<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2900.2604" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV>Hi Tina,</DIV>
<DIV> </DIV>
<DIV>> I enjoy reading your posts, thanks a lot for the great work you
are<BR>> doing for ser users! <BR></DIV>
<DIV>Thanks :-)</DIV>
<DIV> <BR>> I am going to consider that kind of scenario as well (load
balancer<BR>> returning Virtual IP address). I think for such configuration
server<BR>> side has to keep session information and always forward a call to
the<BR>> right server. <BR></DIV>
<DIV>I believe the load balancer can be in front and transparantly
translate addresses in the SIP messages:</DIV>
<DIV>User agent always sees a.b.c.d</DIV>
<DIV>
|</DIV>
<DIV>Load balancer with public IP a.b.c.d</DIV>
<DIV>
|
|</DIV>
<DIV>ser1 (10.0.0.2) ser2
(10.0.0.3)</DIV>
<DIV> </DIV>
<DIV>
<DIV>Paul: You have a similar load-balancing solution, right? Do you use a
commerical load balancer that you can reveal the name of? :-)</DIV></DIV>
<DIV> </DIV>
<DIV>The load balancer need to understand SIP (sort of) and use something (ex.
IP address of user agent) to select SER server to send to and rewrite incoming
to 10.0.0.2 and 10.0.0.3 respectively and outgoing to a.b.c.d. ser1 and
ser2 will happily believe they are alone and communicate directly with user
agent using their own private addresses. Both should know about all users
(usrloc). Paul just recently posted a small patch for a setup where they
use two-way replication between two mysql servers and where each SER instance
will save all REGISTERs it receives from user agents, but only save to memory
those replicated from the other SER peer. Klaus pointed out that the same can be
done by not doing mysql replication, but have one DB for each SER and just use
t_replicate() and save_noreply() (for two servers, for more than two you need to
use forward_tcp()).</DIV>
<DIV> </DIV>
<DIV>I have heard complaints that this approach will generate too much
network traffic. Also, t_replicate() does not guarantee delivery of the
replication (forward_tcp is better, but not fool proof). So, the "best"
solution would probably be to hope that Paul's recent TCP/UDP patch for
network access to FIFO commands get included in the CVS. Then, a new replication
module should use XMLRPC or whatever protocol the patch is using to forward (in
a guaranteed mode) the usrloc information, which then would be injected directly
into SER's memory, just like save() would, and of course in whatever
db_mode you would like. The replicaton module would have to keep a queue
(of course, flushed to a DB or something) in case of network problems, so that
the replication would be done (sooner or later) regardless of problems.</DIV>
<DIV> Of course, this is a major undertaking, so for most
replication scenarioes, I would think the forward_tcp() + save_noreply() would
work just fine.</DIV>
<DIV> </DIV>
<DIV>g-)</DIV>
<DIV><BR>> "Greger V. Teigre" <greger@teigre.com> wrote:<BR>> I was
thinking about a load balancing scenario where the load<BR>> balancer will
replace the IP addresses. <BR>> g-)<BR>> <BR>> ---- Original Message
----<BR>> From: Tina<BR>> To: Greger V. Teigre<BR>> Sent: Wednesday,
April 06, 2005 09:45 PM<BR>> Subject: Re: [Serusers] still no help - usrloc
synchronization<BR>> <BR>>> Thank you for givingme the scenario with
"restricted IP" NAT, I am<BR>>> starting to find some acceptable
solution.<BR>>> Unfortunately, "one-public-IP" approach is not free from
problems<BR>>> also. If your SIP router inserts this "one-public-IP" into
the VIA<BR>>> header, the reply routing goes via wrong SIP
server...<BR>>> If your SIP server inserts its real-IP-address - the
scenario<BR>>> mentioned above is still not resolved.<BR>>> Any
comments?<BR>>> Tina<BR>>> <BR>>> "Greger V. Teigre"
<greger@teigre.com> wrote:<BR>>> See inline.<BR>>>
<BR>>>> If you use DNS server for load balancing... the client receives
one<BR>>>> of your domain IP addresses according to SRV. I don't see
the<BR>>>> problem <BR>>>> with a call here, cause UAC asks t!
he address only once (before<BR>>>> sending INVITE). UAC already has
the IP for BYE/reINVITEs. So why<BR>>>> would you replicate
INVITEs?<BR>>> <BR>>> I would never replicate INVITEs, I would just
make sure that they are<BR>>> proxied through the correct SER server (i.e.
IP).<BR>>> <BR>>> The problems depends on your setup. If you have
SERs with different<BR>>> IPs, ex UA1 has registered with server A
and UA2 has registered with<BR>>> server B: If UA2 wants to call UA1 and
UA is behind an IP restricted<BR>>> NAT, server A is stored in the NAT
table of the NAT in front of UA1.<BR>>> If server B sends an INVITE to
UA1, the INVITE will be refused by<BR>>> UA1's
NAT.<BR>>> This is why a "one public IP" in front
of a load balancing<BR>>> cluster probably is a good way to
go.<BR>>> <BR>>>> If you use IPVS/LVS... I believe you can force
SER to insert it's<BR>>>> public IP into VIA,! so there is no problem
with replies. With<BR>>>> regard <BR>>>> to another requests,
I believe load balancer keeps connection<BR>>>> template, then when
another request comes it would be forwarded to<BR>>>> the same
ser.<BR>>> <BR>>> Yes. There are different "keys" to use to load
balance SIP messages.<BR>>> One good way from a NAT point of view is to
use originating IP<BR>>> address. What you must remember is that the
problem is not on the<BR>>> server side, but on the client side. The
NAT will in many situations<BR>>> stop incoming UDP packets if the
originating ip:port is not already<BR>>> stored in the NAT table.
The Via header does not matter for the NAT.<BR>>> g-)<BR>>>
<BR>>>> Any comments?<BR>>>> <BR>>>> "Greger V.
Teigre" <greger@teigre.com> wrote:<BR>>>> Yes, I believe that is
so. But still you get a problem if the NAT is<BR>>>> restricted,
port-restricted or symmetric... The best would be to<BR>>>>
load>> balance and always make sure that a given client is
handled<BR>>>> through a given <BR>>>> SER (REGISTER and
INVITEs). That includes forwarding INVITEs from<BR>>>> one
<BR>>>> SER to<BR>>>> another... OR you must load balance in
front of your servers with<BR>>>> one <BR>>>>
common<BR>>>> public IP.<BR>>>> g-)<BR>>>>
<BR>>>> Matt Schulte wrote:<BR>>>>> Ack, I didn't even
think about NAT. Would these be added before it<BR>>>>> gets sent
off to the second proxy? ie:<BR>>>>> <BR>>>>> if
(!src_ip==blah.netlogic.net) {<BR>>>>>
add_rcv_param();<BR>>>>> t_replicate("blah.netlogic.net",
"999");<BR>>>>> };<BR>>>>> <BR>>>>>
-----Original Message-----<BR>>>>> From: Greger V. Teigre
[mailto:greger@teigre.com]<BR>>>>> Sent: Tuesday, April 05, 2005
7:49 AM<BR>>>>> To: Matt Schulte;
kramarv@yahoo.com<BR>>>>> ! ; Cc:
serusers@lists.iptel.org<BR>>>>> Subject: ! Re: [Serusers] still no help -
usrloc synchronization<BR>>>>> <BR>>>>>
<BR>>>>> Well, you still have the NAT issues unless you do load
balancing<BR>>>>> and your<BR>>>>> SER servers have the
same public IP.<BR>>>>> Have you looked at 0.9.0 nathelper function
add_rcv_param() ? It<BR>>>>> will add received info to the contact
header for the other SER to<BR>>>>> process. Haven't really tried
yet...<BR>>>>> g-)<BR>>>>> <BR>>>>> Matt
Schulte wrote:<BR>>>>>> I'm starting to lean this direction,
using t_replicate and all. I<BR>>>>>> could never get usrloc (db
mode) to function properly..<BR>>>>>> t_replicate
is<BR>>>>> <BR>>>>>> a dirty but very effective
workaround.<BR>>>>>> <BR>>>>>> -----Original
Message-----<BR>>>>>> From: Greger V. Teigre
[mailto:greger@teigre.com]<BR>>>>>> Sent: Saturday, April 02,!
2005 1:33 AM<BR>>>>>> To:
kramarv@yahoo.com<BR>>>>>> Cc:
serusers@lists.iptel.org<BR>>>>>> Subject: Re: [Serusers] still ! no
help - usrloc synchronization<BR>>>>>> <BR>>>>>>
<BR>>>>>> Have a look at this thread:<BR>>>>>>
http://lists.iptel.org/pipermail/serusers/2005-January/014669.html<BR>>>>>>
g-)<BR>>>>>> <BR>>>>>> Java Rockx
wrote:<BR>>>>>>> Tina,<BR>>>>>>>
<BR>>>>>>> I thought I saw you post the other day that you did
not want to<BR>>>>>>> use t_replicate(), however, this is
probably your best bet to<BR>>>>>>> getting
this<BR>>>>>> <BR>>>>>>> to work,
IMHO.<BR>>>>>>> <BR>>>>>>>
Regards,<BR>>>>>>> Paul<BR>>>>>>>
<BR>>>>>>> On Apr 1, 2005 4:08 PM, Tina
wrote:<BR>>>>>> ! >><BR>>>>>>> ! ;>
Hi, please help me, I'm stuck with it!!!!!<BR>>>>>>>> I am
trying to set up several sers with a shared MySQL
database<BR>>>>>>>> for location
service.<BR>>>>>>>> <BR>>>>>>>> I set
in each ser.cfg:<BR>>>>>>>>
<BR>>>>>>>> modparam("usrloc", "db_mode",
2)<BR>>>>>>>>
modparam("usrloc",<BR>>>>>>>>
"db_url","sql://ser:heslo@192.168.25.163/ser")<BR>>>>>>>>
<BR>>>>>>>> and the servers are not
synchronized.<BR>>>>>>>> The I
set<BR>>>>>>>> modparam("usrloc", "db_mode",
2)<BR>>>>>>>> <BR>>>>>>>>
<BR>>>>>>>> made UAC (Xlite) register to one of the
servers.<BR>>>>>>>> I see it via usrloc, but there is no
record in "location" mySQL<BR>>>>>>>> table....So others do
not see the client and I'm unable to make<BR>>>>>>>>
calls....<BR>>>>>>>> <BR>>>>>>>>
<BR>>>>>>>> Please help how to work with usrloc and
mySQL...<BR>>>>>>>> <BR>>>>>>>>
Tina,<BR>>>>>>>> software
engineer<BR>>>>>>>> <BR>>>>>>>>
________________________________<BR>>>>>>>> Do you
Yahoo!?<BR>>>>>>>> Better first dates. More second dates.
Yahoo! Personals<BR>>>>>>>>
<BR>>>>>>>> <BR>>>>>>>>
_______________________________________________<BR>>>>>>>>
Serusers mailing list<BR>>>>>>>>
serusers@lists.iptel.org<BR>>>>>>>>
http://lists.iptel.org/mailman/listinfo/serusers<BR>>>>>>>>
<BR>>>>>>>> <BR>>>>>>>>
<BR>>>>>>> <BR>>>>>>>
_______________________________________________<BR>>>>>>>
Serusers mailing list<BR>>>>>>>
serusers@lists.iptel.org<BR>>>>>>>
http://lists.iptel.org/mailman/listinfo/serusers <BR>>>>>>
<BR>>>>>>
_______________________________________________<BR>>>>>> Serusers
mailing list<BR>>>>>> serusers@lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers<BR>>>> <BR>>>>
<BR>>>> <BR>>>> <BR>>>> Yahoo!
Messenger<BR>>>> Show us what our next emoticon should look like. Join
the fun.<BR>>> <BR>>> <BR>>> Do you Yahoo!?<BR>>> Better
first dates. More second dates. Yahoo! Personals<BR>> <BR>> <BR>>
Yahoo! Messenger<BR>> Show us what our next emoticon should look like. Join
the fun.</DIV></BODY></HTML>