Bug#894667: beep: CVE-2018-0492

[Tony Hoyle]

The patch vulnerability seems more severe to me, as people apply patches
all the time (they shouldn't do it as root, but people are people).

It's concerning that the holeybeep.ninja site exploited an unrelated
fault for 'fun' without apparently telling anyone.

Tony

Bug#770496: freeipa-client: Missing dependency for ipa-client-install

[Tony Hoyle]

Package: freeipa-client
Version: 4.0.5-1
Severity: normal

Dear Maintainer,

Installing on a clean(ish) machine errors out due to a missing
dependency on python-pyasn1

Traceback (most recent call last):
  File /usr/sbin/ipa-client-install, line 2790, in module
sys.exit(main())
  File /usr/sbin/ipa-client-install, line 2771, in main
rval = install(options, env, fstore, statestore)
  File /usr/sbin/ipa-client-install, line 2472, in install
api.finalize()
  File /usr/lib/python2.7/dist-packages/ipalib/plugable.py, line 708, in fina$
self.__do_if_not_done('load_plugins')
  File /usr/lib/python2.7/dist-packages/ipalib/plugable.py, line 482, in __do$
getattr(self, name)()
  File /usr/lib/python2.7/dist-packages/ipalib/plugable.py, line 645, in load$
self.import_plugins('ipalib')
  File /usr/lib/python2.7/dist-packages/ipalib/plugable.py, line 689, in impo$
__import__(fullname)
  File /usr/lib/python2.7/dist-packages/ipalib/plugins/cert.py, line 30, in $
from ipalib import pkcs10
  File /usr/lib/python2.7/dist-packages/ipalib/pkcs10.py, line 24, in module
from pyasn1.type import univ, char, namedtype, tag
ImportError: No module named pyasn1.type


-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental'), (1, 
'unstable')
Architecture: i386 (x86_64)

Kernel: Linux 3.2.0-2-amd64 (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages freeipa-client depends on:
ii  bind9utils   1:9.9.5.dfsg-5
ii  certmonger   0.75.14-2
ii  dnsutils 1:9.8.1.dfsg.P1-4.1
ii  krb5-user1.12.1+dfsg-14
ii  libc62.17-92
ii  libcomerr2   1.42.4-3
ii  libcurl3 7.38.0-3
ii  libcurl3-nss 7.38.0-3
ii  libk5crypto3 1.12.1+dfsg-14
ii  libkrb5-31.12.1+dfsg-14
ii  libldap-2.4-22.4.40-2
ii  libnss3-tools2:3.17.2-1
ii  libpopt0 1.16-7
ii  libsasl2-2   2.1.25.dfsg1-4+b1
ii  libsasl2-modules-gssapi-mit  2.1.26.dfsg1-12
ii  libxmlrpc-core-c31.33.14-0.2
ii  ntp  1:4.2.6.p5+dfsg-3.1+b1
ii  python   2.7.5-5
ii  python-dnspython 1.12.0-1
ii  python-freeipa   4.0.5-1
ii  python-krbv  1.0.90-1
ii  python-ldap  2.4.10-1
pn  python:any   none
ii  sssd 1.11.7-2
ii  wget 1.13.4-3

freeipa-client recommends no packages.

Versions of packages freeipa-client suggests:
pn  libpam-krb5  none

-- no debconf information


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

Bug#770495: freeipa-client: Does not set NIS domain name although it chaims to

[Tony Hoyle]

Package: freeipa-client
Version: 4.0.5-1
Severity: minor

Dear Maintainer,

During the run of ipa-client-install it claims to set the NIS domain name.
However after running (even after a reboot) nisdomainname is not set.

This breaks sudo (which apparently needs it), but everything else works.

I worked around it by setting it in rc.local.

Tony

-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental'), (1, 
'unstable')
Architecture: i386 (x86_64)

Kernel: Linux 3.2.0-2-amd64 (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages freeipa-client depends on:
ii  bind9utils   1:9.9.5.dfsg-5
ii  certmonger   0.75.14-2
ii  dnsutils 1:9.8.1.dfsg.P1-4.1
ii  krb5-user1.12.1+dfsg-14
ii  libc62.17-92
ii  libcomerr2   1.42.4-3
ii  libcurl3 7.38.0-3
ii  libcurl3-nss 7.38.0-3
ii  libk5crypto3 1.12.1+dfsg-14
ii  libkrb5-31.12.1+dfsg-14
ii  libldap-2.4-22.4.40-2
ii  libnss3-tools2:3.17.2-1
ii  libpopt0 1.16-7
ii  libsasl2-2   2.1.25.dfsg1-4+b1
ii  libsasl2-modules-gssapi-mit  2.1.26.dfsg1-12
ii  libxmlrpc-core-c31.33.14-0.2
ii  ntp  1:4.2.6.p5+dfsg-3.1+b1
ii  python   2.7.5-5
ii  python-dnspython 1.12.0-1
ii  python-freeipa   4.0.5-1
ii  python-krbv  1.0.90-1
ii  python-ldap  2.4.10-1
pn  python:any   none
ii  sssd 1.11.7-2
ii  wget 1.13.4-3

freeipa-client recommends no packages.

Versions of packages freeipa-client suggests:
pn  libpam-krb5  none

-- no debconf information


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

Bug#690227: Netatalk 3.0.6 packages for Debian

[Tony Hoyle]

Hi,

In common with the previous poster to this bug, I needed a newer version 
(3.0.3 was causing errors for me with mountain lion).  I've built 3.0.6 
using the same debian directory as 3.0.3 with minimal changes.  It works 
for me, but no warranty etc.


http://www.hoyle.me.uk/debian/

Tony


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

Bug#690227: Netatalk 3.0.6 packages for Debian

[Tony Hoyle]

On 11/01/2014 16:12, Jonas Smedegaard wrote:

Quoting Tony Hoyle (2014-01-11 16:45:08)

In common with the previous poster to this bug, I needed a newer
version (3.0.3 was causing errors for me with mountain lion).  I've
built 3.0.6 using the same debian directory as 3.0.3 with minimal
changes.  It works for me, but no warranty etc.

http://www.hoyle.me.uk/debian/


Thanks for sharing.

Can you elaborate on the kinds of issues you experienced with 3.0.3?

Netatalk 2.2.2 works fine for my needs, so I am interested if that's
simply because you use some features I don't.


  - Jonas

I had to update to 3.0.3 originally because of frequent corruption 
issues with time machine.  This was fixed somewhere in the 3.x series 
(or rather the google reports of corruption seem to die off somewhere 
after the release of 3.x).  3.0.3 has been fairly stable since install 
(April/May ish).


When I updated OSX recently I was finding the share was randomly coming 
up as not found, which was again breaking time machine (it silently 
failed to backup for over a week before I noticed).  So far 3.0.6 hasn't 
suffered the issue, although it's difficult to judge yet - OTOH being on 
a relatively recent release means I can make meaningful bug reports 
should I need to.


Why 2.2.2?  why not, for example, 2.2.5, if you're sticking with the old 
stable branch?


Tony


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

Bug#672291: pyaimt: Permissions wrong on first run

[Tony Hoyle]

Package: pyaimt
Version: 0.8.0.1-2
Severity: normal

On install pyaimt has incorrect permissions which mean it always fails to run up

It tries to create /var/lib/pyaimt/aim server name but it's running as the 
pyaimt user at
that point and /var/lib/pyaimt is owned by root.

candy:~# /etc/init.d/pyaimt start
Starting AIM transport for Jabber: pyaimt failed!
candy:~# cat /var/log/pyaimt/pyaimt.log 
candy:~# 

candy:~# su - pyaimt -s /bin/bash
pyaimt@candy:~$ python /usr/share/pyaimt/PyAIMt.py 
--config=/etc/pyaimt.conf.xml -o spooldir=/var/lib/pyaimt 
Traceback (most recent call last):
  File /usr/share/pyaimt/PyAIMt.py, line 10, in module
import main
  File /usr/share/pyaimt/src/main.py, line 126, in module
import session
  File /usr/share/pyaimt/src/session.py, line 6, in module
import legacy
  File /usr/share/pyaimt/src/legacy/__init__.py, line 4, in module
from glue import LegacyConnection, LegacyGroupchat, translateAccount
  File /usr/share/pyaimt/src/legacy/glue.py, line 41, in module
defaultAIMAvatar = avatar.AvatarCache().setAvatar(defaultAIMAvatarData)
  File /usr/share/pyaimt/src/avatar.py, line 103, in setAvatar
f = open(self.dir(key) + key, 'wb')
  File /usr/share/pyaimt/src/avatar.py, line 91, in dir
os.makedirs(d)
  File /usr/lib/python2.7/os.py, line 150, in makedirs
makedirs(head, mode)
  File /usr/lib/python2.7/os.py, line 150, in makedirs
makedirs(head, mode)
  File /usr/lib/python2.7/os.py, line 157, in makedirs
mkdir(name, mode)
OSError: [Errno 13] Permission denied: '/var/lib/pyaimt/aim.hoyle.me.uk'

candy:~# ls -ld /var/lib/pyaimt/
drwxr-xr-x 2 root root 4096 May  9 19:35 /var/lib/pyaimt/

A working fix is to change the ownership of /var/lib/pyaimt.

Tony

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

Kernel: Linux 3.0.0-2-amd64 (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages pyaimt depends on:
ii  adduser  3.113+nmu1  add and remove users and groups
ii  lsb-base 4.1+Debian0 Linux Standard Base 4.1 init scrip
ii  python   2.7.2-10interactive high-level object-orie
ii  python-crypto2.5-2   cryptographic algorithms and proto
ii  python-openssl   0.13-1  Python wrapper around the OpenSSL 
ii  python-support   1.0.14  automated rebuilding support for P
ii  python-twisted   11.1.0-1Event-based framework for internet
ii  python-twisted-web   11.1.0-1HTTP protocol implementation toget
ii  python-twisted-words 11.1.0-1Chat and Instant Messaging

Versions of packages pyaimt recommends:
ii  python-imaging1.1.7-4Python Imaging Library
ii  python-mysqldb1.2.3-1Python interface to MySQL
ii  python-nevow  0.10.0-4   Web application templating system 

Versions of packages pyaimt suggests:
ii  ejabberd  2.1.10-2   distributed, fault-tolerant Jabber

-- Configuration Files:
/etc/pyaimt.conf.xml changed [not included]

-- no debconf information



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

Bug#649988: bind9: rndc-confgen fails due to libssl dependency problem

[Tony Hoyle]

Same issue here.. just tried to update to bind 9.8.1 in testing and it
failed with the 'initializing DST: openssl failure' error.

I'm running openssl 1.0.0g, and the temporary fix did not work for me
(it seems to be related to chrooted bind and I'm not using that - it's
just a bog standard install).

Strace shows it loading the infamous libgost.so file correctly, then dying.

I have found no way past this error.. could the logging be improved?
'openssl failure' isn't exactly informative.

Tony

open(/usr/lib/ssl/openssl.cnf, O_RDONLY) = 8
fstat(8, {st_mode=S_IFREG|0644, st_size=10835, ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)
= 0x7f117e4fe000
read(8, #\n# OpenSSL example configuratio..., 4096) = 4096
read(8, Netscape crash on BMPStrings or ..., 4096) = 4096
read(8,  this to avoid interpreting an e..., 4096) = 2643
read(8, , 4096)   = 0
close(8)= 0
munmap(0x7f117e4fe000, 4096)= 0
futex(0x7f117cef50ec, FUTEX_WAKE_PRIVATE, 2147483647) = 0
open(/usr/lib/openssl-1.0.0/engines/libgost.so, O_RDONLY) = 8
read(8,
\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0\0\1\0\0\0\220X\0\0\0\0\0\0...,
832) = 832
fstat(8, {st_mode=S_IFREG|0644, st_size=95568, ...}) = 0
mmap(NULL, 2190728, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 8,
0) = 0x7f11791f1000
mprotect(0x7f1179207000, 2093056, PROT_NONE) = 0
mmap(0x7f1179406000, 8192, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 8, 0x15000) = 0x7f1179406000
close(8)= 0
munmap(0x7f11791f1000, 2190728) = 0
gettimeofday({1329160359, 137859}, NULL) = 0
stat(/etc/localtime, {st_mode=S_IFREG|0644, st_size=3661, ...}) = 0
write(2, 13-Feb-2012 19:12:39.137 initial..., 5913-Feb-2012
19:12:39.137 initializing DST: openssl failure
) = 59
gettimeofday({1329160359, 138046}, NULL) = 0
stat(/etc/localtime, {st_mode=S_IFREG|0644, st_size=3661, ...}) = 0
write(2, 13-Feb-2012 19:12:39.138 exiting..., 5413-Feb-2012
19:12:39.138 exiting (due to fatal error)
) = 54
exit_group(1)   = ?



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

Bug#578070: xl2tpd: [PATCH] Add ability to specify rx/tx speed

[Tony Hoyle]

Package: xl2tpd
Version: 1.2.5+dfsg-1
Severity: wishlist

The attached patch adds rx and tx speed options to the configuration file.  
This helps in situations where
the lns uses the speed information (for example to scale graphs, or for load 
balancing).

The patch is fairly simple as most of the code was already there, just 
hardcoded.

Tony

-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (500, 'testing'), (50, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-trunk-amd64 (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages xl2tpd depends on:
ii  libc6  2.10.2-2  GNU C Library: Shared libraries
ii  libpcap0.8 1.0.0-6   system interface for user-level pa
ii  ppp2.4.4rel-10.1 Point-to-Point Protocol (PPP) - da

xl2tpd recommends no packages.

xl2tpd suggests no packages.

-- no debconf information
diff -rubN xl2tpd-1.2.5+dfsg/control.c xl2tpd-1.2.5+dfsg.1/control.c
--- xl2tpd-1.2.5+dfsg/control.c	2009-12-12 18:32:57.0 +
+++ xl2tpd-1.2.5+dfsg.1/control.c	2010-04-16 13:35:02.606516672 +0100
@@ -204,11 +204,15 @@
 {
 t-ourrws = t-lns-tun_rws;
 t-hbit = t-lns-hbit;
+		t-rxspeed = t-lns-rxspeed;
+		t-txspeed = t-lns-txspeed;
 }
 else if (t-lac)
 {
 t-ourrws = t-lac-tun_rws;
 t-hbit = t-lac-hbit;
+		t-rxspeed = t-lac-rxspeed;
+		t-txspeed = t-lac-txspeed;
 }
 /* This is an attempt to bring up the tunnel */
 t-state = SCCRQ;
@@ -800,14 +804,14 @@
 mk_challenge (t-chal_them.vector, VECTOR_SIZE);
 add_randvect_avp (buf, t-chal_them.vector, VECTOR_SIZE);
 }
-add_txspeed_avp (buf, DEFAULT_TX_BPS);
+add_txspeed_avp (buf, t-txspeed);
 add_frame_avp (buf, c-frame);
 /*		if (c-ourrws = 0)
 			add_avp_rws(buf, c-ourrws); */
 /* FIXME: Packet Processing Delay */
 /* We don't need any kind of proxy PPP stuff */
 /* Can we proxy authenticate ourselves??? */
-add_rxspeed_avp (buf, DEFAULT_RX_BPS);
+add_rxspeed_avp (buf, t-rxspeed);
 /* add_seqreqd_avp (buf); *//* We don't have sequencing code, so
  * don't ask for sequencing */
 add_control_hdr (t, c, buf);
diff -rubN xl2tpd-1.2.5+dfsg/doc/l2tpd.conf.sample xl2tpd-1.2.5+dfsg.1/doc/l2tpd.conf.sample
--- xl2tpd-1.2.5+dfsg/doc/l2tpd.conf.sample	2009-12-12 18:32:57.0 +
+++ xl2tpd-1.2.5+dfsg.1/doc/l2tpd.conf.sample	2010-04-16 13:53:35.027270715 +0100
@@ -43,7 +43,10 @@
 ; tunnel rws = 4		; * RWS for tunnel (must be  0)
 ; flow bit = yes		; * Include sequence numbers
 ; challenge = yes		; * Challenge authenticate peer ; 
-;
+; rx bps = 1000; Receive tunnel speed
+; tx bps = 1000; Transmit tunnel speed
+; bps = 10	; Define both receive and transmit speed in one option
+
 ; [lac marko]			; Example VPN LAC definition
 ; lns = lns.marko.net	; * Who is our LNS?
 ; lns = lns2.marko.net	; * A backup LNS (not yet used)
diff -rubN xl2tpd-1.2.5+dfsg/file.c xl2tpd-1.2.5+dfsg.1/file.c
--- xl2tpd-1.2.5+dfsg/file.c	2009-12-12 18:32:57.0 +
+++ xl2tpd-1.2.5+dfsg.1/file.c	2010-04-16 13:33:14.782518750 +0100
@@ -91,6 +91,8 @@
 tmp-localaddr = 0;
 tmp-tun_rws = DEFAULT_RWS_SIZE;
 tmp-call_rws = DEFAULT_RWS_SIZE;
+tmp-rxspeed = DEFAULT_RX_BPS;
+tmp-txspeed = DEFAULT_TX_BPS;
 tmp-hbit = 0;
 tmp-lbit = 0;
 tmp-authpeer = 0;
@@ -306,6 +308,46 @@
 return 0;
 }
 
+int set_speed (char *word, char *value, int context, void *item)
+{
+if (atoi (value)  1 )
+{
+snprintf (filerr, sizeof (filerr),
+  bps must be greater than zero\n);
+return -1;
+}
+switch (context  ~CONTEXT_DEFAULT)
+{
+case CONTEXT_LAC:
+if (word[0] == 't')
+set_int (word, value, (((struct lac *) item)-txspeed));
+else if (word[0] == 'r')
+set_int (word, value, (((struct lac *) item)-rxspeed));
+else
+	{
+set_int (word, value, (((struct lac *) item)-rxspeed));
+set_int (word, value, (((struct lac *) item)-txspeed));
+	}
+break;
+case CONTEXT_LNS:
+if (word[0] == 't')
+set_int (word, value, (((struct lns *) item)-txspeed));
+else if (word[0] == 'r')
+set_int (word, value, (((struct lns *) item)-rxspeed));
+else
+	{
+set_int (word, value, (((struct lns *) item)-rxspeed));
+set_int (word, value, (((struct lns *) item)-txspeed));
+	}
+break;
+default:
+snprintf (filerr, sizeof (filerr), '%s' not valid in this context\n,
+  word);
+return -1;
+}
+return 0;
+}
+
 int set_rmax (char *word, char *value, int context, void 

Bug#549060: Please enable DHCPv6

[Tony Hoyle]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi,

I can confirm that version 4.1.1 builds cleanly and works with ipv6
(currently 4.1.1 is what apt-source pulls down even though there's only
a 4.1.0 binary).

I also made the startup script start the daemon twice - once for v4 and
once for v6 (since you can't mix modes, for reasons best known to the
authors).  Just added:

start-stop-daemon --start --quiet --pidfile $DHCP6DPID \
 --exec /usr/sbin/dhcpd -- -6 -cf /etc/dhcp/dhcpd6.conf
 -lf /var/lib/dhcp/dhcp6.leases -q $INTERFACES

It's a complete hack but I'm posting it just in case it's helpful to
someone else.

Tony

-BEGIN PGP SIGNATURE-
Version: GnuPG/MacGPG2 v2.0.12 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLmskhAAoJEJ1qCQ6ePCDU2UAH/j/k01J0qsBv6P15/qGJk5Xt
Qq80DPSnEoNTbIeHBRadmRmKW5g8yIIYUnXF36PWJ6qkrH1zJCGAaonN+dDKinPu
VHos4yZRWQBV+bpQNuzGWeEksSrBCS7SOSGrlWTf6y6Azfe2s9ocPmJ5FpH8pa6k
FmJLxK0zYxYC/7urbdYMIo+F/8W8HKp0uRo12YC+CQN1GY3GUZ+2OSWbtRTHwJdc
wAoWwoWq7plyiX80/n8mNzuggAB2MwkxCv07oUaog2h2kmVrB/SI1D459Jt43WrJ
4sSNp4/HkDuN79wwTUXOsKHJcWEBvl5g396+MD7rMqlV2jA5f/jinbSY/Y2oXL0=
=Dm7Q
-END PGP SIGNATURE-



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

Bug#433945: avahi-daemon: .local in unicast detection not perfect

[Tony Hoyle]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Package: avahi-daemon
Version: 0.6.19-2
Severity: normal

I've been thinking about this.. specifically why OSX has no problems
with DNS domains ending in .local and avahi does (and also that .local
in unicast detection can never work because that's not the root of the
problem IMO).

A company I work with has a domain ending in .local, historical, because
when they started they were a purely OSX shop and that's what their
limited knowledge said was good.  It's since sprouted a few active
directory servers, exchange email, etc.  Changing just isn't going to
happen.

In this environment from this OSX box I can ping any box in the domain -
OSX works perfectly.  avahi breaks in various ways, causing all sorts of
wierd effects.

What OSX is doing is distinguishing betweeen foo.bar.local and foo.local
- - the former is purely a DNS domain, and the latter is an mdns query.
This is shown where bar.local has an A record but OSX cannot in fact
resolve it.. it goes to mdns and aborts.  foo.bar.local responds
immediately.

avahi doesn't appear to do this.. it sees the .local, gets somewhat
confused and barfs (even if I rename a machine to foo.bar.local
experimentally and run avahi on it it can't resolve it, so it's not
working at all).

As to why the .local detection can't work - because on domains that
avahi has a problem with .local probably doesn't exist.. bar.local
might, but on a local DNS server it would be unusual to define the root
TLD like that.  I don't *think* anyone is actually using the .local
domain itself (never seen it, myself).

Also the solution proposed in message #5 wouldn't work in the case I'm
seeing - these are domains coming in over VPN not defined in
resolv.conf.. there's probably no way of detecting it automatically.

I propose that avahi be changed to match the OSX behaviour.  It's not
perfect, but it'd break a *lot* less systems.

Tony
-BEGIN PGP SIGNATURE-
Version: GnuPG/MacGPG2 v2.0.12 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLKW1YAAoJEJ1qCQ6ePCDUXuoH/RhpKRQw/LWhDJ/TXf+oj+Jj
9sIrUu3qPOp2NI2JczmS7Y3UMfe9bQsunLZlIyF0GxXWZ3qsuvfQ1L0ghyHfIbL5
HmPgJJ1m6IYcQSSYh5fdhXS5NLmWSAMhvvgQFy57FsLzDfU5XOGevBl7SmAh34yb
kKhX60xyml5KaoSR2yXhc/HptMx2JxhxUzvhLnrWtKOs39st2fYZiHSUXPlesSWD
ORCt/Q1KMp7isGPv/jGiIhbIOC49TUhJi+p2JSYhKol6143HQ6Pn5dMOxMKHAvrt
r3YhJKt3d1juGCXANk40MZFV9cyaEwI1VTzBPO7qvLm9UxfBZeKFb/+z1S83o+I=
=m72g
-END PGP SIGNATURE-



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

Bug#556932: apache2: Out of date comment in ports.conf

[Tony Hoyle]

Package: apache2.2-common
Version: 2.2.14-1
Severity: minor

ports.conf still contains the pre-SNI message in it's mod-ssl section:
# SSL name based virtual hosts are not yet supported, therefore no
# NameVirtualHost statement here

As they are supported, some better wording.. eg. SSL Name based virtual hosts 
require a browser with SNI support.  See 
http://wiki.apache.org/httpd/NameBasedSSLVHostsWithSNI;

It's probably not worth putting a NameVirtualHost in by default at this stage 
as there are plenty of browsers that don't yet support it (eg. pre-vista IE 
versions).

Tony


-- Package-specific info:
List of enabled modules from 'apache2 -M':
  alias auth_basic authn_file authz_default authz_groupfile
  authz_host authz_user autoindex cgi deflate dir env mime
  negotiation php5 setenvif ssl status

-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.18-128.4.1.el5xen (SMP w/4 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages apache2 depends on:
ii  apache2-mpm-prefork   2.2.14-1   Apache HTTP Server - traditional n
ii  apache2.2-common  2.2.14-1   Apache HTTP Server common files

apache2 recommends no packages.

apache2 suggests no packages.

Versions of packages apache2.2-common depends on:
ii  apache2-utils 2.2.14-1   utility programs for webservers
ii  apache2.2-bin 2.2.14-1   Apache HTTP Server common binary f
ii  libmagic1 5.03-2 File type determination library us
ii  lsb-base  3.2-23 Linux Standard Base 3.2 init scrip
ii  mime-support  3.46-1 MIME files 'mime.types'  'mailcap
ii  perl  5.10.1-7   Larry Wall's Practical Extraction 
ii  procps1:3.2.8-2  /proc file system utilities

-- no debconf information



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

Bug#535604: atftpd: Default config does not work without inetd and fails confusingly

[Tony Hoyle]

Package: atftpd
Version: 0.7.dfsg-7
Severity: normal

If you set USE_INETD to no then atftpd refuses to start with no error message 
other than usage(), which appears correct.
I had to dig into the source code to find out that you have to add --daemon 
onto the command line.

Two things wrong here:
1. The app just doesn't tell you what's wrong.. a generic usage() is useless in 
this case.  The man page is no help either.
2. The /etc/init.d/atftpd should be adding --daemon automatically if USE_INETD 
is false.

Tony

-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: powerpc (ppc)

Kernel: Linux 2.6.26-2-powerpc
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages atftpd depends on:
ii  debconf [debconf-2.0] 1.5.26 Debian configuration management sy
ii  libc6 2.9-12 GNU C Library: Shared libraries
ii  libpcre3  7.8-2  Perl 5 Compatible Regular Expressi
ii  libwrap0  7.6.q-16   Wietse Venema's TCP wrappers libra
ii  update-inetd  4.31   inetd configuration file updater

Versions of packages atftpd recommends:
ii  openbsd-inetd [inet-superse 0.20080125-2 The OpenBSD Internet Superserver

Versions of packages atftpd suggests:
ii  logrotate 3.7.7-3Log rotation utility

-- debconf information:
  atftpd/port: 69
  atftpd/tftpd-timeout: 300
  atftpd/mcast_port: 1758
  atftpd/verbosity: 5 (LOG_NOTICE)
  atftpd/timeout: true
  atftpd/tsize: true
  atftpd/retry-timeout: 5
  atftpd/multicast: true
  atftpd/ttl: 1
  atftpd/use_inetd: true
  atftpd/basedir: /var/lib/tftpboot
  atftpd/mcast_addr: 239.239.239.0-255
  atftpd/logfile: /var/log/atftpd.log
  atftpd/blksize: true
  atftpd/logtofile: false
  atftpd/maxthread: 100



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

Bug#532557: powerpc: vmfs tools returns corrupt data

[Tony Hoyle]

Mike Hommey wrote:

Actually, you shouldn't even get this message given that you only have
one extent. Can you provide the output for the following commands:
debugvmfs /dev/sdc1 show_volume
debugvmfs /dev/sdc1 show_fs


And debugvmfs /dev/sdc1 show_file_blocks /sol-10-u6-ga1-x86-dvd.iso, too.



OK. New test.  I installed esxi 3.5 update 4 on a spare hard drive, and 
copied a single file onto datastore1:


# debugvmfs /dev/sdb3 ls -l /
drwxr-xr-t 2 root root   1120 2009-06-11 15:46:19 .
drwxr-xr-t 2 root root   1120 2009-06-11 15:46:19 ..
-r 1 root root 851968 2009-06-11 15:32:29 .fbb.sf
-r 1 root root   63143936 2009-06-11 15:32:30 .fdc.sf
-r 1 root root  255655936 2009-06-11 15:32:30 .pbc.sf
-r 1 root root  260374528 2009-06-11 15:32:30 .sbc.sf
-r 1 root root4194304 2009-06-11 15:32:30 .vh.sf
-rw--- 1 root root 2665758720 2009-06-11 16:12:47 
en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso


This exhibits the identical issue:

# debugvmfs /dev/sdb3 cat 
/en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso 
tmp.iso

VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported


# debugvmfs /dev/sdb3 show_volume
Logical Volume Information:
  - UUID: 4a31157a-86455264-5a53-001b211d59d8
  - Size: 148 GB
  - Blocks  : 595
  - Num. Extents : 1

Physical Volume Information:
  - UUID: 4a31157d-c70d702a-3fb8-001b211d59d8
  - LUN : 0
  - Version : 3
  - Name: ATA HDS722516VLSA80 V34O
  - Num. Segments : 594
  - First Segment : 0
  - Last Segment  : 593

# debugvmfs /dev/sdb3 show_fs
VMFS FS Information:
  - Volume Version   : 4
  - Version  : 31
  - Label: datastore1
  - UUID : 4a31157d-272fa266-7a32-001b211d59d8
  - Block size   : 1048576 (0x10)
  - Subblock size: 65536 (0x1)
  - FDC Header size  : 0x1
  - FDC Bitmap count : 0x20

# debugvmfs /dev/sdb3 show_file_blocks 
en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso

0x0015e101 0x0015e141 0x0015e181 0x0015e1c1
0x0015e201 0x0015e241 0x0015e281 0x0015e2c1
0x0015e301 0x0015e341 0x0015e381 0x0015e3c1
0x0015e401 0x0015e441 0x0015e481 0x0015e4c1
0x0015e501 0x0015e541 0x0015e581 0x0015e5c1
0x0015e601 0x0015e641 0x0015e681 0x0015e6c1
0x0015e701 0x0015e741 0x0015e781 0x0015e7c1
0x0015e801 0x0015e841 0x0015e881 0x0015e8c1
0x0015e901 0x0015e941 0x0015e981 0x0015e9c1
0x0015ea01 0x0015ea41 0x0015ea81 0x0015eac1
0x0015eb01 0x0015eb41 0x0015eb81 0x0015ebc1
0x0015ec01 0x0015ec41 0x0015ec81 0x0015ecc1
0x0015ed01 0x0015ed41 0x0015ed81 0x0015edc1
0x0015ee01 0x0015ee41 0x0015ee81 0x0015eec1
0x0015ef01 0x0015ef41 0x0015ef81 0x0015efc1
0x0015f001 0x0015f041 0x0015f081 0x0015f0c1
0x0015f101 0x0015f141 0x0015f181 0x0015f1c1
0x0015f201 0x0015f241 0x0015f281 0x0015f2c1
0x0015f301 0x0015f341 0x0015f381 0x0015f3c1
0x0015f401 0x0015f441 0x0015f481 0x0015f4c1
0x0015f501 0x0015f541 0x0015f581 0x0015f5c1
0x0015f601 0x0015f641 0x0015f681 0x0015f6c1
0x0015f701 0x0015f741 0x0015f781 0x0015f7c1
0x0015f801 0x0015f841 0x0015f881 0x0015f8c1
0x0015f901 0x0015f941 0x0015f981 0x0015f9c1
0x0015fa01 0x0015fa41 0x0015fa81 0x0015fac1
0x0015fb01 0x0015fb41 0x0015fb81 0x0015fbc1
0x0015fc01 0x0015fc41 0x0015fc81 0x0015fcc1
0x0015fd01 0x0015fd41 0x0015fd81 0x0015fdc1
0x0015fe01 0x0015fe41 0x0015fe81 0x0015fec1
0x0015ff01 0x0015ff41 0x0015ff81 0x0015ffc1
0x00160001 0x00160041 0x00160081 0x001600c1
0x00160101 0x00160141 0x00160181 0x001601c1
0x00160201 0x00160241 0x00160281 0x001602c1
0x00160301 0x00160341 0x00160381 0x001603c1
0x00160401 0x00160441 0x00160481 0x001604c1
0x00160501 0x00160541 0x00160581 0x001605c1
0x00160601 0x00160641 0x00160681 0x001606c1
0x00160701 0x00160741 0x00160781 0x001607c1
0x00160801 0x00160841 0x00160881 0x001608c1
0x00160901 0x00160941 0x00160981 0x001609c1
0x00160a01 0x00160a41 0x00160a81 0x00160ac1
0x00160b01 0x00160b41 0x00160b81 0x00160bc1
0x00160c01 0x00160c41 0x00160c81 0x00160cc1
0x00160d01 0x00160d41 0x00160d81 0x00160dc1
0x00160e01 0x00160e41 0x00160e81 0x00160ec1
0x00160f01 0x00160f41 0x00160f81 0x00160fc1
0x00161001 0x00161041 0x00161081 0x001610c1
0x00161101 0x00161141 0x00161181 0x001611c1
0x0015e001 0x0015e041 0x0015e081 0x00161201
0x00161241 0x00161281 0x001612c1 0x00161301
0x00161341 0x00161381 0x001613c1 0x00161401
0x00161441 0x00161481 0x001614c1 0x00161501
0x00161541 0x00161581 0x001615c1 0x00161601
0x00161641 0x00161681 0x001616c1 0x00161701
0x00161741 0x00161781 0x001617c1 0x00161801
0x00161841 0x00161881 0x001618c1 0x00161901
0x00161941 0x00161981 0x001619c1 0x00161a01
0x00161a41 0x00161a81 0x00161ac1 0x00161b01
0x00161b41 0x00161b81 0x00161bc1 0x00161c01
0x00161c41 0x00161c81 0x00161cc1 0x00161d01
0x00161d41 0x00161d81 0x00161dc1 

Bug#532557: powerpc: vmfs tools returns corrupt data

[Tony Hoyle]

Mike Hommey wrote:


Just untar the file, go into the created directory, run make, then run
./debugvmfs


This one crashes..

# ./debugvmfs /dev/sdb3 ls /
Segmentation fault


This is because of an endianness issue in vmfs_inode_read:

   if (inode-mdh.magic != VMFS_INODE_MAGIC)
  return(-1);

Presumably this should go via read_le32, but I just commented it out for 
my tests.


# ./debugvmfs /dev/sdb3 cat 
/en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso 
/tmp/out.iso

VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported
...

etc.


I put some debugging in there to see what was happenening.  It starts 
off OK:
# ./debugvmfs /dev/sdb3 cat 
/en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso 
/tmp/out.iso

pos=0020 len=0200 pos+len=00200200
vmfs_lvm_extent_size = 2000
pos=0040 len=1000 pos+len=00401000
vmfs_lvm_extent_size = 2000
pos=00418000 len=1000 pos+len=00419000
vmfs_lvm_extent_size = 2000
pos=0050 len=1000 pos+len=00501000
vmfs_lvm_extent_size = 2000
pos=0050 len=1000 pos+len=00501000
vmfs_lvm_extent_size = 2000
pos=0050 len=1000 pos+len=00501000
vmfs_lvm_extent_size = 2000

Then suddenly goes bananas:

pos=02ae len=1000 pos+len=02ae1000
vmfs_lvm_extent_size = 2000
pos=1002 len=1000 pos+len=10021000
vmfs_lvm_extent_size = 2000
pos=57840 len=1000 pos+len=578401000
vmfs_lvm_extent_size = 2000
VMFS: i/o spanned over several extents is unsupported
pos=1002 len=1000 pos+len=10021000
vmfs_lvm_extent_size = 2000
pos=578401000 len=1000 pos+len=578402000
vmfs_lvm_extent_size = 2000
VMFS: i/o spanned over several extents is unsupported
pos=1002 len=1000 pos+len=10021000
vmfs_lvm_extent_size = 2000
pos=578402000 len=1000 pos+len=578403000
vmfs_lvm_extent_size = 2000
VMFS: i/o spanned over several extents is unsupported
pos=1002 len=1000 pos+len=10021000
vmfs_lvm_extent_size = 2000
pos=578403000 len=1000 pos+len=578404000
vmfs_lvm_extent_size = 2000
VMFS: i/o spanned over several extents is unsupported

It seems that every alternate read is completely wrong from that point on.

Hope this helps

Tony



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

Bug#532557: powerpc: vmfs tools returns corrupt data

[Tony Hoyle]

Mike Hommey wrote:

Would you mind trying the newly attached tarball ? (Same build 
instructions as before)



+ a dependency on readline-dev :p

Everything seems to work now:

# ./debugvmfs /dev/sdb3 cat
/en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso 
 out.iso

# md5sum out.iso
27c58cdb3d620f28c36333a5552f271c  out.iso

# ./vmfs-fuse /dev/sdb3 /mnt
# md5sum 
/mnt/en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso
27c58cdb3d620f28c36333a5552f271c 
en_windows_server_2008_datacenter_enterprise_standard_x64_dvd_X14-26714.iso


Looking good :)

Tony



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

Bug#532557: powerpc: vmfs tools returns corrupt data

[Tony Hoyle]

Mike Hommey wrote:

Hi Tony,

On Wed, Jun 10, 2009 at 03:59:52AM +0100, Tony Hoyle wrote:

Package: vmfs-tools Version: 0.1.1-1 Severity: important

On powerpc, any file over 2MB returns essentially random data.
This is testable by running md5sum across any file larger than this
- it comes up with different results each time.

The same package installed on amd64 appears to behave correctly, so
this is a platform specific issue, possibly due to endinness.


Does this happen with debugvmfs of vmfs-fuse ? I can't reproduce with
 debugvmfs on ppc.


Here's the output I'm getting from a simple ms5sum:

# md5sum sol-10-u6-ga1-x86-dvd.iso sol-10-u6-ga1-x86-dvd.iso
sol-10-u6-ga1-x86-dvd.iso
7ac54248f85cd22764dcbd4c503e81ea  sol-10-u6-ga1-x86-dvd.iso
20a9814af68cdf2cc1749fdf540437f5  sol-10-u6-ga1-x86-dvd.iso
b5134e4d505a5cd369912349610895da  sol-10-u6-ga1-x86-dvd.iso

None of these are correct!

debugvmfs also does not work:

# debugvmfs /dev/sdc1 cat sol-10-u6-ga1-x86-dvd.iso /tmp/sol.iso
VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported
VMFS: i/o spanned over several extents is unsupported
...

repeated about 1000 times

Presumably the driver is not handling this condition and silently
corrupting the files.

The same command executed on amd64 with the exact same vmfs image gives
no errors and produces a file that appears correct.

# debugvmfs /dev/sdb1 cat sol-10-u6-ga1-x86-dvd.iso /tmp/sol.iso
# md5sum /tmp/sol.iso
ae8e9aa4a1a3f3fd8846a4caef9f09e2  /tmp/sol.iso

I didn't do anything special to create the drive, and both machines have 
the same version of everything as far as I can tell - kernel, libc 
version, etc.  The only difference (and I'll admit it's a biggie) is the 
different processor.  It was created on esxi 3.5 (I think.. the drive it 
was on failed hence the requirement to extract the data, but 4.0 was 
only released last month and I definately haven't run any updates in 
that time).


Unfortunately this drive has to go back into the machine it came from to 
be reused.. I'll look at preserving the contents but as it's about 200gb 
of data that's going to be difficult.  Hope I've given you enough to 
work with.


Tony



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

Bug#532557: powerpc: vmfs tools returns corrupt data

[Tony Hoyle]

Package: vmfs-tools
Version: 0.1.1-1
Severity: important

On powerpc, any file over 2MB returns essentially random data.  This is 
testable by running md5sum across any file larger than this - it comes 
up with different results each time.


The same package installed on amd64 appears to behave correctly, so this
is a platform specific issue, possibly due to endinness.

Tony

-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: powerpc (ppc)

Kernel: Linux 2.6.26-2-powerpc
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages vmfs-tools depends on:
ii  libc6 2.9-12 GNU C Library: Shared libraries
ii  libfuse2  2.7.4-1.1  Filesystem in USErspace library
ii  libreadline5  5.2-4  GNU readline and history 
libraries





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

Bug#480500: libio-socket-inet6-perl: Incompatible redefines between INET6.pm and, Exporter.pm

[Tony Hoyle]

This bug also affects amavis and spamassassin, making their cron jobs 
extremely noisy.


Example output:

Constant subroutine Net::DNS::Resolver::Base::AF_INET6 redefined at 
/usr/share/perl/5.10/Exporter.pm line 66.

 at /usr/lib/perl5/Net/DNS/Resolver/Base.pm line 65
Prototype mismatch: sub Net::DNS::Resolver::Base::AF_INET6 () vs none at 
/usr/share/perl/5.10/Exporter.pm line 66.

 at /usr/lib/perl5/Net/DNS/Resolver/Base.pm line 65
Constant subroutine Net::DNS::Resolver::Base::PF_INET6 redefined at 
/usr/share/perl/5.10/Exporter.pm line 66.

 at /usr/lib/perl5/Net/DNS/Resolver/Base.pm line 65
Prototype mismatch: sub Net::DNS::Resolver::Base::PF_INET6 () vs none at 
/usr/share/perl/5.10/Exporter.pm line 66.

 at /usr/lib/perl5/Net/DNS/Resolver/Base.pm line 65
bayes: synced databases from journal in 1 seconds: 2386 unique entries 
(3911 total entries)


Tony



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

Bug#479564: libio-socket-inet6-perl: New upstream version (2.54) available

[Tony Hoyle]

Package: libio-socket-inet6-perl
Version: 2.51-1
Severity: minor

A new version is available on CPAN.  This fixes a number of 
spurious warnings - see 
http://www.nntp.perl.org/group/perl.module-authors/2008/01/msg6187.html


System Information:
Debian Release: lenny/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.22-2-amd64 (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages libio-socket-inet6-perl depends on:
ii  libsocket6-perl   0.19-1+b1  Perl extensions for IPv6
ii  perl  5.10.0-9   Larry Wall's Practical Extraction 

libio-socket-inet6-perl recommends no packages.

-- no debconf information



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

Bug#428392: dkim-filter.sock permissions should be more MTA friendly

[Tony Hoyle]

Mike Markley wrote:

What user is Postfix running as when it attempts to connect to the
socket, BTW? I'd like to update README.Debian with explicit instructions
on how to add the correct user to the group.


It's definable (setting mail_owner in main.cf) but by default runs as 
'postfix'.


A wrinkle is the out of the box debian package runs smtpd chrooted so 
you've got to change that line in master.cf so it can get at the socket 
(there's probably a way of putting the socket in the chroot but I don't 
know if there's a debianised way of doing that automatically - might be 
worth asking the postfix maintainer).


Tony





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

Bug#428392: dkim-filter.sock permissions should be more MTA friendly

[Tony Hoyle]

Mike Markley wrote:
You don't specify which MTA you're running; stock Sendmail shouldn't 
have an issue.


Postfix.  It runs as its own unprivileged user (I'm surprised that
sendmail doesn't do this also).


With that said, I'm perfectly fine with adding such a group, but I'll
 need to figure out the best way to make this happen. The socket's
being created and having its permissions set out of libmilter by 
smfi_opensocket().


You could probably just set the sgid bit on the parent directory... then 
there's just the actual permissions on the socket to change (is it 
obeying umask by any chance??).


Tony


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

Bug#428392: dkim-filter.sock permissions should be more MTA friendly

[Tony Hoyle]

Package: dkim-filter
Version: 1.0.0.dfsg-1
Severity: normal

The default permissions of /var/run/dkim-filter/dkim-filter.sock only 
allow the dkim-filter user to write to it.  This precludes the MTA being 
able to use it as a milter, since that runs as its own user.

Also default permissions allow execute with is pointless (you can't of 
course execute a socket).  That's a minor issue though.

I suggest creating a dkim-milter group at install and giving 
that write access to the socket.

Tony

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

Kernel: Linux 2.6.20-1-xen-amd64 (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages dkim-filter depends on:
ii  adduser   3.102  Add and remove users and groups
ii  libc6 2.5-10 GNU C Library: Shared libraries
ii  libdb4.5  4.5.20-4   Berkeley v4.5 Database Libraries [
ii  libmilter18.14.1-4   Sendmail Mail Filter API (Milter)
ii  libssl0.9.8   0.9.8e-5   SSL shared libraries

dkim-filter recommends no packages.

-- no debconf information


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

Bug#403195: installation: Finnish timezone not selected by the etch installer

[Tony Hoyle]

Martin Michlmayr wrote:

So I don't see how you'd get Europe/Andorra at all.  I'm fairly sure I
saw this myself in the past but maybe something changed in the
installer.  (This menu where you can choose timezones after the
partitioner definitely wasn't there in the past).


It's possible it got 'fixed' at some point and the binary build hasn't 
been updated.  Both me and the original reporter saw it though..


Which image did you use?  I used the unofficial one (with the network 
firmware in it) since not having ethernet seemed kinda a biggie for me 
with the official builds... (http://www.slug-firmware.net/d-dls.php)


If there's a later one and they haven't rebuilt it that would explain 
the difference.


IMO the menu should show all timezones though - maybe the d-i guys can 
add a 'world' country or something so that can happen (I'm only guessing 
that it's that easy... if not then just ignore me :) )


Tony


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

Bug#403195: installation: Finnish timezone not selected by the etch installer

[Tony Hoyle]

Martin Michlmayr wrote:

I cannot do a test install to check right now but I'm pretty sure the
dialog you see only talks about the mirror (i.e. which country do you
live in so we can select a proper mirror).


No it's another one.  About 2/3rds of the way through a dialog pops up 
saying 'You are in Europe/Andorra timezone'* with no option to override 
or anything, just a confirmation button.


The mirrors screen is OK - it lists all of them and you select your 
local one.


Tony

* That may not be the exact wording.


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

Bug#403195: installation: Finnish timezone not selected by the etch installer

[Tony Hoyle]

Hi

Just encountered this with the UK timezone as well.. same distribution.

Selected UK, got the correct mirror list etc. but the timzone was 
'Europe/Andorra'.  Is it selecting the first (alphabetically) in the list?


Seems to be a bug in the NSLU2 distribution - probably needs pushing to 
whoever maintains that one.


Tony


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

Bug#403195: installation: Finnish timezone not selected by the etch installer

[Tony Hoyle]

Christian Perrier wrote:

Could you give details about the successive actions you had in the
language choosing step:


Hmm.. difficult to remember as I eventually decided to use a different 
image to boot the slug.



-chosen language
-chosen country


Would have been UK English.


Which priority did you use for the install? Was it the default
priority (ie booting by just hitting Enter at the installer boot
prompt), or something else like epxert?


This device doesn't have an installer boot prompt.

Basically the slug has no console or serial port.  What happens is you 
boot it with the bootstrap image and eventually it beeps and you can ssh 
in.  At that point it's in the debian installer in 'Low memory mode' and 
you have to select a short list of installer components to load.


See http://www.cyrius.com/debian/nslu2/install.html for the installation 
instructions (as far as I know the same ones everyone uses).


One possibility is the instructions are missing something important (eg. 
a missing installer component) and it's confusing the timezones somehow.


Tony



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

Bug#403195: installation: Finnish timezone not selected by the etch installer

[Tony Hoyle]

Martin Michlmayr wrote:

The long term solution is to somehow make the country/timezone
selection available after the net is brought up.



Anyway, this is documented on
http://cyrius.com/debian/nslu2/install.html

After the installation, you need to configure your timezone because a
default value was chosen for this during the installation. You can do
this with the tzconfig command. Apart from this, your new Debian
system is fully configured.


How about  skipping the dialog entirely if the timezone isn't set?  It's 
 just confusing having it there at the moment (confused both me and the 
original reporter).  Or at least having it say 'Defaulting to UTC, you 
need to set timezone manually' or similar.


Andorra seems just like an odd default to me...

Tony


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

Bug#379285: Incorrect location for asterisk.pid and asterisk.ctl

[Tony Hoyle]

Package: asterisk
Version: 1:1.2.7.1.dfsg-3
Severity: normal

In the default configuration asterisk runs as an unprivileged user, but
the binary tries to create and use /var/run/asterisk.pid and 
/var/run/asterisk.ctl and fails (without the .ctl the -r function will 
not work).  

There is a /var/run/asterisk directory presumably setup for this purpose 
but it's not trying to use it.

The only fix currently is to run asterisk as root, which is not ideal.

Tony

-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (500, 'stable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.16-2-xen-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages asterisk depends on:
ii  adduser 3.87 Add and remove users and groups
ii  asterisk-classic1:1.2.7.1.dfsg-3 Open Source Private Branch Exchang

asterisk recommends no packages.

-- no debconf information


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

Bug#379285: Incorrect location for asterisk.pid and asterisk.ctl

[Tony Hoyle]

Tzafrir Cohen wrote:


The directory for the pid file is set in /etc/asterisk/asterisk.conf ,
in the section [directories], value 'astrundir'

What do you have there?


Ahh... never heard of that setting before.  changing that worked.. thanks.

/etc/init.d/asterisk still has:

PIDFILE=/var/run/asterisk/asterisk.pid

which needs changing manually (and perhaps should be in the default build) but 
it does seem to work as it is.


I guess you can close this bug.  Thanks for the help.

Tony



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

Bug#379285: Incorrect location for asterisk.pid and asterisk.ctl

[Tony Hoyle]

Tzafrir Cohen wrote:

I believe you got thing wrong: The PID should not be in
/var/run/asterisk.pid (if you want to be able to run asterisk as
non-root).


That's what I'm saying - the default /etc/init.d/asterisk has that path 
hardcoded in it, so that script needs changing.


Tony








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

Bug#291883: libhowl-dev: howl/howl.h is incorrect

[Tony Hoyle]

Package: libhowl-dev
Version: 0.9.8-2
Severity: important

howl/howl.h does not have the correct prefix within it for including
its files, so it tries to include /usr/include/salt/salt.h instead
of /usr/include/howl/salt/salt.h.  It's hard to autoconf for this, since other 
platforms do this correctly.

Please either fix the header or move the files into /usr/include where the 
upstream packages expect them to be.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (1, 'experimental')
Architecture: i386 (i686)
Kernel: Linux 2.6.7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages libhowl-dev depends on:
ii  libhowl0  0.9.8-2Library for Zeroconf service disco

-- no debconf information


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

Bug#281247: assert messages

[Tony Hoyle]

 The author of HOWL has been mailed and he tells that it happens
 because libhowl0 has been compiled without -DNDEBUG.
This is in fact an upstream bug (there's no occurrence of NDEBUG in the 
howl code, so the above could never be true).

The following patch fixes it:
[EMAIL PROTECTED]:~/howl-0.9.8/src/lib/howl$ diff -u debug.c.old debug.c
--- debug.c.old 2005-01-23 20:11:18.0 +
+++ debug.c 2005-01-23 20:11:46.0 +
@@ -47,6 +47,7 @@
char message[1024];
char string[512];
+#ifdef _DEBUG
if (code)
{
sprintf(message, [assert] error: %d %s\n[assert] 
where: \%s\, \%s\, line: %d\n\n, code, sw_format_error_string(code, 
string), file, func, line);
@@ -61,6 +62,8 @@
 #if defined(WIN32)
OutputDebugString(message);
 #endif
+
+#endif
 }

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