packaged version of neon (libneon27-gnutls) instead of the one shipped with
 cadaver itself, it does not make any difference if you use this option when
 compiling cadaver.
 
  I noticed someone assigned the same bug to libneon26 in ubuntu:
  https://bugs.launchpad.net/ubuntu/+source/neon26/+bug/104777
 
 As noted in this bug report it seems to be sufficient to add
 --with-ca-bundle=/etc/ssl/certs/ca-certificates.crt to the configure options
 when compiling libneon. Imho libneon and libneon-gnutls should recommend
 ca-certificates then. A simple patch against version 0.27.2-1 is attached to
 this email. This is untested though, so it would be nice if Marc could verify
 that it works as expected.

Thanks Sebastian for your quick reaction, your patch works perfectly !
I did try to rebuild neon27, but I must have messed up something...

From my point of view, your solution can be incorporated into the package
and this bug can be closed.

By the way, it would probably make sense to patch neon25 and neon26 the
same way ?

Thanks again !
Marc




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]