gmane.comp.freedesktop.avahi

Multiple Interface (avahi 0.6.25)

John Tobias — 2012-05-14T22:34:18

Hello All,

I have a device with multiple interfaces (eth0, and wireless(ra0) ) and
they are configured with avahi-autoipd IP addresses.

eth0:avahi -> 169.254.66.17
ra0:avahi ->169.254.129.129

In my host machine, the wireless (ra0) interface is only available at this
time and I am able to communicate to the said device thru adhoc.
I can ping the 169.254.129.129 which is good but, the problem is I can ping
the eth0 (169.254.66.17) too.

Is there a way to fix the said problem?.

Thanks,

john

How to add an alias for the .local domain

Timothy Madden — 2012-05-10T00:59:55

Hello

I would like to give an additional name to my .local domain, so any 
avahi-daemon host is accessible as either
hostname.local
or
hostname.xe-test

Is that possible with avahi ? I could not find any documentation about 
it. I have an /etc/mdns.allow file containing both names:

.local
.xe-test

Still no name resolving for .xe-test.

I have avahi 0.6.25 on CentOS 6.2 (0.6.25-11.el6) x86_64 and nss-mdns 
0.10 (0.10-8.el6)

Thank you,
Timothy Madden

[PATCH] build-sys: fix parallel install in avahi-utils

Diego Elio Pettenò — 2012-05-01T00:09:45

The moment install-exec-local is called, we might still not have
created ${DESTDIR}/${bindir} so we should make sure to create it
first, and then try to chdir into it.

Signed-off-by: Diego Elio Pettenò <flameeyes< at >flameeyes.eu>
---
 avahi-utils/Makefile.am |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/avahi-utils/Makefile.am b/avahi-utils/Makefile.am
index a644b4a..1abc79a 100644
--- a/avahi-utils/Makefile.am
+++ b/avahi-utils/Makefile.am
< at >< at > -54,7 +54,8 < at >< at > avahi_set_host_name_CFLAGS = $(AM_CFLAGS)
 avahi_set_host_name_LDADD = $(AM_LDADD) ../avahi-client/libavahi-client.la ../avahi-common/libavahi-common.la
 
 install-exec-local:
-cd $(DESTDIR)/$(bindir) && \
+$(mkdir_p) $(DESTDIR)/$(bindir) && \
+cd $(DESTDIR)/$(bindir) && \
 rm -f avahi-resolve-host-name avahi-resolve-address avahi-browse-domains avahi-publish-address avahi-publish-service && \
 $(LN_S) avahi-resolve avahi-resolve-host-name && \
 $(LN_S) avahi-resolve avahi-resolve-address && \

Package dbus-1 was not found in the pkg-config search path.

Robert McCullough — 2012-04-30T14:00:15

*Hi,*
*
*
*I am trying to build avahi for our embedded MPC5200B powerpc running Linux
2.6.34.4.*
*I am getting the following configure error.*

Package dbus-1 was not found in the pkg-config search path.
Perhaps you should add the directory containing `dbus-1.pc'
to the PKG_CONFIG_PATH environment variable
No package 'dbus-1' found
Error: Couldn't determine the version of your D-Bus package.
  This is probably an error in this script, please report it
  along with the following information:
      Base D-Buss version =''
      DBUS_VERSION_MAJOR='0'
      DBUS_VERSION_MINOR='0'
      DBUS_VERSION_MICRO='0'

*Can someone help me to locate where my problem is?*
*
*
*
*
*Here is my build script:*

#!/bin/sh


##
# Builds  avahi for Linux x86
##
AVAHI_ROOT=/opt/vendors/avahi
AVAHI_VER=avahi-0.6.30
AVAHI_BUILD=linux_x86_build
AVAHI_INSTALL=linux_x86_install
CONFIG_LOG=config-linux_x86.log
BUILD_LOG=build-linux_x86.log
INSTALL_LOG=install-linux_x86.log

EXPAT_INCLUDE=/opt/vendors/Expat/include/expat-2_0_1
EXPAT_LIB

Re: Memory consumption

Lennart Poettering — 2012-04-12T10:46:13


As mentioned, this will only minimize address space, not memory
usage. Allocating a biggre stack comes at virtually no cost -- as long
as it is never accessed. Unless you turned off lazy allocation in Linux
chasing this allocation is a waste of time. You are chasing a ghost.

Have you turned off lazy allocation on Linux?

(Note that the event loop logic in avahi is fully abstracted, you can
implement your own logic, and run the stuff in your own thread with your
own parameters. AvahiThreadedPoll is just one implementation, but
nothing you actually have to use.

Lennart

Re: Memory consumption

Frank Lahm — 2012-04-12T09:45:04

Am 11. April 2012 17:44 schrieb Lennart Poettering <lennart-mdGvqq1h2p+GdvJs77BJ7Q< at >public.gmane.org>:

This limits the size of the stack of the main thread in the first
place and then that value is used by pthread_create() for the size of
the stack allocation. I've tried setting setrlimit(RLIMIT_STACK,...)
to a smaller value before calling avahi_threaded_poll_new|start(), but
that doesn't alter the allocation.


Yes.

Regards!
-f

Re: Memory consumption

Lennart Poettering — 2012-04-11T15:44:50



This might simply be the stack for the thread, which you can influence
with "ulimit -s" in its size.

However, note that having a large stack doesn't actually mean that this
uses a lot of memory, it just reserves address space, which is only
backed when needed. Hence, most likely this is really a
non-issue. (i.e. don't confuse reservation of memory with reservation of
address space!)

Lennart

Re: Memory consumption

Frank Lahm — 2012-04-11T15:41:25

Am 11. April 2012 16:50 schrieb Frank Lahm <franklahm< at >googlemail.com>:

Would it make sense to extend the Avahi API so that a parameter could
be passed which is then used to call pthread_attr_setstacksize() for
the thread?

-f
_______________________________________________
avahi mailing list
avahi< at >lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/avahi

Re: Memory consumption

Frank Lahm — 2012-04-11T14:50:52

Am 10. April 2012 22:26 schrieb Lennart Poettering <lennart-mdGvqq1h2p+GdvJs77BJ7Q< at >public.gmane.org>:

I had checked with Valgrind memcheck before to no avail. I've looked
at massif as to your recommendation and that is indeed quite revealing
(this requires the --pages-as-heap=yes option as the large allocation
is apparently not done via malloc et al):

$ sudo valgrind --tool=massif --pages-as-heap=yes --detailed-freq=1
/usr/local/netatalk/sbin/afpd -d
...
^C

Looking at the last snapshot:

$ ms_print ...
...
--------------------------------------------------------------------------------
  n        time(i)         total(B)   useful-heap(B) extra-heap(B)    stacks(B)
--------------------------------------------------------------------------------
 61  1,283,953,675       84,291,584       84,291,584             0            0
100.00% (84,291,584B) (page allocation syscalls) mmap/mremap/brk,
--alloc-fns, etc.
...
->12.45% (10,493,952B) 0x6B32F99: mmap (in /lib/libc-2.7.so)
| ->12.44% (10,489,856B) 0x6848AEF: p

Re: Memory consumption

Lennart Poettering — 2012-04-10T20:26:47


I don't see where the Avahi client libraries cuould do such a big
allocation. My only guess is that this is actually the D-Bus library
that does this (it maintains a message cache). To figure this out it
might be worth to plot the memory usage with a tool like valgrind's
massif tool?

Lennart

Memory consumption

Frank Lahm — 2012-04-10T13:23:56

Hi all,

I'm one of the Netatalk (OpenSource AFP fileserver) devs. We use Avahi
via avahi_threaded_poll and friends in order to register the AFP
server with mdns.
We call the Avahi setup stuff once in our main process `afpd` here
[1]. Handling user sessions is done by forking childs.

As it seems, when calling the Avahi setup functions, a big chunk of
memory is taken:

With Avahi registration:

$ s pmap 14083
14083:   /usr/local/netatalk/sbin/afpd -d -F /usr/local/netatalk/etc/afp.conf
0000000000400000    308K r-x--  /usr/local/netatalk/sbin/afpd
000000000064c000     24K rw---  /usr/local/netatalk/sbin/afpd
0000000000652000    116K rw---    [ anon ]
000000001fd34000    280K rw---    [ anon ]
000000004068d000      4K -----    [ anon ]
000000004068e000  10240K rw---    [ anon ]

Without Avahi registration:

$ s pmap 14233
14233:   /usr/local/netatalk/sbin/afpd -d -F /usr/local/netatalk/etc/afp.conf
0000000000400000    308K r-x--  /usr/local/netatalk/sbin/afpd
000000000064c000     24K rw---  /usr/local/netatalk

Multiple AvahiThreadedPolls in the same process

Joe Hasbani — 2012-03-19T19:52:43

Hi all,

Is it possible/safe to have multiple AvahiThreadedPolls running in the
same process?   (In an ideal world, I wouldn't be doing this, but I do
have my reasons.)

I ask because I am running into trouble.  In my program, each poll has
an associated client and service browser for the same service.
Everything looks to be functioning fine until I attempt to resolve a
service.  At that point, any number of crashes may be experienced.  If
I run with only one poll, I don't experience these crashes, and
resolutions complete successfully.  I am using no globals to track the
poll, client, or browser, so I don't anticipate conflict due to my
code.

I guess I am wondering if I have a bug in my code or if this simply
isn't expected to work.

Please find below some sample stack traces.

Thanks for your time,

Joe Hasbani
Pure Storage


Program received signal SIGSEGV, Segmentation fault.
#0  0x00007fffeeb64bfc in ?? () from /lib/x86_64-linux-gnu/libdbus-1.so.3
#1  0x00007fffeeb64c69 in ?? () from /lib/x86_64-linux-

Re: The reflector option in networks with loops

David Dyja — 2012-03-14T17:26:20

Anybody?

2012/3/13 David Dyja <d.dyja-oe7qfRrRQfekNM68Nv64ybNAH6kLmebB< at >public.gmane.org>

Re: Bonjour emulation by Avahi on Ubuntu

Gaspard Bucher — 2012-03-14T07:49:04

Avahi can be difficult to grok... I just went through this transition and
it was not easy due to missing docs for low level stuff such as the PollApi.

Here is some advice depending on your situation:

A. You want to avoid creating new threads and use your own select/poll:

You have to create a AvahiWatch and AvahiTimer classes that will insert
file descriptors into your event loop. See poll API for
Qt<http://avahi.sourcearchive.com/documentation/0.6.10/qt-watch_8cpp-source.html>
or
Lua bindings for
lubyk<https://github.com/lubyk/lubyk/tree/master/modules/mdns/core>
(has
both mac and linux code) as examples.

B. You can live with some extra threads:

Use the AvahiThreadedPoll <http://avahi.org/wiki/RunningAvahiClientAsThread> to
run avahi in its own event loop.

C. You use Qt or Glib as event loop: you can use libavahi-qt
and libavahi-glib respectively.

Good luck.

Gaspard

On Tue, Feb 21, 2012 at 1:31 AM, Kirk Reistroffer <kreistroffer-KtrgwEVz+4Y< at >public.gmane.org>wrote:

The reflector option in networks with loops

David Dyja — 2012-03-13T13:10:37

Hello,

I'm doing some testing in a network with multiple routers and loops.

In the simplest form, I have three routers connected in a triangle running
Avahi with the "reflector" option activated.

Everything seems to be working fine, the hosts connected to the three
subnets see each other just fine and there is no packet storm due to the
loop.

My question to the developpers is why isn't there a problem? since the
reflector just forwards the multicast packets and changes the source
address the queries should loop infinitetly, right?

So is there a mechanism to stop the packet storm so it's ok to use the
reflector option in looped networks, or it should be avoided?

Kind regards,
David Dyja

Re: PTR query responses

Markus Rathgeb — 2012-03-12T18:42:39

2012/3/12 Christiano F. Haesbaert <haesbaert-/WJDqAo4zMcBXFe83j6qeQ< at >public.gmane.org>:

I understand the point of view, first get the PTR response, and then
resolve what you want to use.
But it does not explain, why the same avahi versions and the same
configurations does act differently (and normally more than a PTR
record is send).

Re: PTR query responses

Christiano F. Haesbaert — 2012-03-12T17:33:41

Because the PTR response is what takes part in the enumeration phase,
when you get the SRV/TXT/A records, you actually resolving the
services.

In a normal cenario, you don't want every SRV/TXT/A (you don't want to
resolve them all), you want to enumerate all possible services, and
you (the user) picks one, you resolve it (ask for the SRV/TXT/A).

It's been over a year since I implemented mdns/dns-sd and my mind is a bit fusy.

Re: PTR query responses

Markus Rathgeb — 2012-03-12T17:28:57

2012/3/12 Markus Rathgeb <maggu2810-gM/Ye1E23mwN+BqQ9rBEUg< at >public.gmane.org>:

I have had a look at the sending a response packet logic:

===
[reponse-sched.c] static void
send_response_packet(AvahiResponseScheduler *s, AvahiResponseJob *rj)
calls
[reponse-sched.c] static int
packet_add_response_job(AvahiResponseScheduler *s, AvahiDnsPacket *p,
AvahiResponseJob *rj)

[reponse-sched.c] static int
packet_add_response_job(AvahiResponseScheduler *s, AvahiDnsPacket *p,
AvahiResponseJob *rj)
calls
[server.c] void avahi_server_enumerate_aux_records(AvahiServer *s,
AvahiInterface *i, AvahiRecord *r, void (*callback)(AvahiServer *s,
AvahiRecord *r, int flush_cache, void* userdata), void* userdata) {

For class AVAHI_DNS_CLASS_IN and type AVAHI_DNS_TYPE_PTR
the "existing" AVAHI_DNS_TYPE_SRV and AVAHI_DNS_TYPE_SRV entries for
the name should be added.

For class AVAHI_DNS_CLASS_IN and type AVAHI_DNS_TYPE_SRV
the "existing" AVAHI_DNS_TYPE_A and AVAHI_DNS_TYPE_AAAA entries for
the name should be added

...
===

So, why c

Re: PTR query responses

Markus Rathgeb — 2012-03-12T16:25:41

2012/3/12 Christiano F. Haesbaert <haesbaert-/WJDqAo4zMcBXFe83j6qeQ< at >public.gmane.org>:

Thank you for the reply.

avahi-browse shows the service.
avahi-resolve-hostname could resolve the A record.

The windows command line tool dns-sd seems also to be working (-B and
-G v4 hostname).

I got the error with some tools written by colleagues.
So I do not say it is an error of avahi (it could be an error in the
implementation of the tools).

The mutlicast PTR question for the special service is done to get a
list of all machines that are offering that service (no "known answer"
included).
So it is possible to request the A record for every machine that sends
an answer and no A record in the answer was included.

The question is: Why are always the four record types in the answer
and now one machine replies with only one record type in the answer (I
never seen this before and after restart of avahi I see four record
types again for this machine)?
The restart of avahi was done after about four hours.

Re: PTR query responses

Christiano F. Haesbaert — 2012-03-12T16:12:04

There is a feature called something like "Known Answer Suppression",
the querierer includes the known answers in the query packet, under
the AN section, so the the one getting queried won't include the known
answers in the reply.

Not sure if I made myself clear, try using avahi-browse, if the
service ain't there, it's wrong.

PTR query responses

Markus Rathgeb — 2012-03-12T16:08:33

Hello!
I have here multiple machines with the same setup.
Normally if a multicast question for a PTR "_srv._proto.local" is
made, the machines reply with a multicast response containing four
records, the types PTR, TXT, SRV, A.
This morning I see (wireshark) a machine that reply contains only the
PTR record -- missing the TXT, SRV and A type.
If a multicast question for the A record of the machine's name is
requested, the machine answered with a multicast respone given the A
record.
After a restart of the avahi daemon of this machine, the machine's
reply to the PTR question contains all four records (PTR, TXT, SRV, A)
again.
What's going on?
Could someone explain me, how I could run in that situation and
whether it is correct or not (why)?
Thanks in advance,
Markus