Bug#1017557: libmilter1.0.1: Missing symlinks to libmilter.so.1.0.1
Package: libmilter1.0.1 Version: 8.15.2-22 Severity: important Dear Maintainer, * What led up to the situation? Trying to use a pre-compiled binary for `ctmilter`, an application that dynalinks with `libmilter`. * What exactly did you do (or not do) that was effective (or ineffective)? The application wouldn't start because it failed to find `libmilter.so.1`. Checking with $ ldd ./ctmilter.bin linux-vdso.so.1 (0x7ffd14911000) libmilter.so.1.0 => not found libstdc++.so.6 => /lib/x86_64-linux-gnu/libstdc++.so.6 (0x7f07c7812000) libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x7f07c76ce000) libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x7f07c76b4000) libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x7f07c7692000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7f07c74cd000) /lib64/ld-linux-x86-64.so.2 (0x7f07c79e7000) led me to look at the contents of package `libmilter1.0.1` $ dpkg -L libmilter1.0.1 /. /usr /usr/lib /usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/libmilter.so.1.0.1 /usr/share /usr/share/bug /usr/share/bug/libmilter1.0.1 /usr/share/bug/libmilter1.0.1/control /usr/share/bug/libmilter1.0.1/script /usr/share/doc /usr/share/doc/libmilter1.0.1 /usr/share/doc/libmilter1.0.1/README.gz /usr/share/doc/libmilter1.0.1/changelog.Debian.gz /usr/share/doc/libmilter1.0.1/changelog.gz /usr/share/doc/libmilter1.0.1/copyright and then tried # cd /usr/lib/x86_64-linux-gnu # ln -s libmilter.so.1.0.1 libmilter.so.1.0 $ ldd ./ctmilter.bin linux-vdso.so.1 (0x7ffc71dd8000) libmilter.so.1.0 => /lib/x86_64-linux-gnu/libmilter.so.1.0 (0x7fc3c7e99000) libstdc++.so.6 => /lib/x86_64-linux-gnu/libstdc++.so.6 (0x7fc3c7ccc000) libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x7fc3c7b88000) libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x7fc3c7b6e000) libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x7fc3c7b4c000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x7fc3c7987000) /lib64/ld-linux-x86-64.so.2 (0x7fc3c7eb4000) After that, `ctmilter` worked fine. * What outcome did you expect instead? After reading Debian Policy, specifically https://www.debian.org/doc/debian-policy/ch-sharedlibs.html#s-sharedlibs-runtime I believe the package is missing symlinks libmilter.so.1 -> libmilter.so.1.0.1 libmilter.so.1.0 -> libmilter.so.1.0.1 -- Package-specific info: Output of /usr/share/bug/libmilter1.0.1/script: ls -alR /etc/mail: /etc/mail: total 348 drwxr-sr-x 7 smmta smmsp 4096 Sep 4 2021 . drwxr-xr-x 187 root root 12288 Aug 17 12:02 .. -rwxr-xr-- 1 root smmsp 10016 Sep 4 2021 Makefile -rw--- 1 root root 4265 Sep 4 2021 access -rw-r- 1 smmta smmsp 12288 Sep 4 2021 access.db -rw-r--r-- 1 root root281 Aug 25 2019 address.resolve lrwxrwxrwx 1 root smmsp10 Sep 14 2019 aliases -> ../aliases -rw-r- 1 smmta smmsp 12288 Sep 4 2021 aliases.db -rw-r--r-- 1 root root 3223 Sep 4 2021 databases -rw-r--r-- 1 root root 5659 Aug 25 2019 helpfile -rw-r--r-- 1 root smmsp32 Sep 14 2019 local-host-names drwxr-sr-x 2 smmta smmsp 4096 Sep 14 2019 m4 drwxr-xr-x 2 root root 4096 Sep 4 2021 peers drwxr-xr-x 2 root smmsp 4096 Aug 25 2019 sasl -rw-r--r-- 1 root smmsp 65170 Sep 4 2021 sendmail.cf -rw-r--r-- 1 root root 65313 Sep 4 2021 sendmail.cf.old -rw-r--r-- 1 root root 12235 Sep 4 2021 sendmail.conf -rw-r--r-- 1 root smmsp 4235 Sep 4 2021 sendmail.mc -rw-r--r-- 1 root root148 Aug 25 2019 service.switch -rw-r--r-- 1 root root179 Aug 25 2019 service.switch-nodns drwxr-sr-x 2 smmta smmsp 4096 Sep 4 2021 smrsh -rw-r--r-- 1 root smmsp 44527 Sep 4 2021 submit.cf -rw-r--r-- 1 root root 44662 Sep 4 2021 submit.cf.old -rw-r--r-- 1 root smmsp 2383 Sep 4 2021 submit.mc drwxr-xr-x 2 smmta smmsp 4096 Sep 14 2019 tls -rw-r--r-- 1 root smmsp 0 Sep 14 2019 trusted-users /etc/mail/m4: total 8 drwxr-sr-x 2 smmta smmsp 4096 Sep 14 2019 . drwxr-sr-x 7 smmta smmsp 4096 Sep 4 2021 .. -rw-r- 1 root smmsp0 Sep 14 2019 dialup.m4 -rw-r- 1 root smmsp0 Sep 14 2019 provider.m4 /etc/mail/peers: total 12 drwxr-xr-x 2 root root 4096 Sep 4 2021 . drwxr-sr-x 7 smmta smmsp 4096 Sep 4 2021 .. -rw-r--r-- 1 root root 328 Aug 25 2019 provider /etc/mail/sasl: total 8 drwxr-xr-x 2 root smmsp 4096 Aug 25 2019 . drwxr-sr-x 7 smmta smmsp 4096 Sep 4 2021 .. /etc/mail/smrsh: total 8 drwxr-sr-x 2 smmta smmsp 4096 Sep 4 2021 . drwxr-sr-x 7 smmta smmsp 4096 Sep 4 2021 .. lrwxrwxrwx 1 root smmsp 32 Sep 4 2021 mail.local -> /usr/libexec/sendmail/mail.local lrwxrwxrwx 1 root smmsp 17 Sep 14 2019 procmail -> /usr/bin/procmail /etc/mail/tls: total 48 drwxr-xr-x 2 smmta smmsp 4096 Sep 14 2019 . drwxr-sr-x 7
Bug#935013: Latest upstream (was: new upstream)
Any progress on this? I see the patch didn't make it into Debian 11. I've been using the patch for custom-built packages targeting Debian 10 and Debian 11, the latter with a simple `compat 10` change. There haven't been any changes to upstream since I provided the patch. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 0064 ADF5 EB5C DE16 99C1 6C56 F2A3 86B5 A757 E5A1
Bug#935013: validns: new upstream
On Sun, 18 Aug 2019 05:54:56 +0200 Joost van =?utf-8?Q?Baal-Ili=C4=87?=
wrote:
> Package: validns
> Version: 0.8+git20160720-3.1
> Severity: normal
>
> Prof. Ernesto Hernández-Novich (emhn-guest) offered help in private
> communication, and donated a patch in private communication.
> (Thanks!) His patch, which builds a current upstream, should get
> applied.
Attached you'll find the patch mentioned.
This patch provides updated `quilt` patches so upstream `validns`
commit f423245b9867359398f83e8a60fea167ad7694ca
Author: Anton Berezin
Date: Fri Aug 4 16:27:44 2017 +0200
builds against OpenSSL 1.1. I've been using this patch for over a
month with a Debian 9 targeted package, and no issues.
I wrote upstream asking whether or not they would update their code to
work with OpenSSL 1.1 but haven't gotten an answer.
Regards,
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 0064 ADF5 EB5C DE16 99C1 6C56 F2A3 86B5 A757 E5A1
From bc36d7d8b48874199118f786060a117a208c1f9e Mon Sep 17 00:00:00 2001
From: Ernesto Hernández-Novich
Date: Thu, 18 Jul 2019 16:00:44 -0700
Subject: Make latest `validns` build on Debian 9 (and later)
---
.../patches/fix-compilation-on-openssl-1.1.patch | 441 +++--
debian/patches/fix-dont-overwrite-cflags.patch | 12 +-
debian/patches/fix-makefile-clean.patch| 4 +-
...ddress-possible-string-truncation-warning.patch | 37 +-
4 files changed, 256 insertions(+), 238 deletions(-)
diff --git a/debian/patches/fix-compilation-on-openssl-1.1.patch b/debian/patches/fix-compilation-on-openssl-1.1.patch
index f20384c..45d815a 100644
--- a/debian/patches/fix-compilation-on-openssl-1.1.patch
+++ b/debian/patches/fix-compilation-on-openssl-1.1.patch
@@ -11,182 +11,129 @@ Signed-off-by: Sebastian Andrzej Siewior
rrsig.c | 69 ++-
3 files changed, 42 insertions(+), 65 deletions(-)
-diff --git a/dnskey.c b/dnskey.c
-index fecc62abfd21..fda220c14d08 100644
a/dnskey.c
-+++ b/dnskey.c
-@@ -154,6 +154,7 @@ int dnskey_build_pkey(struct rr_dnskey *rr)
- unsigned int e_bytes;
- unsigned char *pk;
- int l;
-+ BIGNUM *n, *e;
-
- rsa = RSA_new();
- if (!rsa)
-@@ -174,11 +175,15 @@ int dnskey_build_pkey(struct rr_dnskey *rr)
- if (l < e_bytes) /* public key is too short */
- goto done;
-
-- rsa->e = BN_bin2bn(pk, e_bytes, NULL);
-+ e = BN_bin2bn(pk, e_bytes, NULL);
- pk += e_bytes;
- l -= e_bytes;
-
-- rsa->n = BN_bin2bn(pk, l, NULL);
-+ n = BN_bin2bn(pk, l, NULL);
-+ if (!e || !n)
-+ goto done;
-+
-+ RSA_set0_key(rsa, n, e, NULL);
-
- pkey = EVP_PKEY_new();
- if (!pkey)
-diff --git a/nsec3checks.c b/nsec3checks.c
-index 69c655345bad..2abac9efa1bf 100644
a/nsec3checks.c
-+++ b/nsec3checks.c
-@@ -28,7 +28,7 @@
- static struct binary_data name2hash(char *name, struct rr *param)
- {
- struct rr_nsec3param *p = (struct rr_nsec3param *)param;
-- EVP_MD_CTX ctx;
-+ EVP_MD_CTX *ctx;
- unsigned char md0[EVP_MAX_MD_SIZE];
- unsigned char md1[EVP_MAX_MD_SIZE];
- unsigned char *md[2];
-@@ -45,26 +45,31 @@ static struct binary_data name2hash(char *name, struct rr *param)
-
- /* XXX Maybe use Init_ex and Final_ex for speed? */
-
-- EVP_MD_CTX_init();
-- if (EVP_DigestInit(, EVP_sha1()) != 1)
-+ ctx = EVP_MD_CTX_new();
-+ if (ctx == NULL)
- return r;
-- digest_size = EVP_MD_CTX_size();
-- EVP_DigestUpdate(, wire_name.data, wire_name.length);
-- EVP_DigestUpdate(, p->salt.data, p->salt.length);
-- EVP_DigestFinal(, md[mdi], NULL);
-+ if (EVP_DigestInit(ctx, EVP_sha1()) != 1)
-+ goto out;
-+ digest_size = EVP_MD_CTX_size(ctx);
-+ EVP_DigestUpdate(ctx, wire_name.data, wire_name.length);
-+ EVP_DigestUpdate(ctx, p->salt.data, p->salt.length);
-+ EVP_DigestFinal(ctx, md[mdi], NULL);
-
- for (i = 0; i < p->iterations; i++) {
-- if (EVP_DigestInit(, EVP_sha1()) != 1)
-- return r;
-- EVP_DigestUpdate(, md[mdi], digest_size);
-+ if (EVP_DigestInit(ctx, EVP_sha1()) != 1)
-+ goto out;
-+
-+ EVP_DigestUpdate(ctx, md[mdi], digest_size);
- mdi = (mdi + 1) % 2;
-- EVP_DigestUpdate(, p->salt.data, p->salt.length);
-- EVP_DigestFinal(, md[mdi], NULL);
-+ EVP_DigestUpdate(ctx, p->salt.data, p->salt.length);
-+ EVP_DigestFinal(ctx, md[mdi], NULL);
- }
-
- r.length = digest_size;
- r.data = getmem(digest_size);
- memcpy(r.data, md[mdi], digest_size);
-+out:
-+ EVP_MD_CTX_free(ctx);
- return r;
- }
-
-diff --git a/rrsig.c b/rrsig.c
-index 81f24b4c49da..0a9e864285d0 100644
--- a/rrsig.c
+++ b/rrsig.c
-@@ -26,7 +26,7 @@
+@@ -27,7 +27,7 @@
struct verification_data
{
- struct verification_data *next;
-- EVP_MD_CTX ctx;
-+ EVP_MD_CTX *ctx;
- struct rr_dnskey *key;
- struct rr_rrsig *rr;
- int ok;
-@@ -180,7 +180,7 @@ void *verification_thread(vo
Bug#789381: libpoe-api-peek-perl: FTBFS with perl 5.22: test failures
On Sat, 30 Jul 2016 14:35:09 +0200 intrigeri <intrig...@debian.org> wrote: > Hi, > > Prof. Ernesto Hernández-Novich: > > On Sun, 26 Jun 2016 15:27:05 +0200 > > intrigeri <intrig...@debian.org> wrote: > >> Prof. Ernesto Hernández-Novich wrote (23 May 2016 11:49:55 GMT) : > >> > On Sat, 21 May 2016 19:03:54 +0300 Niko Tyni <nt...@debian.org> > >> > wrote: webgui has never been part of any stable release. WebGUI's > >> > upstream has been very slow for the last year or so, with only a > >> > couple of releases. The latest release, about a month and a half > >> > ago, still lists POE::Component::IKC as a dependency and it's > >> > used for a core workflow component -- it requires 0.2001 > >> > >> > I've written upstream about their thoughts on the issue. > >> > >> Did you get any feedback from them yet? > > > I've gotten feedback. > > > They were not aware of the issue because they haven't tested WebGUI > > with 5.22. They will be reviewing the actual dependency to see if > > it is indeed needed or not. Perhaps we coud wait a week or so > > before removing the packages. > > A month later: any update on this side? Nothing. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 0064 ADF5 EB5C DE16 99C1 6C56 F2A3 86B5 A757 E5A1
Bug#789381: libpoe-api-peek-perl: FTBFS with perl 5.22: test failures
On Sun, 26 Jun 2016 15:27:05 +0200 intrigeri <intrig...@debian.org> wrote: > Hi, > > Prof. Ernesto Hernández-Novich wrote (23 May 2016 11:49:55 GMT) : > > On Sat, 21 May 2016 19:03:54 +0300 Niko Tyni <nt...@debian.org> > > wrote: webgui has never been part of any stable release. WebGUI's > > upstream has been very slow for the last year or so, with only a > > couple of releases. The latest release, about a month and a half > > ago, still lists POE::Component::IKC as a dependency and it's used > > for a core workflow component -- it requires 0.2001 > > > I've written upstream about their thoughts on the issue. > > Did you get any feedback from them yet? I've gotten feedback. They were not aware of the issue because they haven't tested WebGUI with 5.22. They will be reviewing the actual dependency to see if it is indeed needed or not. Perhaps we coud wait a week or so before removing the packages. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 0064 ADF5 EB5C DE16 99C1 6C56 F2A3 86B5 A757 E5A1
Bug#789381: libpoe-api-peek-perl: FTBFS with perl 5.22: test failures
On Sat, 21 May 2016 19:03:54 +0300 Niko Tyni <nt...@debian.org> wrote: > On Sat, Jun 20, 2015 at 01:54:30PM +0100, Dominic Hargreaves wrote: > > Source: libpoe-api-peek-perl > > Version: 2.2000-1 > > Severity: important > > User: debian-p...@lists.debian.org > > Usertags: perl-5.22-transition > > Tags: sid stretch upstream > > Forwarded: https://rt.cpan.org/Public/Bug/Display.html?id=103803 > > > > This package FTBFS with perl 5.22 (currently in experimental), since > > perl 5.22 is not supported by the package: > > There's been no reaction to this upstream in more than a year. > We discussed this package at the Debian Perl Team Sprint in Zürich, > and we think it should be removed from Debian. However, it has a > reverse dependency chain to webgui through libpoe-component-ikc-perl. > > I see webgui has been RC-buggy for a long time and is not part > of any recent stable release. I'm cc'ing the maintainer, Ernesto > Hernández-Novich. Ernesto, what do you think we should do with > libpoe-api-peek-perl ? Should webgui get removed as well? webgui has never been part of any stable release. WebGUI's upstream has been very slow for the last year or so, with only a couple of releases. The latest release, about a month and a half ago, still lists POE::Component::IKC as a dependency and it's used for a core workflow component -- it requires 0.2001 I've written upstream about their thoughts on the issue. I've no particular opinion on what to do with libpoe-api-peek-perl -- I certainly don't know how to fix it. If removal is the sane choice, go ahead and remove both. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 0064 ADF5 EB5C DE16 99C1 6C56 F2A3 86B5 A757 E5A1
Bug#733961: Migration to yui3
On Thu, 2014-01-02 at 19:06 +0100, Moritz Muehlenhoff wrote: [...] Please migrate from src:yui to src:yui3. src:yui is abandoned, see http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=730104 Hi Moritz. Thanks for the report. WebGUI uses YUI 2.9 and since YUI 3 is not backwards compatible with YUI 2.x, I'll have to wait for upstream to transition. There hasn't been upgrades to WebGUI 7.10 for a while now, so I'm not sure when will that happen, if it happens at all. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 0064 ADF5 EB5C DE16 99C1 6C56 F2A3 86B5 A757 E5A1 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#732913: Add MariaDB as an alternative dependency
Hi Otto On Sun, 2013-12-22 at 19:58 +0200, Otto Kekäläinen wrote: Package: webgui Severity: wishlist MariaDB is an drop in replacement for MySQL. As MariaDB has just landed in Debian unstable it would be a good time to include it in the dependencies as an alternative to MySQL. Please change in the debian/control any occurences of mysql-server and mysql-client to mariadb-server | mysql-server and mariadb-client | mysql-client. [...] This is a very quick and safe change to do, and there is no urgency to upload the package just for this small thing. Just update the dependency (or suggest or recommends) lines in the debian/control file and let the change propagate when there is something else to push too. I've already made the change and committed to collab-maint's SVN repo. The package is still in an UNRELEASED state since there has been no new upstream release for a while. I pinged upstream today to see if they expect to make a release in the near future. Regards. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 0064 ADF5 EB5C DE16 99C1 6C56 F2A3 86B5 A757 E5A1 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#710558: closed by Christoph Berg m...@debian.org (Bug#710558: fixed in pgpool2 3.3.0-1)
On Thu, 2013-08-15 at 10:16 +0200, Christoph Berg wrote: Re: Ernesto Hernández-Novich 2013-08-14 1376489094.7.3.ca...@trillian.itverx.com.ve We've filed the bug upstream http://www.pgpool.net/mantisbt/view.php?id=60 We haven't tried newer versions of pgpool-II, as we must keep on using pgpool-II 3.1.3 as shipped by Debian Stable for ease of operation and deployment. Since newer 3.1.x releases only add bug-fixes, we are hoping either for a future package update (proposed update) on Wheezy as soon as we have resolution upstream, or a patched 3.1.3 version. Can you propose and update for Wheezy including the latest 3.1.8? It only provides bug-fixes for 3.1.3, without API or functionality changes. Hi, Hi Christoph I'm not sure I will have time to take care of this - as an alternative, there are wheezy packages for 3.3.0 on apt.postgresql.org. I'm aware of apt.postgresql.org and I've used them quite often, however in this particular case it doesn't help due to non-technical reasons (not my deployment, not my rules, sadly). But that's my problem, not yours. Which patches exactly from 3.1.8 would you want to have applied to 3.1.5? At least the one that fixes the problem noted in this bug. I believe pushing 3.1.8 as an upgrade would be way better. I don't know what exactly does a proposed update for stable entail, but I've seen it happen regularly with PostgreSQL packages: it's not just patches, it's the latest upstream that's only bug-fixes. If you need help packaging, I can lend a hand with that: I could build the 3.1.8 package and you propose the update. Thanks for your time. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#719719: ITP: bind10 -- ISC's Internet Domain Name Server
On Wed, 2013-08-14 at 12:22 -0600, LaMont Jones wrote: On Wed, Aug 14, 2013 at 09:47:14AM -0430, Ernesto Hernández-Novich wrote: * Package name: bind10 Version : 1.1.0 Upstream Author : ISC Software Distribution pack...@isc.org * URL : http://bind10.isc.org * License : BSD Programming Lang: C++, Python Description : ISC's Internet Domain Name Server BIND 10 is the next generation of BIND, the most widely-used name server software on the Internet, and is supported by the Internet Software Consortium, www.isc.org. I had meant to get bind10 uploaded about the time it hit RC, but clearly this has not happened. Since it may take me a few weeks to get a package uploaded, I would certainly welcome some team effort on this. I'll be working more on it this weekend, I expect. Hi Lamont. I've actually been working with ISC's people on getting bind10 packaged for Debian, so it's not a start from scratch thing. I did review what was being done with bind9 and used most of the ideas with bind10. BIND10 is quite different to BIND9 and I had (and have) direct contact with ISC's development team to make sure BIND10 is straightforward to package without having to resort to patching. It still isn't as easy as we would like, but it's getting there. I've used git-buildpackage for my work, mostly on a local repository. I will have a mostly-working 1.1.0 package over the weekend (albeit not lintian clean since the hardening stuff is not finished) so we could proceed from that. I'd like to help with BIND packaging in general. Regards, -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#710558: closed by Christoph Berg m...@debian.org (Bug#710558: fixed in pgpool2 3.3.0-1)
On Wed, 2013-08-14 at 09:06 +, Debian Bug Tracking System wrote: This is an automatic notification regarding your Bug report which was filed against the pgpool2 package: #710558: pgpool2: Exceptions on COMMIT cause abnormal pgpool process termination and disconnects It has been closed by Christoph Berg m...@debian.org. Their explanation is attached below along with your original report. If this explanation is unsatisfactory and you have not received a better one in a separate message then please contact Christoph Berg m...@debian.org by replying to this email. [...] email message attachment Source: pgpool2 Source-Version: 3.3.0-1 We believe that the bug you reported is fixed in the latest version of pgpool2, which is due to be installed in the Debian FTP archive. Thanks! [...] We've filed the bug upstream http://www.pgpool.net/mantisbt/view.php?id=60 We haven't tried newer versions of pgpool-II, as we must keep on using pgpool-II 3.1.3 as shipped by Debian Stable for ease of operation and deployment. Since newer 3.1.x releases only add bug-fixes, we are hoping either for a future package update (proposed update) on Wheezy as soon as we have resolution upstream, or a patched 3.1.3 version. Can you propose and update for Wheezy including the latest 3.1.8? It only provides bug-fixes for 3.1.3, without API or functionality changes. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#710558: Latest release of PgPool-II includes patch
PgPool-II 3.1.8 released on 2013-07-10 [1] includes a more general fix for this problem. It also provides several bugfixes without adding new functionality. Perhaps it's better to propose updating pgpool from 3.1.3 to 3.1.8 in Debian Wheezy (stable). [1] http://www.pgpool.net/docs/pgpool-II-3.1.8/NEWS.txt -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#710558: pgpool2: Exceptions on COMMIT cause abnormal pgpool process termination and disconnects
Upstream acknowledged and fixed this bug on all current PgPool versions,
applying a patch for a broader set of cases
http://www.pgpool.net/mantisbt/view.php?id=60
It would be great if the latest 3.1 release was proposed as a Wheezy
update, since it doesn't add new functionality, just bugfixes. In any
event, I backported the patch to 3.1.3 as shipped by Debian (attached),
and it works fine.
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
--- a/pool_process_query.c
+++ b/pool_process_query.c
@@ -77,6 +77,7 @@
static POOL_STATUS add_lock_target(POOL_CONNECTION *frontend, POOL_CONNECTION_POOL *backend, char* table);
static bool has_lock_target(POOL_CONNECTION *frontend, POOL_CONNECTION_POOL *backend, char* table, bool for_update);
static POOL_STATUS insert_oid_into_insert_lock(POOL_CONNECTION *frontend, POOL_CONNECTION_POOL *backend, char* table);
+static bool is_all_slaves_command_complete(unsigned char *kind_list, int num_backends, int master);
/* timeout sec for pool_check_fd */
static int timeoutsec;
@@ -3453,6 +3454,27 @@
}
/*
+ * returns true if all slaves status are 'C' (Command Complete)
+ */
+static bool is_all_slaves_command_complete(unsigned char *kind_list, int num_backends, int master)
+{
+ int i;
+ int ok = true;
+
+ for (i=0;inum_backends;i++)
+ {
+ if (i == master || kind_list[i] == 0)
+ continue;
+ if (kind_list[i] != 'C')
+ {
+ ok = false;
+ break;
+ }
+ }
+ return ok;
+}
+
+/*
* read_kind_from_backend: read kind from backends.
* the frontend parameter is used to send kind mismatch error message to the frontend.
* the out parameter decided_kind is the packet kind decided by this function.
Bug#710558: pgpool2: Exceptions on COMMIT cause abnormal pgpool process termination and disconnects
Disregard the last patch. The complete working patch is attached.
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
--- a/pool_process_query.c
+++ b/pool_process_query.c
@@ -77,6 +77,7 @@
static POOL_STATUS add_lock_target(POOL_CONNECTION *frontend, POOL_CONNECTION_POOL *backend, char* table);
static bool has_lock_target(POOL_CONNECTION *frontend, POOL_CONNECTION_POOL *backend, char* table, bool for_update);
static POOL_STATUS insert_oid_into_insert_lock(POOL_CONNECTION *frontend, POOL_CONNECTION_POOL *backend, char* table);
+static bool is_all_slaves_command_complete(unsigned char *kind_list, int num_backends, int master);
/* timeout sec for pool_check_fd */
static int timeoutsec;
@@ -3453,6 +3454,27 @@
}
/*
+ * returns true if all slaves status are 'C' (Command Complete)
+ */
+static bool is_all_slaves_command_complete(unsigned char *kind_list, int num_backends, int master)
+{
+ int i;
+ int ok = true;
+
+ for (i=0;inum_backends;i++)
+ {
+ if (i == master || kind_list[i] == 0)
+ continue;
+ if (kind_list[i] != 'C')
+ {
+ ok = false;
+ break;
+ }
+ }
+ return ok;
+}
+
+/*
* read_kind_from_backend: read kind from backends.
* the frontend parameter is used to send kind mismatch error message to the frontend.
* the out parameter decided_kind is the packet kind decided by this function.
@@ -3563,7 +3585,23 @@
* not all backends agree with kind. We need to do decide by majority
*/
- if (max_count = NUM_BACKENDS / 2.0)
+/*
+ * However here is a special case. In master slave mode, if
+ * master gets an error at commit, while other slaves are
+ * normal at commit, we don't need to degenrate any backend
+ * because it is likely that the error was caused by a
+ * deferred trigger.
+ */
+if (MASTER_SLAVE query_context-parse_tree
+is_commit_query(query_context-parse_tree)
+kind_list[MASTER_NODE_ID] == 'E'
+is_all_slaves_command_complete(kind_list, NUM_BACKENDS, MASTER_NODE_ID))
+{
+*decided_kind = kind_list[MASTER_NODE_ID];
+pool_log(read_kind_from_backend: do not degenerate because it is likely caused by a delayed commit);
+return POOL_CONTINUE;
+}
+else if (max_count = NUM_BACKENDS / 2.0)
{
/* no one gets majority. We trust master node's kind */
trust_kind = kind_list[MASTER_NODE_ID];
Bug#681564: webgui: apache2 segfaults
On Wed, 2013-05-22 at 13:42 +1000, Dean Hamstead wrote:
according to the code of Data::Visitor the Data::Alias dependency is
entirely optional...
http://cpansearch.perl.org/src/DOY/Data-Visitor-0.28/lib/Data/Visitor.pm
---
use constant HAS_DATA_ALIAS = load_optional_class('Data::Alias');
---
I was able to apt-get remove libdata-alias-perl without dpkg complaining.
Now apache isnt crashing... and webgui seems to be working!
Yes it works, and it passes all regression tests provided with WebGUI.
That's the workaround others found and I confirmed here
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=681564#80
It seems to be an interaction between the latest Data::Alias and
mod_perl. However, unless we come up with a simple Data::Alias using
mod_perl brokenness test, we can't reassing this bug to Data::Alias
(where I think the bug is FWIW).
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#681564: webgui: apache2 segfaults
On Thu, 2013-05-16 at 08:52 +, Dean Hamstead wrote: Package: webgui Version: 7.10.28-1 Followup-For: Bug #681564 New debian install, same problem still... I have this wrapped up in an LXC which i am happy to grant you access to for any probing people may want to try... Thank you for your report. Turns out the problem is Data::Alias (*not* used by WebGUI at all) when pulled in by Data::Visitor (used indirectly by WebGUI via Net::Twitter). I haven't had the time to create a mod_perl runnable snippet using Data::Visitor that would crash Apache. If you can come up with one, then we could reassing this bug to the proper library. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#684290: Unable to use Apache2::SizeLimit
On Mon, 2013-05-06 at 09:01 -0430, Ernesto Hernández-Novich wrote:
After upgrading from Squeeze to Wheezy, I'm unable to use my mod_perl
based application again. Apache's error log shows
[Mon May 06 08:36:18 2013] [error] [client 192.168.0.202] failed to
resolve handler `Apache2::SizeLimit': You must install Linux::Pid for
Apache::SizeLimit to work on your platform.
at /usr/lib/perl5/Apache/SizeLimit/Core.pm line 128.\nBEGIN
failed--compilation aborted at /usr/lib/perl5/Apache/SizeLimit/Core.pm
line 171.\nCompilation failed in require
at /usr/lib/perl5/Apache2/SizeLimit.pm line 48.\nBEGIN
failed--compilation aborted at /usr/lib/perl5/Apache2/SizeLimit.pm line
48.\nCompilation failed in require at (eval 1489) line 2.\n
Application works fine after removing the reference to
Apache2::SizeLimit in Apache's configuration file and restarting.
Working without Apache2::SizeLimit makes it very hard to control the
resource consumption of the mod_perl application, so it's not an
acceptable workaround.
After installing Linux::Pid everything works.
Looking at /usr/lib/perl5/Apache/SizeLimit/Core.pm, there's a BEGIN
block (line 134) having
elsif ($Config{'osname'} eq 'linux') {
_load('Linux::Pid');
so it does use Linux::Pid. It also *tries* to use Linux::Smaps a couple
of lines after, but this time it's optional and doesn't fail if it isn't
there.
Neither Linux::Pid nor Linux::Smaps is available on Debian Wheezy nor
Sid, at the time of this writing. I believe Linux::Pid should've been a
part of Debian Wheezy for Apache2::SizeLimit to work properly.
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#684290: Unable to use Apache2::SizeLimit
After upgrading from Squeeze to Wheezy, I'm unable to use my mod_perl based application again. Apache's error log shows [Mon May 06 08:36:18 2013] [error] [client 192.168.0.202] failed to resolve handler `Apache2::SizeLimit': You must install Linux::Pid for Apache::SizeLimit to work on your platform. at /usr/lib/perl5/Apache/SizeLimit/Core.pm line 128.\nBEGIN failed--compilation aborted at /usr/lib/perl5/Apache/SizeLimit/Core.pm line 171.\nCompilation failed in require at /usr/lib/perl5/Apache2/SizeLimit.pm line 48.\nBEGIN failed--compilation aborted at /usr/lib/perl5/Apache2/SizeLimit.pm line 48.\nCompilation failed in require at (eval 1489) line 2.\n Application works fine after removing the reference to Apache2::SizeLimit in Apache's configuration file and restarting. Working without Apache2::SizeLimit makes it very hard to control the resource consumption of the mod_perl application, so it's not an acceptable workaround. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#703696: Patch to ensure /var/run/postgresql is present
Tags: patch
The following patch to pgpool2's init script makes
sure /var/run/postgresql exists and sets proper permissions so it works
with and without PostgreSQL coexisting on the same machine.
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
diff -Naur pgpool2-3.1.3.orig/debian/pgpool2.init.d pgpool2-3.1.3/debian/pgpool2.init.d
--- pgpool2-3.1.3.orig/debian/pgpool2.init.d 2012-10-12 04:49:23.0 -0430
+++ pgpool2-3.1.3/debian/pgpool2.init.d 2013-04-09 19:16:46.698088278 -0430
@@ -38,6 +38,11 @@
d_start() {
+# dirs under /var/run can go away on reboots
+mkdir -p /var/run/postgresql
+chmod u+rwx,g+rwxs,o+rw /var/run/postgresql/# 2775
+chown postgres:postgres /var/run/postgresql/
+
if is_running; then
:
else
Bug#703696: pgpool2: Startup script assumes /var/run/postgresql exists
Debian Policy (9.1.4) states that /run is cleared on boot, so packages must not assume that any files or directories under /run exist unless the package has arranged to create those files or directories since the last reboot. Even more so, pgpool2's default configuration sets it's socket file to exist under /var/run/postgresql. pgpool2's init script does not set up this directory, so even if one were to create it manually, the problem persists after a reboot. When pgpool2 coexists with PostgreSQL, the latter makes sure that directory exists, so the problem doesn't show. But it will happen when using pgpool2 by itself in a machine without PostgreSQL. Since it is a violation of Debian Policy and makes the package unusable when installed by itself, I believe the severity of this bug should be increased to 'serious' -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#681564: fix
On Fri, 2013-02-08 at 11:12 +0100, Arne Wichmann wrote: begin quotation from Marek Andricik (in 20130123090843.ge3...@mail.vychod.net): Removing the libdata-alias-perl package helps. Apache and WebGUI works again. Tested on fresh Sid and Wheezy. No other package depended on libdata-alias-perl. I would not call this solution but rather clue for someone more experienced in the matter who can really find the cause and possibly fix it. Thanks, this workaround helped for the time being. I can replicate the problem. When using webgui 7.10.27 on Squeeze *with* libdata-alias-perl 1.07 the problem doesn't show up. But when using WebGUI 7.10 on Wheezy/Sid *with* libdata-alias-perl 1.16 the problem shows up. WebGUI does not use Data::Alias, so this must be an interaction between Data::Alias and mod_perl or another Perl module. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#681564: fix
On Fri, 2013-02-08 at 10:51 -0430, Ernesto Hernández-Novich wrote: [...] When using webgui 7.10.27 on Squeeze *with* libdata-alias-perl 1.07 the problem doesn't show up. But when using WebGUI 7.10 on Wheezy/Sid *with* libdata-alias-perl 1.16 the problem shows up. WebGUI does not use Data::Alias, so this must be an interaction between Data::Alias and mod_perl or another Perl module. Found the reason Data::Alias is being pulled in. * WebGUI uses Net::Twitter for authentication. * Net::Twitter uses Data::Visitor. * Data::Visitor Recommends Data::Alias and uses it when available. I believe this is definitely not a WebGUI bug, but instead a Data::Alias bug when interacting with mod_perl. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#681564: fix
On Wed, 2013-01-23 at 10:08 +0100, Marek Andricik wrote: Removing the libdata-alias-perl package helps. Apache and WebGUI works again. Tested on fresh Sid and Wheezy. No other package depended on libdata-alias-perl. What version of libdata-alias-perl were you using at the time? I have 1.16-1 installed and can't reproduce the problem. I would not call this solution but rather clue for someone more experienced in the matter who can really find the cause and possibly fix it. Older versions of Data::Alias were not compatible with Perl 5.12 onwards. and that was fixed in 1.16, FWIW. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#691087: webgui: deprecated syntax in create.sql.gz
On Sun, 2012-10-21 at 10:13 +, Dean Hamstead wrote: [...] the create.sql.gz file contains the TYPE=MyISAM syntax, but this is deprecated and should be ENGINE=MyISAM TYPE= was completely removed in mysql 5.5 so the sql file is broken see also http://dev.mysql.com/doc/refman/5.1/en/create-table.html Thanks for the report. I forwarded the bug upstream since create.sql is generated automatically as part of upstream's release workflow. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#681564: latest webgui release available
On Wed, 2012-08-15 at 23:45 +1000, Dean Hamstead wrote: commenting out Apache2::Cookie in preload.perl had no effect :( I haven't been able to reproduce your problem. I've tried with Wheezy and Sid, both on i386 and amd64, but it always works as expected. The latest release of WebGUI (7.10.26) was packaged and uploaded a couple of days ago. Maybe you'd like to try with it, perhaps with a fresh installation, and see if it fixes your problem. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#690486: twidge: gives a user error (Bad response: 404)
Greetings. I hade this same problem. I simply ran twidge setup and re-registered the application with Twitter, and everything started working normally again. -- Ernesto Hernández-Novich - @iamemhn - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#681564: Info received (Bug#681564: Info received (Bug#681564: Acknowledgement (webgui: apache2 segfaults)))
On Mon, 2012-08-13 at 21:32 +1000, Dean Hamstead wrote: [...] Apache2/Cookie.pm APR/Request/Cookie.pm root@webgui:~# though, if i just drop all the webgui loading and just run a perl script that loads Apache2::Cookie, i dont see a segfault. if i load them early in perload.perl - i still see the segfault. I'd like to start by pointing out that I haven't tried to run WebGUI with Perl 5.14 yet, so I thank you for your debugging efforts. Please try removing or commenting out the require Apache2::Cookie; line from /etc/webgui/preload.perl and restart Apache. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#680734: webgui: warnings in error.log
On Sun, 2012-07-08 at 20:22 +1000, Dean Hamstead wrote: Package: webgui Version: 7.10.24-1 Severity: minor As per the subject Use of qw(...) as parentheses is deprecated at /usr/share/webgui/lib/WebGUI/Asset/Wobject/Survey.pm line 1402. Use of qw(...) as parentheses is deprecated at /usr/share/webgui/lib/WebGUI/Asset/Wobject/Survey.pm line 2188. Very minor code syntax fixes Thanks for the report. This bug has been fixed upstream already. I'm waiting for their 7.10.25 release to package and upload. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#653099: ITP: liburi-encode-perl -- Perl module to encode and decode strings to URIs
On Fri, 2011-12-23 at 12:10 -0800, Don Armstrong wrote: On Fri, 23 Dec 2011, Ernesto Hernández-Novich wrote: * Package name: liburi-encode-perl Version : 0.04 Upstream Author : Mithun Ayachit mit...@cpan.org * URL : http://search.cpan.org/dist/URI-Encode/ * License : Artistic Programming Lang: Perl Description : Perl module to encode and decode strings to URIs URI::Encode provides an easy method to encode strings (mainly URLs) into a format which can be pasted into a plain text, so that those links are 'clickable' by the person reading it. If you are looking for speed and want to encode reserved characters, use URI::Escape::XS Why is anyone using URI::Encode instead of URI::Escape, which handles all of this, is faster to boot, and is in core? If you actually wanted its main feature, you'd just use: use URI::Escape qw(uri_escape_utf8); uri_escape_utf8($string,^A-Za-z0-9\-\._~:\/\?\#\[\]\@\!\$\\'\(\)\*\+\,\;\=); I agree. I've packaged URI::Encode because it's a dependency for a larger module (Facebook::Graph). I've already opened a bug upstream suggesting that URI::Escape be used instead of URI::Encode. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#610434: Has anyone attempted a tomcat7 package?
On Fri, 2011-05-13 at 10:25 +0200, James Page wrote: I'll make time next week and get it uploaded to subversion/git (anyone have any opinion on which - I personally prefer git). Debian's Java Packaging effort already uses SVN http://svn.debian.org/wsvn/pkg-java/ The tomcat6 effort is already there, so is natural for this tomcat7 effort to be there too, considering this is a Debian Bug being addressed. -- Ernesto Hernández-Novich - @iamemhn - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#610434: Has anyone attempted a tomcat7 package?
On Fri, 2011-05-13 at 08:34 -0430, Miguel Landaeta wrote: I think there is no problem using git since Debian Java team already has many repositories available at git://git.debian.org/pkg-java. Fair enough. I personally prefer to use git, if everyone who is interested on maintaining tomcat7 agrees on using git, I can create repository for it soon. I won't be maintaining it, so feel free to use what you prefer. Thanks! -- Ernesto Hernández-Novich - @iamemhn - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#610434: Has anyone attempted a tomcat7 package?
Well, I managed to get the thing to work (or at least I think so). Attached you'll find the diff that works against Tomcat7 7.0.12. I started with a fresh changelog that you should feel free to change in whatever way you see fit. I refreshed the copyright file into DEP parsable format and did a quick check on copyright dates while adding my own stanza. Finally, I found out what were the missing classes that prevented tomcat from starting and changed debian/rules to do the needed dh_installjars. I'd really appreciate if you could do a quick review and test the package. My users will start doing so probably this afternoon my time. -- Ernesto Hernández-Novich - @iamemhn - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 tomcat7_7.0.12-1.debian.tar.gz Description: application/compressed-tar
Bug#610434: Has anyone attempted a tomcat7 package?
On Thu, 2011-05-05 at 13:10 +0100, James Page wrote: I'm hoping to start packaging tomcat7 in the next few weeks (indeed I'm running a session on it at the Ubuntu Developer Summit next week if you want to tune in - see [0]). If you can wait a bit it would be great to collaborate on this work. I know enough Java to turn away from it, so I'll pass. I'm building this package at a customer's request; they're too stubborn to realize that using tomcat6 is enough for their purposes. So you can count on my Debian packaging abilities, including dh8 if needed (I've worked with the Debian Perl and Haskell Groups, and also maintain webgui). Looking at this issue specifically; this class should be contained in tomcat-util.jar that the build process produces. It looks like the jar's produced by the tomcat build has changed quite a bit since tomcat6 so the packaging will need a thorough review to ensure that its picking up all the right bits and pieces. My packaging effort started by reviewing the tomcat6 package, renaming what had to be renamed and fixing some of the jar names that, indeed, have been renamed. I'm attaching what I've done so far. The changelog is minimal and the copyright file hasn't been updated, and I dropped a lot of people from Uploaders (this was supposed to be an internal package in the meantime). I also dropped some of the CVE-related patches and refreshed the rest; I'm not sure if the patch for build.xml is useful, but I've kept it on account of ignorance. The package builds and is lintian clean, except for not closing a bug. From memory a standard tomcat install relies on jar files being installed to the correct locations in the CATALINA_BASE directory so you might want to start looking there. That's what I'll attempt later today (around 17:00 UTC or later). I generally lurk on the debian-java IRC channel if you want to discuss in more detail. Good to know. -- Ernesto Hernández-Novich - @iamemhn - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 tomcat7_7.0.12-1.debian.tar.gz Description: application/compressed-tar
Bug#620433: correction
On Sat, 2011-04-02 at 20:21 +0100, Nicholas Bamber wrote: Ernesto, It's in the queue now and Gregor has seen enough of it I am not expecting any issues. Anyway I am sure it's okay as long as it does not make it into tetsing and you could always raise the severity of the bug. WebGUI the application does not use HTML::Form. A few regression tests do use HTML::Form. This tests are intended for people hacking on WebGUI and are never used while running the application. I've added the dependency as you suggested, but it will be included in the next release of the package. I've also contacted upstream, and their intention is to stop using HTML::Form altogether. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#620433: correction
On Fri, 2011-04-01 at 22:30 +0100, Nicholas Bamber wrote: Sorry that should be Depends: libhtml-form-perl | libwww-perl ( 6) The next upstream release (7.9.26) will probably take place by the end of next week. Upstream is pretty regular doing stable releases and they happen every other week. When is libwww-perl 6.01 being uploaded? I'd rather wait for the 7.9.26 release instead of having a 7.9.25-2 for a few days. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#595047: twidge: Doesn't shorten URLs sometimes
On Tue, 2010-08-31 at 15:35 -0500, John Goerzen wrote:
On 08/31/2010 03:01 PM, Ernesto Hernández-Novich wrote:
I agree with you. However, shortening _every_ URL _every_ time seems to
be the common practice for Twitter clients, and that's the behaviour
implied by the man page.
[...]
I would accept:
* A patch to the manpage to make this clear
* A patch that adds a setting for shorten all URLs, defaulting to
false, and documents it
Attached you'll find two patches against 1.0.6-1:
shortenall.patch - Patches the manpage to make clear how URL shortening
works, along with patches to the code. Basically, shortenurls states
whether or not URLs should be shortened, with the same semantics it has
already (length 140), whilst the new shortenall option shortens
everything everytime if set to yes. As you requested, shortenall
defaults to 'no'.
debian.patch - Improvements to the Debian packaging: install the
bash-completion in place; install the needed doc-base fragment for the
PDF manual (this gets rid of one lintian warning); use dh_install
instead of manually copying in debian/rules; use dh_auto_clean instead
of manually invoking make; use dh_prep instead of dh_clean. These
changes required bumping the versioned B-D for debhelper to 7.
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
diff -Naur twidge-1.0.6/Commands/Update.hs twidge-1.0.6+emhn/Commands/Update.hs
--- twidge-1.0.6/Commands/Update.hs 2010-06-03 08:09:19.0 -0430
+++ twidge-1.0.6+emhn/Commands/Update.hs 2010-11-06 22:39:57.482906834 -0430
@@ -103,9 +103,12 @@
procStatus cp section status =
do poststatus - case get cp section shortenurls of
- Right True | length status 140
- - shortenUrls cp status
- _ - return status
+ Right True - case get cp section shortenall of
+ Right True - shortenUrls cp status
+ _ | length status 140
+ - shortenUrls cp status
+ _ - return status
+ _ - return status
when (length poststatus 140)
(permFail $ Your status update was ++
show (length poststatus) ++
diff -Naur twidge-1.0.6/Config.hs twidge-1.0.6+emhn/Config.hs
--- twidge-1.0.6/Config.hs 2010-09-01 11:30:08.0 -0430
+++ twidge-1.0.6+emhn/Config.hs 2010-11-06 22:02:15.102889901 -0430
@@ -47,6 +47,7 @@
cp - set cp DEFAULT oauthauthorize %(serverbase)s/oauth/authorize
cp - set cp DEFAULT sendmail /usr/sbin/sendmail
cp - set cp DEFAULT shortenurls yes
+ cp - set cp DEFAULT shortenall no
return cp
startingcp = emptyCP {accessfunc = interpolatingAccess 10}
diff -Naur twidge-1.0.6/doc/twidge-manpage.sgml twidge-1.0.6+emhn/doc/twidge-manpage.sgml
--- twidge-1.0.6/doc/twidge-manpage.sgml 2010-10-25 20:33:26.0 -0430
+++ twidge-1.0.6+emhn/doc/twidge-manpage.sgml 2010-11-06 22:50:03.563888668 -0430
@@ -559,9 +559,20 @@
/para
para
By default, Twidge will attempt to shorten URLs in your
-updates via the is.gd service. You can disable this by
-setting shortenurls = no in the [DEFAULT] or [dmsend] section of your
-configuration file.
+updates via an URL shortening service, but only if your
+update's length exceeds 140 characters. This means long
+URLs shorter than 140 characters WILL NOT be shortened.
+ /para
+ para
+If you want all your URLs to be shortened all the time
+regardless of their length, then set shortenall = yes
+in the [DEFAULT] or [dmsend] section of your configuration
+file.
+ /para
+ para
+You can disable URL shortening altogether by setting
+shortenurls = no in the [DEFAULT] or [dmsend] section of
+your configuration file.
/para
/refsect2
@@ -626,9 +637,20 @@
/para
para
By default, Twidge will attempt to shorten URLs in your
-updates via the TinyURL.com service. You can disable this by
-setting shortenurls = no in the [DEFAULT] or [update] section of your
-configuration file.
+updates via an URL shortening service, but only if your
+update's length exceeds 140 characters. This means long
+URLs shorter than 140 characters WILL NOT be shortened.
+ /para
+ para
+If you want all your URLs to be shortened all the time
+regardless of their length, then set shortenall = yes
+in the [DEFAULT] or [dmsend] section of your
Bug#593698: #593698 - Status update ? [WebGUI]
On Fri, 2010-09-17 at 14:18 +0200, Didier 'OdyX' Raboud wrote: Could you please communicate trough this bugreport what is your current plan for the resolution of this RC bug ? The bug has been solved in the 7.9.14-1 version. I consider 7.8.24 not to be suitable for Squeeze as per the reasons I gave on my message to Debian Release so I'm not backporting it. I am aware of your previous mail to debian-release [0] to which you haven't gotten an answer yet, but status updates (or +pending tags at least) to the RC bugs affecting WebGUI would be appreciated. Given that 7.9 works out of the box and is smoothly upgradeable from 7.8 (593885 has been solved in 7.9.14 too), I hope the Debian Release considers WebGUI's case and allows 7.9 to get into Squeeze as soon as 591208 gets closed by the 7.9.15 release, since 7.8 will be both old and potentially vulnerable to have in Squeeze. Besides, 591208 cannot be solved in 7.8 since the offending .swf files are essential for 7.8 but irrelevant in 7.9. In the meantime, you could upload the new webgui upstream release to experimental for example. I will probably upload 7.9.15 as soon at it's released upstream. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#595047: twidge: Doesn't shorten URLs sometimes
On Tue, 2010-08-31 at 12:32 -0500, Kumar Appaiah wrote: On Tue, Aug 31, 2010 at 12:51:49PM -0430, Ernesto Hernández-Novich wrote: The manpage states that Twidge will attempt to shorten URLs in your updates via the is.gd, bit.ly, j.ump or TinyURL services, if one sets the shortenurl option to 'yes' in ~/.twidgerc. However, if the update's text length is less than or equal to 140 characters, the URL shortening does not happen, because of the way the function procStatus (taken from Commands/Update.hs) has been written: [snip] But what if I don't want the URLs to be shortened unless the text is actually larger than 140 characters? I'd prefer complete URLs whenever possible, and shorten only if needed. Of course, this is _my_ preference, and need not match everyone else's. I agree with you. However, shortening _every_ URL _every_ time seems to be the common practice for Twitter clients, and that's the behaviour implied by the man page. If there where an additional preference for conditional URL shortening, I'd probably use it, if only to annoy those who think every URL should be shortened or ponies die, but that would require extra code plus changes in the manpage. It's easier to fix the code to match the manpage :-) -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#332595: This bug can be closed
Package: debhelper
Version: 8.0.0
Lines 108-112 of /usr/bin/dh_fixperms are
# ADA ali files should be mode 444 to avoid recompilation
if (-d $tmp/usr/lib/ada) {
complex_doit(find $tmp/usr/lib/ada -type f,
-name '*.ali' $find_options -print0,
2/dev/null | xargs -0r chmod uga-w);
}
so I propose closing this bug.
--
Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#576912: Patch to upgrade to 2.26
Attached you'll find a patch to refresh and upgrade liblog-dispatch-perl to the latest upstream release (2.26 as of this writing). The patch upgrades to DH7, source format 3.0, the parseable copyright file format and latest standards version. It builds cleanly on an up-to-date Sid pbuilder and lintian is happy about it. You only need to fix the copyright dates on the packaging and release the package through your sponsor. As I wrote in a previous message, if you'd like to hand over the package to the Debian Perl Group we'll be more than happy to adopt it, as long as you write an e-mail agreeing, either to this bug and/or debian-p...@lists.debian.org Thanks for your time! -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 liblog-dispatch-perl_2.26-1.debian.tar.gz Description: application/compressed-tar
Bug#576912: liblog-dispatch-perl: new upstream version available
I'm building a package for Log::Any::Adapter::Dispatch (liblog-any-adapter-dispatch-perl), but it requires at least Log::Dispatch 2.26. Log::Any::Adapter::Dispatch is a dependency of my webgui package, that I'm trying to transition from 7.8 to 7.9 just in time to Squeeze. I'd appreciate that the latest upstream of Log::Dispatch were available soon. If you're interested, you might want to move your package to the Debian Perl Packaging Group [1] where packages for Perl libraries are maintained by a team, with several tools available to make package updating easier. [1] http://pkg-perl.alioth.debian.org -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#413066: Packaging TinyMCE locales
I'd be interested in that too, in order to have the spanish locale for TinyMCE. I can help with packaging, hopefully based on Andreas work; a single source package (tinymce-locales or something like that) building several binary packages named tinymce-i18-LANG and we add locales as required by the user base. -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#585471: webgui: License of www/extras/yui/examples/treeview/assets/js/json.js is considered non-free
WebGUI's upstream [1] decided on removing the YUI examples from their tarball. The change will be effective starting on their 7.8.24 release, probably happening over the weekend. [1] http://www.webgui.org/use/bugs/tracker/11710 -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#585471: webgui: License of www/extras/yui/examples/treeview/assets/js/json.js is considered non-free
On Thu, 2010-06-10 at 20:20 +0200, Alexander Reichle-Schmehl wrote: Package: webgui Version: 7.8.19-1, 7.8.20-1, 7.8.21-1 Severity: serious Justification: contains non-free files Hi! The json license of www/extras/yui/examples/treeview/assets/js/json.js is considered non-free (The Software shall be used for Good, not Evil.) See the recent discussions on debian-legal. The file belongs to the redistribution of the YUI 2.8.1 examples included with the WebGUI sources. The webgui binary package does not use that redistribution of YUI 2.8.1, using libjs-yui instead, so the file can only be found in webgui source package. WebGUI's upstream has been notified even though it's not their fault that YUI 2.8.1 examples include that file. They are considering removal of the YUI 2.8.1 examples, since they're not used by WebGUI at all. I'll wait for the next release (next weekend) before doing anything else. I've also filed a ticket with YUI [1] that has been promptly addressed, although it's targeted for 2.9.0. [1] http://yuilibrary.com/projects/yui2/ticket/2529037 -- Prof. Ernesto Hernández-Novich - MYS-220C - @iamemhn Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#522790: libsvn-web-perl: svnweb-server crash when attepmt to browse repository
I had the same problem, but using SVN::Web through Apache2 and mod_perl to create a customized SVN browser. After a straight dist-upgrade from Debian Etch to Lenny, the Web SVN access stopped working and I started getting error messages like the ones mentioned in the original bug report [1]. I fixed it as suggested by Christian Hammers [2] and would also appreciate the upload of 0.53-3 from unstable to stable-proposed-updates in the near future, so it becomes part of the next Lenny update. [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=522790#5 [2] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=522790#15 -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#548544: libhtml-template-expr-perl: New upstream version available
Any progress in getting the latest upstream release packaged? The offer to take the package over to the Debian Perl Group still stands. -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#547564: webgui: Incorrect dependencies in init.d scripts
On Wed, 2009-09-30 at 22:11 +0200, Petter Reinholdtsen wrote: Hi. Any hope of having an upload with a fix for this in unstable soon? Please object if we should not NMU to fix it. I'm waiting for the latest libarchive-zip-perl to hit Sid to upload the latest 7.7.20 version of WebGUI, which includes your fix. -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#548963: libarchive-zip-perl: New upstream release
On Wed, 2009-09-30 at 08:50 +1000, Ben Burton wrote: Hi Ernesto, Regarding libarchive-zip-perl: If you feel you don't have the time or interest to keep up with libarchive-zip-perl, I can transfer it to the pkg-perl project myself. This is probably best.. alas my free time is not what it once was. :) I've already transfered the package to the Project's SVN. Would you like to be kept as Uploader? Thanks for looking after this, You're welcome! -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#521889: Regarding libimage-exiftool-perl in Debian
Hi. I'm the maintainer of the webgui [1] package for Debian. I'm attempting to upgrade to the latest stable release (7.7.x) but am unable to because it requires a version of Image::ExifTool newer than the one in Debian Sid. I noticed there are several bugs open [2], one of them (#521889) being the 'New version available', but have had no activity since posted. I work with the Debian Perl Packaging team [3], and if you feel you don't have the time or interest on keeping up with libimage-exiftool-perl, I can transfer it to the pkg-perl project where there are several maintainers that are willing to maintain it. Please let me know if you're fine with this. You can CC debian-p...@lists.debian.org so the rest of the group knows about the package transfer. Thank you! [1] http://www.webgui.org [2] http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libimage-exiftool-perl [3] http://pkg-perl.alioth.debian.org -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#528214: Offer to take over libxml-feedpp-perl
On Wed, 2009-08-26 at 15:57 -0430, Ernesto Hernández-Novich wrote: I filed this bug because WebGUI (already in Debian Sid) requires at least XML::FeedPP 0.37 to properly handle UTF-8 feeds. It's been a while since I filed the bug and would like to know if you're planning to upload a newer version soon. The latest version of XML::FeedPP (0.40) depends on XML::TreePP 0.38. I contacted David Moreno Garza (damog) yesterday, and he has already uploaded libxml-treepp-perl 0.39 to Sid so everything is set for the latest XML::FeedPP. -- Ernesto Hernández-Novich - Linux 2.6.28 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#528214: Offer to take over libxml-feedpp-perl
Hi. I filed this bug because WebGUI (already in Debian Sid) requires at least XML::FeedPP 0.37 to properly handle UTF-8 feeds. It's been a while since I filed the bug and would like to know if you're planning to upload a newer version soon. I would also like to suggest moving this package into Debian's Perl Group [1] SVN, so you end up having co-maintainers for free. If you're fine with this, just let me know. [1] http://pkg-perl.alioth.debian.org/ -- Ernesto Hernández-Novich - Linux 2.6.28 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#524136: Bug reported in CPAN with patch
http://rt.cpan.org/Public/Bug/Display.html?id=45067 -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#524136: libtimedate-perl: Formatting with %z gives wrong result for half-hour timezones
On Tue, 2009-04-14 at 22:37 -0430, Ernesto Hernández-Novich wrote: Package: libtimedate-perl Version: 1.1600-9 Severity: important Tags: patch Using Date::Format to format a date including the timezone as an offset (%z) does not work correctly if the offset is 30 minutes. I noticed the problem since I need VET (-4:30) and several mail headers started to be very wrong. A simple test shows #!/usr/bin/perl use Date::Format; print time2str(%Y-%m-%d %T %Z %z, $^T, +0400).\n; print time2str(%Y-%m-%d %T %Z %z, $^T, +0430).\n; print time2str(%Y-%m-%d %T %Z %z, $^T, +).\n; print time2str(%Y-%m-%d %T %Z %z, $^T, -0400).\n; print time2str(%Y-%m-%d %T %Z %z, $^T, -0430).\n; $ perl ~/z.pl 2009-04-15 06:58:32 +0400 +0400 2009-04-15 07:28:32 +0430 +041800 2009-04-15 02:58:32 + + 2009-04-14 22:58:32 -0400 -0400 2009-04-14 22:28:32 -0430 -041800 The problem is in function format_z in /usr/share/perl5/Date/Format.pm line 239. I've replaced it with sprintf(%+05d, int($o/3600)*100 + ($o-int($o/3600)*3600)/60 ); and it seems to work out ok. But after some thoughts and inputs from Gunnar Wolf on IRC, I think sprintf(%+03d%02d, int($o / 3600), int(abs($o) % 3600) / 60); would be better because is less of a change to the original line and its intentions are clearer. Besides, it also works for timezones with other weird offsets like +1245 (Catham Islands) +0545 (Kathmandu) and I even tried with things like -0417 Because I Can (tm). -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#470890: Please increase severity to grave
This bug should be marked grave because any application using CGI::Application::Plugins::ActionDispatch and having actions with capturing regexes (almost all :-) will be broken because of it. Line 150 is a debugging line that uses Data::Dumper to dump the current application status to STDOUT. This line is in the normal flow of processing therefore will be reached in every execution, thus breaking most applications. Depending on how the programmer set up the CGI::Application module, it may break the application with anything ranging from a webserver's 500 error to unexpected output on the generated page. The bug can be fixed either by removing line 150 or upgrading to a later upstream release. -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#494164: New upstream release
Package: libdatetime-locale-perl Version: 1:0.35-1+1.5.0 Severity: wishlist There's a new upstream version of DateTime::Locale (0.41). The Debian Perl Group [1] is currently trying to update a related package, libdatetime-format-strptime-perl, to its latest upstream release, but it depends on having 0.41 of libdatetime-locale-perl available. [1] http://pkg-perl.alioth.debian.org/ -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#479658: ikiwiki: Your session has expired after Saving any edited page
On Fri, 2008-05-30 at 23:39 -0400, Joey Hess wrote: Ernesto Hernández-Novich wrote: Does not work if I set anonok = 0 httpauth = 1 Ok, I've finally managed to reproduce your problem. And it is a cookie problem. Since httpauth avoids the need to log in, the login cookie is never set, so the session id varies every time. The recent security fix exposed this latent bug, which was fixed a long time ago in newer versions of ikiwiki. It may take a while for the security team to get a fixed version, in the meantime, I've attached a patch. I've just installed the patch from Security and everything seems to be working fine in the three sites I run with ikiwiki, so I believe the problem has been solved. Thank you for your time. -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#479658: ikiwiki: Your session has expired after Saving any edited page
On Tue, 2008-05-06 at 17:46 -0400, Joey Hess wrote: Ernesto Hernández-Novich wrote: If you're using http auth, my guess is that is the problem somehow. Indeed. If I downgrade to 1.33.3 everything works fine again, so I figure the problem has to do with the newly added code for session management in 1.33.4 and/or 1.33.5 I assume so, but I don't really see how. Does it work if you disable httpauth? Works fine if I set anonok = 1 httpauth = 0 Works fine if I set anonok = 0 httpauth = 0 Does not work if I set anonok = 0 httpauth = 1 -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#479658: ikiwiki: Your session has expired after Saving any edited page
On Mon, 2008-05-05 at 22:20 -0400, Joey Hess wrote: I don't have a stable system handy, but I installed the old ikiwiki on unstable and am editing files via the web w/o problems. So this is looking like a local configuration problem. Perhaps with the editpage.tmpl. Perhaps you should try using the standard one and not your modified version. I did a fresh install of Debian Etch and brought it up to date with the latest security patches. Installed ikiwiki (1.33.5) apache2 (2.2.3-4+etch4) libapache2-mod-perl2 (2.0.2-2.4) subversion (1.4.2dfsg1-2) and then # mkdir /var/www/wiki # mkdir /home/wiki # cd /home/wiki # cp /usr/share/doc/ikiwiki/html/ikiwiki.setup . # vi ikiwiki.setup Set srcdir = '/home/wiki' url = 'http://wiki/wiki' cgiurl = 'http://wiki/ikiwiki.cgi' svnrepo = '/var/lib/svn/wiki' svnpath = 'trunk' Enabled BOTH wrappers as they are in the sample setup, just changing the appropiate paths, enabled httpauth and verbose logging, but left everything else untouched (hence using the default templates as packaged in Debian Etch latest security upgrade). # ikiwiki --seutp /home/wiki/ikiwiki.setup The content gets generated properly. Using good old vi to change stuff, svn commit, the hook runs and the content gets updated fine. Enabled Apache's basic authentication and tried to use ikiwiki.cgi to edit. It shows the edit window, allows me to edit and preview as many times as I want, but as soon as I press Save, it goes to a blank page (no header, no content) and I get [Tue May 06 11:06:23 2008] [error] [client 172.18.100.200] Undefined subroutine IkiWiki::gettext called at /usr/share/perl5/IkiWiki/CGI.pm line 615., referer: http://wiki/wiki/ikiwiki.cgi?page=indexdo=edit [Tue May 06 11:06:23 2008] [error] [client 172.18.100.200] Premature end of script headers: ikiwiki.cgi, referer: http://wiki/wiki/ikiwiki.cgi?page=indexdo=edit in the Apache logs. There is a gettext() routine being called that it's nowhere to be found in IkiWiki's code (missing module or dependency?). Checking line 615 on the mentioned file shows that is the one that says 'Your session has expired', and the default templates I'm using do have the FIELD-SID var in there. I repeated the debugging technique of printing SID and Session ID, and again their values are different. I don't think it's a local configuration problem. -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#479658: ikiwiki: Your session has expired after Saving any edited page
On Tue, 2008-05-06 at 13:32 -0400, Joey Hess wrote: Ernesto Hernández-Novich wrote: ... You don't mention creating the svn repo and checking the source in, but I assume you did. Sure thing. I've replicated this setup according to your instructions in a stable chroot, and editing works fine. Only potential difference I can imagine is that you said you enabled httpauth. I'm registering and logging in the regular way, since I don't have http authentication set up (but I did enable it, for what it's worth). (Also, you installed libapache2-mod-perl2, so I did also, but AFAIK ikiwiki cannot use it.) If you're using http auth, my guess is that is the problem somehow. Indeed. If I downgrade to 1.33.3 everything works fine again, so I figure the problem has to do with the newly added code for session management in 1.33.4 and/or 1.33.5 Regarding mod_perl, it could be used to enhance ikiwiki's performance, but I haven't configured it to do so. I mentioned just because I always install it. I purged it and tried without, but got the same erroneous behavior. Enabled Apache's basic authentication and tried to use ikiwiki.cgi to edit. It shows the edit window, allows me to edit and preview as many times as I want, but as soon as I press Save, it goes to a blank page (no header, no content) and I get [Tue May 06 11:06:23 2008] [error] [client 172.18.100.200] Undefined subroutine IkiWiki::gettext called at /usr/share/perl5/IkiWiki/CGI.pm line 615., referer: http://wiki/wiki/ikiwiki.cgi?page=indexdo=edit [Tue May 06 11:06:23 2008] [error] [client 172.18.100.200] Premature end of script headers: ikiwiki.cgi, referer: http://wiki/wiki/ikiwiki.cgi?page=indexdo=edit in the Apache logs. There is a gettext() routine being called that it's nowhere to be found in IkiWiki's code (missing module or dependency?). This is in fact a bug introduced in 1.33.5, but it is unrelated to whatever is making editing fail in the first place. Agreed. I was just mentioning it for completeness. I repeated the debugging technique of printing SID and Session ID, and again their values are different. I edited /usr/share/perl5/IkiWiki/CGI.pm and added this line after line 613: print STDERR Content-Type: text/html\n\nDebug: sid = $sid ; session id: .$session-id.\n; Now if I save a page, it prints in /var/log/apache2/error.log: Debug: sid = 1b22ac10538cbdb4fc01ec29d87e207d ; session id: 1b22ac10538cbdb4fc01ec29d87e207d Same ids.. Same change here, different IDs... [Tue May 06 13:59:55 2008] [error] [client 172.18.100.200] Content-Type: text/html, referer: http://wiki/wiki/ikiwiki.cgi?page=indexdo=edit [Tue May 06 13:59:55 2008] [error] [client 172.18.100.200] , referer: http://wiki/wiki/ikiwiki.cgi?page=indexdo=edit [Tue May 06 13:59:55 2008] [error] [client 172.18.100.200] Debug: sid = 12274c12ec8deb148d8629ffa6d77611 ; session id: ece12dbcb60ec4fba25ab57603d64a18, referer: http://vigilante.unesr.edu.ve/wiki/ikiwiki.cgi?page=indexdo=edit -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#479658: ikiwiki: Your session has expired after Saving any edited page
On Mon, 2008-05-05 at 20:49 -0400, Joey Hess wrote: Do you have cookies enabled in your browser? If you don't, you'll get a new session id, which won't match the old, and could possibly fail like this. Yes, I had cookies enabled on the browsers while testing. -- Prof. Ernesto Hernández-Novich - MYS-220C Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#422481: libemail-mime-modifier-perl: FTBFS: tests failed
On Wed, 2007-07-25 at 11:29 -0500, Gunnar Wolf wrote: Ernesto Hernández-Novich dijo [Tue, Jul 24, 2007 at 04:04:18PM -0400]: FWIW, this is fixed by upgrading to (at least) Email::MIME 1.858. However, in the meantime, the libemail-simple-perl upgrade to 2.002-1 has caused more test failures that are fixed upstream in Email::MIME::Modifier 1.442. I have both libemail-mime-perl 1.860 and libemail-mime-modifier-perl 1.442 packaged, just waiting for ANY sponsor to upload. ...In your site (http://itverx.com.ve/debian/pool/main/libe/) you have 1.441-2 only. Just uploaded 1.442-1 there. It actually needed a versioned Depends on libemail-mime-perl 1.860 in order to build correctly. Since libemail-mime-perl 1.860 is already in Sid and Lenny, there should be no problems uploading it. Thanks in advance. -- Ernesto Hernández-Novich - Linux 2.6.18 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#422481: libemail-mime-modifier-perl: FTBFS: tests failed
On Tue, 2007-07-24 at 22:16 +0300, Niko Tyni wrote: On Sun, May 06, 2007 at 12:23:23PM +0200, Lucas Nussbaum wrote: Package: libemail-mime-modifier-perl Version: 1.441-1 Severity: serious Justification: FTBFS on i386, very likely to fail everywhere else Usertags: grid5000 rebuild During a rebuild of all packages in sid, I discovered that your package failed to build on i386. Relevant parts: Failed Test Stat Wstat Total Fail Failed List of Failed --- t/parts.t 11 281622 11 50.00% 6 8-9 15-22 2 tests skipped. Failed 1/7 test scripts, 85.71% okay. 11/68 subtests failed, 83.82% okay. make[1]: *** [test_dynamic] Error 255 FWIW, this is fixed by upgrading to (at least) Email::MIME 1.858. However, in the meantime, the libemail-simple-perl upgrade to 2.002-1 has caused more test failures that are fixed upstream in Email::MIME::Modifier 1.442. I have both libemail-mime-perl 1.860 and libemail-mime-modifier-perl 1.442 packaged, just waiting for ANY sponsor to upload. -- Ernesto Hernández-Novich - Linux 2.6.18 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#422479: Processed: Re: Bug#422479: libemail-mime-perl: FTBFS: tests failed
On Wed, 2007-07-18 at 19:15 +, Debian Bug Tracking System wrote: Processing commands for [EMAIL PROTECTED]: tags 422479 patch fixed-upstream Bug#422479: libemail-mime-perl: FTBFS: tests failed Tags were: sid lenny Tags added: patch, fixed-upstream I've packaged 1.859 about a week ago. I'm waiting for any sponsor who can upload it. -- Ernesto Hernández-Novich - Linux 2.6.18 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#406998: Default configuration doesn't match amavisd-new-milter
Package: amavisd-new Version: 2.4.2-5 File /etc/amavis/conf.d/25-amavis_helpers has $unix_socketname = /var/run/amavis/amavisd.sock; but it should be $unix_socketname = /var/lib/amavis/amavisd.sock; since the latter pathname is hardcoded in /usr/sbin/amavis-milter as can be verified using 'strings'. This causes amavis-milter to be unable to communicate with amavisd, therefore any MTA using amavis-milter (sendmail in my case) will report a 451 error. I also needed to comment the _complete_ interface_policy block in order to get it to work properly. -- Ernesto Hernández-Novich - Linux 2.6.17 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#406586: GVIM should be opening files instead of URLs when invoked from Nautilus
Package: vim-gui-common Version: 7.0-122+1 The current /usr/share/applications/gvim.desktop has Exec=gvim -f %U meaning Nautilus will pass one o more URLs to gvim when invoking it. If you invoke GVIM from Nautilus on one or more filenames with UTF-8 encodings usin %U, the filenames will not be correctly handled and GVIM is not able to open them. For instance, I have a file named /home/emhn/Guías/ldap.txt (note the 'i acute', under en_US.UTF-8 in my case). If I use Nautilus to open this file using GVIM, it fails miserably arguing it cannot open /home/emhn/GuÃas/ldap.txt If I use %F instead of %U in the gvim.desktop file, it works fine opening multiple files. I believe this could also help bug #370082 -- Ernesto Hernández-Novich - Linux 2.6.17 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#403088: libemail-mime-modifier-perl: FTBFS: Can't locate object method _headers_as_string via package Email::MIME
On Thu, 2006-12-14 at 16:15 +0100, Lucas Nussbaum wrote: Package: libemail-mime-modifier-perl Version: 1.440-1 Severity: serious Justification: FTBFS on i386, very likely to fail everywhere else Usertags: grid5000 Hi, During a rebuild of all packages in etch, I discovered that your package failed to build on i386. Relevant parts: # Add commands to install the package into debian/ACKAGE_NAME here /usr/bin/make test make[1]: Entering directory `/build/root/libemail-mime-modifier-perl-1.440' PERL_DL_NONLAZY=1 /usr/bin/perl -MExtUtils::Command::MM -e test_harness(0, 'blib/lib', 'blib/ar ch') t/*.t t/content_id..Use of uninitialized value in pattern match (m//) at /usr/share/perl5/Email/MIME/C ontentType.pm line 24. Can't locate object method _headers_as_string via package Email::MIME at /usr/share/perl5/Email/ MIME.pm line 20. The error is actually in Email::MIME (1.851 in Etch, also maintained by me). The method as_string in Email::MIME is using it's superclass' (Email::Simple) _headers_as_string method which does not exist (it has been deprecated). Email::MIME 1.857 fixes this problem and I've already packaged it. I just need a sponsor to upload it. FWIW, I've also packaged Email::MIME::Modifier 1.441 -- Ernesto Hernández-Novich - Linux 2.6.17 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#401161: libgraphics-colornames-www-perl: description is lacking an end:
On Fri, 2006-12-01 at 11:33 +0100, Gerfried Fuchs wrote: Package: libgraphics-colornames-www-perl Version: 0.01-1 Severity: minor Hi! The long description of the package ends in a colon, which suggests that there should be something still coming up, but it isn't: Actually, it's just a typo. Fixed and waiting for my sponsor to upload. -- Ernesto Hernández-Novich - Linux 2.6.17 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#397189: New upstream release
Package: libpoe-component-client-http-perl
Version: 0.65-1.1
The latest upstream version is 0.79. I've packaged and tested id, .diff
attached.
--
Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die!
Geek by nature, Linux by choice, Debian of course.
If you can't aptitude it, it isn't useful or doesn't exist.
GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
--- libpoe-component-client-http-perl-0.79.orig/debian/changelog
+++ libpoe-component-client-http-perl-0.79/debian/changelog
@@ -0,0 +1,117 @@
+libpoe-component-client-http-perl (0.79-1) unstable; urgency=medium
+
+ * Non-maintainer upload.
+ * New upstream release.
+ * Updated Standards Version.
+
+ -- Ernesto Hernández-Novich [EMAIL PROTECTED] Sun, 05 Nov 2006 14:08:30
-0400
+
+libpoe-component-client-http-perl (0.65-1.1) unstable; urgency=medium
+
+ * Non-maintainer upload.
+ * Do not run tests as they require network access which is not
+acceptable. Upload based on patch by Thijs Kinkhorst, thanks.
+Closes: #393899
+
+ -- Michael Ablassmeier [EMAIL PROTECTED] Fri, 27 Oct 2006 09:14:54 +0200
+
+libpoe-component-client-http-perl (0.65-1) unstable; urgency=low
+
+ * New upstream release.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Sun, 2 Jan 2005 03:42:17 +1100
+
+libpoe-component-client-http-perl (0.64-2) unstable; urgency=low
+
+ * Add netbase to Build-Depends. (Closes: #265724)
+ * Add libscan to Makefile.PL to not include TLA files.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Mon, 16 Aug 2004 12:11:20 +1000
+
+libpoe-component-client-http-perl (0.64-1) unstable; urgency=low
+
+ * New upstream release.
+ * Remove obselete local-variables.
+ * Rejig the debian/rules file.
+ * Jump to debhelper version 4.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Fri, 6 Aug 2004 17:56:08 +1000
+
+libpoe-component-client-http-perl (0.52-1) unstable; urgency=low
+
+ * New upstream release.
+ * Change Section from interpreters to perl.
+ * Kick to Standards-Version 3.5.9.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Tue, 1 Apr 2003 13:38:48 +1000
+
+libpoe-component-client-http-perl (0.51-1) unstable; urgency=low
+
+ * New upstream release.
+ * Fix up package so that isn't built as Debian-native.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Fri, 27 Dec 2002 18:05:32 +1100
+
+libpoe-component-client-http-perl (0.48-1) unstable; urgency=low
+
+ * New upstream version. (Closes: #170787)
+ * Move from Build-Depends to Build-Depends-Indep.
+ * Bump Standards-Version to 3.5.8.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Wed, 27 Nov 2002 00:38:23 +1100
+
+libpoe-component-client-http-perl (0.46-3) unstable; urgency=low
+
+ * Fix up Build-Depends, since libpoe-perl, liburi-perl, libwww-perl,
+and libnet-ssleay-perl needed to build. (Closes: #162741)
+ * Also Depend on the same packages.
+ * Bump to Standards-Version: 3.5.7.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Sun, 29 Sep 2002 23:58:15 +1000
+
+libpoe-component-client-http-perl (0.46-2) unstable; urgency=low
+
+ * Finally change Maintainer address.
+ * Fix copyright. (Closes: #157641)
+
+ -- Steve Kowalik [EMAIL PROTECTED] Wed, 21 Aug 2002 23:44:42 +1000
+
+libpoe-component-client-http-perl (0.46-1) unstable; urgency=low
+
+ * New upstream release
+ * Build in build-indep. (Closes: #157503)
+
+ -- Steve Kowalik [EMAIL PROTECTED] Wed, 21 Aug 2002 23:29:43 +1000
+
+libpoe-component-client-http-perl (0.41-2) unstable; urgency=low
+
+ * Duh, depend on libpoe-perl. (Closes: #155765)
+
+ -- Steve Kowalik [EMAIL PROTECTED] Wed, 7 Aug 2002 22:25:24 +1000
+
+libpoe-component-client-http-perl (0.41-1) unstable; urgency=low
+
+ * New upstream release
+ * Guess which bonehead missed 5 upstream releases?
+
+ -- Steve Kowalik [EMAIL PROTECTED] Fri, 29 Mar 2002 19:04:52 +1100
+
+libpoe-component-client-http-perl (0.36-3) unstable; urgency=low
+
+ * Fix copyright.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Sat, 14 Jul 2001 17:40:42 +1000
+
+libpoe-component-client-http-perl (0.36-2) unstable; urgency=low
+
+ * Bugger. Missed a document entry.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Thu, 5 Jul 2001 22:24:06 +1000
+
+libpoe-component-client-http-perl (0.36-1) unstable; urgency=low
+
+ * Initial Release.
+
+ -- Steve Kowalik [EMAIL PROTECTED] Thu, 5 Jul 2001 22:16:57 +1000
+
+
--- libpoe-component-client-http-perl-0.79.orig/debian/compat
+++ libpoe-component-client-http-perl-0.79/debian/compat
@@ -0,0 +1 @@
+4
--- libpoe-component-client-http-perl-0.79.orig/debian/control
+++ libpoe-component-client-http-perl-0.79/debian/control
@@ -0,0 +1,20 @@
+Source: libpoe-component-client-http-perl
+Section: perl
+Priority: optional
+Build-Depends-Indep: debhelper (= 4.0.0), perl (= 5.6.0-17), libpoe-perl,
liburi-perl, libwww-perl, libnet-ssleay-perl, netbase
+Maintainer: Steve Kowalik [EMAIL PROTECTED]
+Standards-Version: 3.7.2
+
+Package: libpoe-component-client-http-perl
+Architecture: all
+Depends: ${perl:Depends}, libpoe-perl, liburi-perl, libwww-perl,
libnet
Bug#391264: New upstream version
Package: libtest-deep-perl Version: 0.093-1 The latest available version of Test::Deep is 0.096. -- Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#377522: fgfs-atlas closes X session
Before running Atlas, what's the output of $ xdpyinfo | grep -i glx You should have the GLX extension loaded in your X server in order to use Atlas. What video card are you using? -- Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#388839: dom0 crashes and reboots machine when domU is started.
Package: xen-linux-system-2.6.17-2-xen-686 Version: 2.6.17-9 I have a testing (Etch) system up to date as of today. I installed ii libc6-xen 2.3.6.ds1-4 ii linux-image-2.6.17-2-xen-686 2.6.17-9 ii linux-modules-2.6.17-2-xen-686 2.6.17-9 ii xen-docs-3.0-unstable 3.0-unstable+hg11292-2 ii xen-hypervisor-3.0-unstable-1-i386 3.0-unstable+hg11292-2 ii xen-ioemu-3.0-unstable 3.0-unstable+hg11292-2 ii xen-linux-system-2.6.17-2-xen-686 2.6.17-9 ii xen-tools 2.5-2 ii xen-utils-3.0-unstable-1 3.0-unstable+hg11292-2 ii xen-utils-common 3.0+hg11292-2 Hypervisor boots and loads dom0 fine. I had to add 'netloop' to /etc/modules in order for the bridge to be configured properly. After that, I built a couple of domU images, one with Sarge and the other with Etch. When either domU is started, the dom0 panics and reboots. If I remove the 'vif' from the domu.cfg, then the domU boots but without network interfaces. I'm using (network-script network-bridge) (vif-script vif-bridge) in xend-config.sxp, and kernel = '/boot/vmlinuz-2.6.17-2-xen-686' ramdisk = '/boot/initrd.img-2.6.17-2-xen-686' vif = [ 'ip=192.168.0.30' ] in domu.cfg; note that commenting out the last line allows the domU to boot, but without network interfaces. -- Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#384788: console-tools startup script could allow setting fore and background colors
Package: console-tools Version: 0.2.3dbs-64 Severity: wishlist The startup script /etc/init.d/console-screen.sh gets several environment variables from /etc/console-tools/config. Some of this environment variables are passed as parameters to setterm in order to configure the console's screensaver. It would be nice if a couple of variables were added to the script so that default fore and background colors could be set for consoles. -- Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#381287: libcgi-ajax-perl: New uplink version 0.697
Martin von Oertzen wrote: Package: libcgi-ajax-perl Version: 0.694-1 Severity: normal There is a new Version at CPAN: 0.697 (134) | bpederse | 2006-07-15 It has been packaged. Just waiting for my sponsor to upload them. -- Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#362648: Fix for bug #362648
I've managed to fix the bug by editing the desktop file /usr/share/applications/podbrowser.desktop attached. I've also added spanish, german and hungarian (magyar) translations (the latter two courtesy of Andrés Gazsó agazso at ius dot cc). The file must have UTF-8 encoding. -- Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3 [Desktop Entry] Version=1.0 Name=Pod Browser Name[de]=Pod Browser Name[en]=Pod Browser Name[es]=Explorador para POD Name[hu]=POD keresö Comment=Browse Perl documentation Comment[de]=Perl Dokumantation Browser Comment[en]=Browse Perl documentation Comment[es]=Explorar la documentación de Perl Comment[hu]=POD keresö Categories=GNOME;Application;Development; Encoding=UTF-8 Exec=podbrowser Icon=podbrowser.png Terminal=false Type=Application
Bug#362648: podbrowser doesn't show on Gnome menu
Package: podbrowser Version: 0.03-1 On Debian Sarge, podbrowser shows up under Gnome's Applications - Programming menu. After upgrading to Etch/Sid, it doesn't show anymore. It doesn't show inside Debians menu either. I've tried purging and reinstalling de package. I also tried changing /usr/share/applications/podbrowser.desktop to no avail. -- Ernesto Hernández-Novich - On Linux 2.6.16 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't aptitude it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#320993: Squirrelmail has improper dependencies
Package: squirrelmail Version: 2:1.4.4-6sarge1 Current dependencies are apache2 | httpd, libapache2-mod-php4 | php4 | php4-cgi, perl, squirrelmail-locales Using Debian Sarge with Security updates, this is incorrect at least in two ways: 1. If it depends on 'httpd' or 'apache2', then it should also be apache2 | apache | httpd 2. Should depend on libapache2-mod-php4 | libapache-mod-php4 | php4 | php4-cgi I noticed the problem when attempted installation of squirrelmail on a Debian Sarge system that had apache and libapache-mod-php4, and both apt-get and aptitude attempted installing apache2, libapache2-mod-php4 and some more stuff. I worked around it installing php4, given that apache provides 'httpd' so the first dependency was satisfied. I also think it should only recommend squirrelmail-locales, not depend on it. -- Ernesto Hernández-Novich - On Linux 2.6.12 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't apt-get it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
Bug#295393: tpb doesn't display on screen unless locale is C
Package: tpb Version: 0.6.3-1 After changing de system's locale to anything other than C (tested with en_US.*, es_ES.* and es_VE.*), tpb will not show anything on screen. Same thing happens if the system's locale is C but the user's locale is anything other than C. If the system's/user's locale is anything other than C, doing $ LANG=C tpb -d show proper behavior. I have all the xfonts-*-transcoded packages installed. I also tried with a TrueType font having an UTF-8 encoding. Seems tpb doesn't handle locales very well. As a workaround, I hardcoded the LANG=C in /etc/X11/Xsession.d/90tpb. -- Ernesto Hernández-Novich - On Linux 2.6.10 i686 - Unix: Live free or die! Geek by nature, Linux by choice, Debian of course. If you can't apt-get it, it isn't useful or doesn't exist. GPG Key Fingerprint = 438C 49A2 A8C7 E7D7 1500 C507 96D6 A3D6 2F4C 85E3
