16 Mar
2005
16 Mar
'05
5:04 p.m.
How can we restrict the registration to only one user per login?.
Currently, you can register with the same login/password in different
machines at the same time. But what we require is that only the first
registration is valid and the rest ones are invalidated by the SER.
Thanks
16 Mar
16 Mar
5:11 p.m.
Hi Jffdsaklj! (must be a hard life living with such a name)
And how will you distinguish between an invalid registration and a valid
re-registration from a new IP address?
regards,
klaus
Jffdsaklj dsjk wrote:
How can we restrict the registration to only one user
per login?.
Currently, you can register with the same login/password in different
machines at the same time. But what we require is that only the first
registration is valid and the rest ones are invalidated by the SER.
Thanks
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
22 Mar
22 Mar
10:37 a.m.
Finally I found a solution. I've attached the block I included in the
SER cfg file.
The problem (big problem) is that after a day (more or less), there
are defunct processes. I think it's a problem of the exec_msg
sentence, but I'm not sure. With these defunct processes, SER crashed.
:(
######################################################################
# This block of code controls the user registration
if (method=="REGISTER") {
# Enable authentication
if (!www_authorize("galeno.hi.inet","subscriber")) {
www_challenge("galeno.hi.inet","0");
break;
};
####
# This block prevents a user from registering with a login
previously registered from another machine
#
# Verify it's not a deregistration message
if (!search("Expires: 0")) {
# It's a registration message
# Verify that there's no register in the table location with the
same user but from a differente machine
if (
!exec_msg('
USUARIO=`echo $SIP_HF_FROM | sed -n "s/<sip:\(.*\)@.*/\1/p"`;
QUERY="select username from location where
username=\"$USUARIO\" and contact not like
\"sip:$SIP_SRCIP:%\"";
RESULT=`mysql -uXXX -pXXX -e "$QUERY" ser`;
if [ -z "$RESULT" ]; then exit 0; else echo "The user
$USUARIO is already registered" >> /var/log/serprueba.log; exit 1;
fi;')
) {
# The user is already registered
sl_send_reply("404", "Not Found");
break;
};
}
####
save("location");
break;
};
#####################################################################
On Wed, 16 Mar 2005 17:11:01 +0100, Klaus Darilion
<klaus.mailinglists(a)pernau.at> wrote:
Hi Jffdsaklj! (must be a hard life living with such a
name)
And how will you distinguish between an invalid registration and a valid
re-registration from a new IP address?
regards,
klaus
Jffdsaklj dsjk wrote:
How can we restrict the registration to only one
user per login?.
Currently, you can register with the same login/password in different
machines at the same time. But what we require is that only the first
registration is valid and the rest ones are invalidated by the SER.
Thanks
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
12:49 p.m.
FYI, exec_msg is highly not recommended because it is "expensive" in
terms of processing.
Regards,
Paul
On Tue, 22 Mar 2005 10:37:58 +0100, Jffdsaklj dsjk <jjjaime(a)gmail.com> wrote:
Finally I found a solution. I've attached the
block I included in the
SER cfg file.
The problem (big problem) is that after a day (more or less), there
are defunct processes. I think it's a problem of the exec_msg
sentence, but I'm not sure. With these defunct processes, SER crashed.
:(
######################################################################
# This block of code controls the user registration
if (method=="REGISTER") {
# Enable authentication
if (!www_authorize("galeno.hi.inet","subscriber")) {
www_challenge("galeno.hi.inet","0");
break;
};
####
# This block prevents a user from registering with a login
previously registered from another machine
#
# Verify it's not a deregistration message
if (!search("Expires: 0")) {
# It's a registration message
# Verify that there's no register in the table location with the
same user but from a differente machine
if (
!exec_msg('
USUARIO=`echo $SIP_HF_FROM | sed -n
"s/<sip:\(.*\)@.*/\1/p"`;
QUERY="select username from location where
username=\"$USUARIO\" and contact not like
\"sip:$SIP_SRCIP:%\"";
RESULT=`mysql -uXXX -pXXX -e "$QUERY" ser`;
if [ -z "$RESULT" ]; then exit 0; else echo "The user
$USUARIO is already registered" >> /var/log/serprueba.log; exit 1;
fi;')
) {
# The user is already registered
sl_send_reply("404", "Not Found");
break;
};
}
####
save("location");
break;
};
#####################################################################
On Wed, 16 Mar 2005 17:11:01 +0100, Klaus Darilion
<klaus.mailinglists(a)pernau.at> wrote:
Hi Jffdsaklj! (must be a hard life living with
such a name)
And how will you distinguish between an invalid registration and a valid
re-registration from a new IP address?
regards,
klaus
Jffdsaklj dsjk wrote:
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
How can we restrict the registration to only one
user per login?.
Currently, you can register with the same login/password in different
machines at the same time. But what we require is that only the first
registration is valid and the rest ones are invalidated by the SER.
Thanks
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
23 Mar
23 Mar
10:11 a.m.
Write a SER module if you wish the code to be fast. -jiri
At 03:16 PM 3/22/2005, Dana Olson wrote:
Right, but what alternative is there?
On Tue, 22 Mar 2005 06:49:08 -0500, Java Rockx <javarockx(a)gmail.com> wrote:
--
Jiri Kuthan http://iptel.org/~jiri/
FYI, exec_msg is highly not recommended because
it is "expensive" in
terms of processing.
Regards,
Paul
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
9:03 a.m.
So, what do you recommend us to use instead of exec_msg?. What's the
best alternative?.
On Tue, 22 Mar 2005 06:49:08 -0500, Java Rockx <javarockx(a)gmail.com> wrote:
FYI, exec_msg is highly not recommended because it is
"expensive" in
terms of processing.
Regards,
Paul
On Tue, 22 Mar 2005 10:37:58 +0100, Jffdsaklj dsjk <jjjaime(a)gmail.com> wrote:
Finally I found a solution. I've attached the
block I included in the
SER cfg file.
The problem (big problem) is that after a day (more or less), there
are defunct processes. I think it's a problem of the exec_msg
sentence, but I'm not sure. With these defunct processes, SER crashed.
:(
######################################################################
# This block of code controls the user registration
if (method=="REGISTER") {
# Enable authentication
if (!www_authorize("galeno.hi.inet","subscriber")) {
www_challenge("galeno.hi.inet","0");
break;
};
####
# This block prevents a user from registering with a login
previously registered from another machine
#
# Verify it's not a deregistration message
if (!search("Expires: 0")) {
# It's a registration message
# Verify that there's no register in the table location with the
same user but from a differente machine
if (
!exec_msg('
USUARIO=`echo $SIP_HF_FROM | sed -n
"s/<sip:\(.*\)@.*/\1/p"`;
QUERY="select username from location where
username=\"$USUARIO\" and contact not like
\"sip:$SIP_SRCIP:%\"";
RESULT=`mysql -uXXX -pXXX -e "$QUERY" ser`;
if [ -z "$RESULT" ]; then exit 0; else echo "The user
$USUARIO is already registered" >> /var/log/serprueba.log; exit 1;
fi;')
) {
# The user is already registered
sl_send_reply("404", "Not Found");
break;
};
}
####
save("location");
break;
};
#####################################################################
On Wed, 16 Mar 2005 17:11:01 +0100, Klaus Darilion
<klaus.mailinglists(a)pernau.at> wrote:
Hi Jffdsaklj! (must be a hard life living with
such a name)
And how will you distinguish between an invalid registration and a valid
re-registration from a new IP address?
regards,
klaus
Jffdsaklj dsjk wrote:
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
How can we restrict the registration to only one
user per login?.
Currently, you can register with the same login/password in different
machines at the same time. But what we require is that only the first
registration is valid and the rest ones are invalidated by the SER.
Thanks
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
2:09 p.m.
Hi Paul,
You are right - maybe a better approach will be to use the registered()
function - it returns true if the AOR in the Request-URI is registered.
To do the trick for REGISTER requests, use avpops so copy the From or To
uri into RURI before checking - then you can to revert_uri() to restore
to original value,
I haven't tried it , it's just an idea.
Best regards,
Marian
Java Rockx wrote:
FYI, exec_msg is highly not recommended because it is
"expensive" in
terms of processing.
Regards,
Paul
On Tue, 22 Mar 2005 10:37:58 +0100, Jffdsaklj dsjk <jjjaime(a)gmail.com> wrote:
>Finally I found a solution. I've attached the block I included in the
>SER cfg file.
>The problem (big problem) is that after a day (more or less), there
>are defunct processes. I think it's a problem of the exec_msg
>sentence, but I'm not sure. With these defunct processes, SER crashed.
>:(
>
> ######################################################################
> # This block of code controls the user registration
> if (method=="REGISTER") {
>
> # Enable authentication
> if (!www_authorize("galeno.hi.inet","subscriber")) {
> www_challenge("galeno.hi.inet","0");
> break;
> };
>
> ####
> # This block prevents a user from registering with a login
>previously registered from another machine
> #
> # Verify it's not a deregistration message
> if (!search("Expires: 0")) {
> # It's a registration message
> # Verify that there's no register in the table location with the
>same user but from a differente machine
> if (
> !exec_msg('
> USUARIO=`echo $SIP_HF_FROM | sed -n
"s/<sip:\(.*\)@.*/\1/p"`;
> QUERY="select username from location where
>username=\"$USUARIO\" and contact not like
\"sip:$SIP_SRCIP:%\"";
> RESULT=`mysql -uXXX -pXXX -e "$QUERY" ser`;
> if [ -z "$RESULT" ]; then exit 0; else echo "The user
>$USUARIO is already registered" >> /var/log/serprueba.log; exit 1;
>fi;')
> ) {
> # The user is already registered
> sl_send_reply("404", "Not Found");
> break;
> };
> }
> ####
>
> save("location");
> break;
> };
> #####################################################################
>
>On Wed, 16 Mar 2005 17:11:01 +0100, Klaus Darilion
><klaus.mailinglists(a)pernau.at> wrote:
>
>>Hi Jffdsaklj! (must be a hard life living with such a name)
>>
>>And how will you distinguish between an invalid registration and a valid
>>re-registration from a new IP address?
>>
>>regards,
>>klaus
>>
>>Jffdsaklj dsjk wrote:
>>
>>>How can we restrict the registration to only one user per login?.
>>>Currently, you can register with the same login/password in different
>>>machines at the same time. But what we require is that only the first
>>>registration is valid and the rest ones are invalidated by the SER.
>>>Thanks
--
Voice System
http://www.voice-system.ro
2:58 p.m.
avpops is only in CVS, right? CVS also has max_contacts for the
registrar module, which kinda does the same thing. Stable has no such
option to my knowledge.
--
Dana
On Wed, 23 Mar 2005 14:09:22 +0100, Marian Dumitru
<marian.dumitru(a)voice-sistem.ro> wrote:
Hi Paul,
You are right - maybe a better approach will be to use the registered()
function - it returns true if the AOR in the Request-URI is registered.
To do the trick for REGISTER requests, use avpops so copy the From or To
uri into RURI before checking - then you can to revert_uri() to restore
to original value,
I haven't tried it , it's just an idea.
Best regards,
Marian
Java Rockx wrote:
FYI, exec_msg is highly not recommended because
it is "expensive" in
terms of processing.
Regards,
Paul
On Tue, 22 Mar 2005 10:37:58 +0100, Jffdsaklj dsjk <jjjaime(a)gmail.com> wrote:
>Finally I found a solution. I've attached the block I included in the
>SER cfg file.
>The problem (big problem) is that after a day (more or less), there
>are defunct processes. I think it's a problem of the exec_msg
>sentence, but I'm not sure. With these defunct processes, SER crashed.
>:(
>
> ######################################################################
> # This block of code controls the user registration
> if (method=="REGISTER") {
>
> # Enable authentication
> if (!www_authorize("galeno.hi.inet","subscriber")) {
> www_challenge("galeno.hi.inet","0");
> break;
> };
>
> ####
> # This block prevents a user from registering with a login
>previously registered from another machine
> #
> # Verify it's not a deregistration message
> if (!search("Expires: 0")) {
> # It's a registration message
> # Verify that there's no register in the table location with the
>same user but from a differente machine
> if (
> !exec_msg('
> USUARIO=`echo $SIP_HF_FROM | sed -n
"s/<sip:\(.*\)@.*/\1/p"`;
> QUERY="select username from location where
>username=\"$USUARIO\" and contact not like
\"sip:$SIP_SRCIP:%\"";
> RESULT=`mysql -uXXX -pXXX -e "$QUERY" ser`;
> if [ -z "$RESULT" ]; then exit 0; else echo "The user
>$USUARIO is already registered" >> /var/log/serprueba.log; exit 1;
>fi;')
> ) {
> # The user is already registered
> sl_send_reply("404", "Not Found");
> break;
> };
> }
> ####
>
> save("location");
> break;
> };
> #####################################################################
>
>On Wed, 16 Mar 2005 17:11:01 +0100, Klaus Darilion
><klaus.mailinglists(a)pernau.at> wrote:
>
>>Hi Jffdsaklj! (must be a hard life living with such a name)
>>
>>And how will you distinguish between an invalid registration and a valid
>>re-registration from a new IP address?
>>
>>regards,
>>klaus
>>
>>Jffdsaklj dsjk wrote:
>>
>>>How can we restrict the registration to only one user per login?.
>>>Currently, you can register with the same login/password in different
>>>machines at the same time. But what we require is that only the first
>>>registration is valid and the rest ones are invalidated by the SER.
>>>Thanks
--
Voice System
http://www.voice-system.ro
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
3:14 p.m.
IMHO setting max_contacts is very dangerous because if you reboot your
SIP UA and your NAT device gives you a new port assignment, the you
will __fail__ to register if your previous contact AOR has not
expired.
I think this is the hardest part of implementing a "per-seat"
configuration. You just don't know when a NAT will assign a new port
to a SIP UA.
I gave up on this for this very reason. Our solution is to lock down
the SIP UAs and never let a customer get in to the settings (except
for the LAN/WAN settings).
Regards,
Paul
On Wed, 23 Mar 2005 08:58:09 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
avpops is only in CVS, right? CVS also has
max_contacts for the
registrar module, which kinda does the same thing. Stable has no such
option to my knowledge.
--
Dana
On Wed, 23 Mar 2005 14:09:22 +0100, Marian Dumitru
<marian.dumitru(a)voice-sistem.ro> wrote:
Hi Paul,
You are right - maybe a better approach will be to use the registered()
function - it returns true if the AOR in the Request-URI is registered.
To do the trick for REGISTER requests, use avpops so copy the From or To
uri into RURI before checking - then you can to revert_uri() to restore
to original value,
I haven't tried it , it's just an idea.
Best regards,
Marian
Java Rockx wrote:
FYI, exec_msg is highly not recommended because
it is "expensive" in
terms of processing.
Regards,
Paul
On Tue, 22 Mar 2005 10:37:58 +0100, Jffdsaklj dsjk <jjjaime(a)gmail.com> wrote:
>Finally I found a solution. I've attached the block I included in the
>SER cfg file.
>The problem (big problem) is that after a day (more or less), there
>are defunct processes. I think it's a problem of the exec_msg
>sentence, but I'm not sure. With these defunct processes, SER crashed.
>:(
>
> ######################################################################
> # This block of code controls the user registration
> if (method=="REGISTER") {
>
> # Enable authentication
> if (!www_authorize("galeno.hi.inet","subscriber")) {
> www_challenge("galeno.hi.inet","0");
> break;
> };
>
> ####
> # This block prevents a user from registering with a login
>previously registered from another machine
> #
> # Verify it's not a deregistration message
> if (!search("Expires: 0")) {
> # It's a registration message
> # Verify that there's no register in the table location with the
>same user but from a differente machine
> if (
> !exec_msg('
> USUARIO=`echo $SIP_HF_FROM | sed -n
"s/<sip:\(.*\)@.*/\1/p"`;
> QUERY="select username from location where
>username=\"$USUARIO\" and contact not like
\"sip:$SIP_SRCIP:%\"";
> RESULT=`mysql -uXXX -pXXX -e "$QUERY" ser`;
> if [ -z "$RESULT" ]; then exit 0; else echo "The user
>$USUARIO is already registered" >> /var/log/serprueba.log; exit 1;
>fi;')
> ) {
> # The user is already registered
> sl_send_reply("404", "Not Found");
> break;
> };
> }
> ####
>
> save("location");
> break;
> };
> #####################################################################
>
>On Wed, 16 Mar 2005 17:11:01 +0100, Klaus Darilion
><klaus.mailinglists(a)pernau.at> wrote:
>
>>Hi Jffdsaklj! (must be a hard life living with such a name)
>>
>>And how will you distinguish between an invalid registration and a valid
>>re-registration from a new IP address?
>>
>>regards,
>>klaus
>>
>>Jffdsaklj dsjk wrote:
>>
>>>How can we restrict the registration to only one user per login?.
>>>Currently, you can register with the same login/password in different
>>>machines at the same time. But what we require is that only the first
>>>registration is valid and the rest ones are invalidated by the SER.
>>>Thanks
--
Voice System
http://www.voice-system.ro
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
4:05 p.m.
On Wed, 23 Mar 2005 09:14:42 -0500, Java Rockx <javarockx(a)gmail.com> wrote:
IMHO setting max_contacts is very dangerous because if
you reboot your
SIP UA and your NAT device gives you a new port assignment, the you
will __fail__ to register if your previous contact AOR has not
expired.
I think this is the hardest part of implementing a "per-seat"
configuration. You just don't know when a NAT will assign a new port
to a SIP UA.
I gave up on this for this very reason. Our solution is to lock down
the SIP UAs and never let a customer get in to the settings (except
for the LAN/WAN settings).
Regards,
Paul
Hey Paul, I hear you. I'm just saying that max_contacts is basically
what we're attempting to accomplish, only in the stable banch of SER.
And I've basically got it down to the point where the only time it
locks me out is if A) someone is already logged in with that account,
or B) my IP address changes. I also wrote a real quick CGI that will
list all the users logged in, and if you click on any of them, it'll
force them out of the location table, allowing someone else to login
(or re-login, if the IP changed).
However, if my superiors find this to not be an acceptable solution,
I'd like to know what you are using, and how did you lock it down? We
tried with SJphone and had some success, but the quality just doesn't
compare to eyeBeam. We managed to lockdown eyeBeam a bit, but users
can still get into the settings. I'd appreciate your insight.
--
Dana
4:16 p.m.
We're shipping UTstarcom iAN-02EX ATAs and these are fully remote
configurable as they query our Apache farm for their configuration
upon bootup (based on their MAC address).
So we have full control of the customer IAD, and the customer doesn't.
For softphones --- this is bad. Xten (eyebeam) AFAIK, do not have a
way to remotely configure them and therefore you have to give the SIP
proxy UID and PWD to the customer and we decided under no
circumstances will this be done on our network.
The solution for a softphone is to find one that will remotely
configure like the UTstarcom IAD. We're not shipping softphones so I
can't really help you there.
Regards,
Paul
On Wed, 23 Mar 2005 10:05:26 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
On Wed, 23 Mar 2005 09:14:42 -0500, Java Rockx
<javarockx(a)gmail.com> wrote:
IMHO setting max_contacts is very dangerous
because if you reboot your
SIP UA and your NAT device gives you a new port assignment, the you
will __fail__ to register if your previous contact AOR has not
expired.
I think this is the hardest part of implementing a "per-seat"
configuration. You just don't know when a NAT will assign a new port
to a SIP UA.
I gave up on this for this very reason. Our solution is to lock down
the SIP UAs and never let a customer get in to the settings (except
for the LAN/WAN settings).
Regards,
Paul
Hey Paul, I hear you. I'm just saying that max_contacts is basically
what we're attempting to accomplish, only in the stable banch of SER.
And I've basically got it down to the point where the only time it
locks me out is if A) someone is already logged in with that account,
or B) my IP address changes. I also wrote a real quick CGI that will
list all the users logged in, and if you click on any of them, it'll
force them out of the location table, allowing someone else to login
(or re-login, if the IP changed).
However, if my superiors find this to not be an acceptable solution,
I'd like to know what you are using, and how did you lock it down? We
tried with SJphone and had some success, but the quality just doesn't
compare to eyeBeam. We managed to lockdown eyeBeam a bit, but users
can still get into the settings. I'd appreciate your insight.
--
Dana
5:54 p.m.
Hi
I never tried max_contacts, but would it not be possible to logoff the
previous contact b4 the new one registers, that way you wont have a
locking issue.
This is the kind of thing I guess yahoo IM etc all do...kind of
Iqbal
PS is max contacts in CVS head, or in 0.9 also
On 3/23/2005, "Java Rockx" <javarockx(a)gmail.com> wrote:
We're shipping UTstarcom iAN-02EX ATAs and these
are fully remote
configurable as they query our Apache farm for their configuration
upon bootup (based on their MAC address).
So we have full control of the customer IAD, and the customer doesn't.
For softphones --- this is bad. Xten (eyebeam) AFAIK, do not have a
way to remotely configure them and therefore you have to give the SIP
proxy UID and PWD to the customer and we decided under no
circumstances will this be done on our network.
The solution for a softphone is to find one that will remotely
configure like the UTstarcom IAD. We're not shipping softphones so I
can't really help you there.
Regards,
Paul
On Wed, 23 Mar 2005 10:05:26 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
On Wed, 23 Mar 2005 09:14:42 -0500, Java Rockx
<javarockx(a)gmail.com> wrote:
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
IMHO setting max_contacts is very dangerous
because if you reboot your
SIP UA and your NAT device gives you a new port assignment, the you
will __fail__ to register if your previous contact AOR has not
expired.
I think this is the hardest part of implementing a "per-seat"
configuration. You just don't know when a NAT will assign a new port
to a SIP UA.
I gave up on this for this very reason. Our solution is to lock down
the SIP UAs and never let a customer get in to the settings (except
for the LAN/WAN settings).
Regards,
Paul
Hey Paul, I hear you. I'm just saying that max_contacts is basically
what we're attempting to accomplish, only in the stable banch of SER.
And I've basically got it down to the point where the only time it
locks me out is if A) someone is already logged in with that account,
or B) my IP address changes. I also wrote a real quick CGI that will
list all the users logged in, and if you click on any of them, it'll
force them out of the location table, allowing someone else to login
(or re-login, if the IP changed).
However, if my superiors find this to not be an acceptable solution,
I'd like to know what you are using, and how did you lock it down? We
tried with SJphone and had some success, but the quality just doesn't
compare to eyeBeam. We managed to lockdown eyeBeam a bit, but users
can still get into the settings. I'd appreciate your insight.
--
Dana
8:29 p.m.
As far as I knew, 0.9.0 wasn't officially released as stable yet. Am I wrong?
And yes, what you said to prevent the locking issue is exactly
correct. But my scripting does this on 0.8.14 as well. The only time
it will lock is if an IP address changes and then a reregister request
is sent without unregistering first. I don't even know if that is a
possibility, but I assume it is.
Ideally, we would be able to know if it's the same phone
re-registering, but you can't tell the difference if it's the same
person with a new IP, or, for example, if you open up task manager,
kill the softphone (which means its registration doesn't force
expire/unregister), change IP addresses, and then log in again.
My script does this in its current state. It will only reject if the
registration hasn't expired and a new IP address is trying to
register. I'm not sure, but I assume, this is also what happens with
max_contacts.
--
Dana
On Wed, 23 Mar 2005 16:54:21 GMT, Iqbal <iqbal(a)gigo.co.uk> wrote:
Hi
I never tried max_contacts, but would it not be possible to logoff the
previous contact b4 the new one registers, that way you wont have a
locking issue.
This is the kind of thing I guess yahoo IM etc all do...kind of
Iqbal
PS is max contacts in CVS head, or in 0.9 also
On 3/23/2005, "Java Rockx" <javarockx(a)gmail.com> wrote:
>We're shipping UTstarcom iAN-02EX ATAs and these are fully remote
>configurable as they query our Apache farm for their configuration
>upon bootup (based on their MAC address).
>
>So we have full control of the customer IAD, and the customer doesn't.
>
>For softphones --- this is bad. Xten (eyebeam) AFAIK, do not have a
>way to remotely configure them and therefore you have to give the SIP
>proxy UID and PWD to the customer and we decided under no
>circumstances will this be done on our network.
>
>The solution for a softphone is to find one that will remotely
>configure like the UTstarcom IAD. We're not shipping softphones so I
>can't really help you there.
>
>Regards,
>Paul
>
>
>On Wed, 23 Mar 2005 10:05:26 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
>> On Wed, 23 Mar 2005 09:14:42 -0500, Java Rockx <javarockx(a)gmail.com>
wrote:
>> > IMHO setting max_contacts is very dangerous because if you reboot your
>> > SIP UA and your NAT device gives you a new port assignment, the you
>> > will __fail__ to register if your previous contact AOR has not
>> > expired.
>> >
>> > I think this is the hardest part of implementing a "per-seat"
>> > configuration. You just don't know when a NAT will assign a new port
>> > to a SIP UA.
>> >
>> > I gave up on this for this very reason. Our solution is to lock down
>> > the SIP UAs and never let a customer get in to the settings (except
>> > for the LAN/WAN settings).
>> >
>> > Regards,
>> > Paul
>>
>> Hey Paul, I hear you. I'm just saying that max_contacts is basically
>> what we're attempting to accomplish, only in the stable banch of SER.
>> And I've basically got it down to the point where the only time it
>> locks me out is if A) someone is already logged in with that account,
>> or B) my IP address changes. I also wrote a real quick CGI that will
>> list all the users logged in, and if you click on any of them, it'll
>> force them out of the location table, allowing someone else to login
>> (or re-login, if the IP changed).
>>
>> However, if my superiors find this to not be an acceptable solution,
>> I'd like to know what you are using, and how did you lock it down? We
>> tried with SJphone and had some success, but the quality just doesn't
>> compare to eyeBeam. We managed to lockdown eyeBeam a bit, but users
>> can still get into the settings. I'd appreciate your insight.
>> --
>> Dana
8:37 p.m.
It would be really nice if somehow the SIP UA's MAC address could be
discovered. If that were possible you could really do per-seat
connections.
Regards,
Paul
On Wed, 23 Mar 2005 14:29:37 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
As far as I knew, 0.9.0 wasn't officially released
as stable yet. Am I wrong?
And yes, what you said to prevent the locking issue is exactly
correct. But my scripting does this on 0.8.14 as well. The only time
it will lock is if an IP address changes and then a reregister request
is sent without unregistering first. I don't even know if that is a
possibility, but I assume it is.
Ideally, we would be able to know if it's the same phone
re-registering, but you can't tell the difference if it's the same
person with a new IP, or, for example, if you open up task manager,
kill the softphone (which means its registration doesn't force
expire/unregister), change IP addresses, and then log in again.
My script does this in its current state. It will only reject if the
registration hasn't expired and a new IP address is trying to
register. I'm not sure, but I assume, this is also what happens with
max_contacts.
--
Dana
On Wed, 23 Mar 2005 16:54:21 GMT, Iqbal <iqbal(a)gigo.co.uk> wrote:
Hi
I never tried max_contacts, but would it not be possible to logoff the
previous contact b4 the new one registers, that way you wont have a
locking issue.
This is the kind of thing I guess yahoo IM etc all do...kind of
Iqbal
PS is max contacts in CVS head, or in 0.9 also
On 3/23/2005, "Java Rockx" <javarockx(a)gmail.com> wrote:
>We're shipping UTstarcom iAN-02EX ATAs and these are fully remote
>configurable as they query our Apache farm for their configuration
>upon bootup (based on their MAC address).
>
>So we have full control of the customer IAD, and the customer doesn't.
>
>For softphones --- this is bad. Xten (eyebeam) AFAIK, do not have a
>way to remotely configure them and therefore you have to give the SIP
>proxy UID and PWD to the customer and we decided under no
>circumstances will this be done on our network.
>
>The solution for a softphone is to find one that will remotely
>configure like the UTstarcom IAD. We're not shipping softphones so I
>can't really help you there.
>
>Regards,
>Paul
>
>
>On Wed, 23 Mar 2005 10:05:26 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
>> On Wed, 23 Mar 2005 09:14:42 -0500, Java Rockx <javarockx(a)gmail.com>
wrote:
>> > IMHO setting max_contacts is very dangerous because if you reboot your
>> > SIP UA and your NAT device gives you a new port assignment, the you
>> > will __fail__ to register if your previous contact AOR has not
>> > expired.
>> >
>> > I think this is the hardest part of implementing a "per-seat"
>> > configuration. You just don't know when a NAT will assign a new port
>> > to a SIP UA.
>> >
>> > I gave up on this for this very reason. Our solution is to lock down
>> > the SIP UAs and never let a customer get in to the settings (except
>> > for the LAN/WAN settings).
>> >
>> > Regards,
>> > Paul
>>
>> Hey Paul, I hear you. I'm just saying that max_contacts is basically
>> what we're attempting to accomplish, only in the stable banch of SER.
>> And I've basically got it down to the point where the only time it
>> locks me out is if A) someone is already logged in with that account,
>> or B) my IP address changes. I also wrote a real quick CGI that will
>> list all the users logged in, and if you click on any of them, it'll
>> force them out of the location table, allowing someone else to login
>> (or re-login, if the IP changed).
>>
>> However, if my superiors find this to not be an acceptable solution,
>> I'd like to know what you are using, and how did you lock it down? We
>> tried with SJphone and had some success, but the quality just doesn't
>> compare to eyeBeam. We managed to lockdown eyeBeam a bit, but users
>> can still get into the settings. I'd appreciate your insight.
>> --
>> Dana
9:28 p.m.
I thought of this too, and I think it really is the only way to do it. *sigh*
I suppose that this could be done if we develop our own custom
softphone, but I doubt we will do that (not entirely true, we're
considering it, but it won't be immediately deployable either way).
Ah well, I'm close enough I think. :)
--
Dana
On Wed, 23 Mar 2005 14:37:29 -0500, Java Rockx <javarockx(a)gmail.com> wrote:
It would be really nice if somehow the SIP UA's
MAC address could be
discovered. If that were possible you could really do per-seat
connections.
Regards,
Paul
On Wed, 23 Mar 2005 14:29:37 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
> As far as I knew, 0.9.0 wasn't officially released as stable yet. Am I wrong?
>
> And yes, what you said to prevent the locking issue is exactly
> correct. But my scripting does this on 0.8.14 as well. The only time
> it will lock is if an IP address changes and then a reregister request
> is sent without unregistering first. I don't even know if that is a
> possibility, but I assume it is.
>
> Ideally, we would be able to know if it's the same phone
> re-registering, but you can't tell the difference if it's the same
> person with a new IP, or, for example, if you open up task manager,
> kill the softphone (which means its registration doesn't force
> expire/unregister), change IP addresses, and then log in again.
>
> My script does this in its current state. It will only reject if the
> registration hasn't expired and a new IP address is trying to
> register. I'm not sure, but I assume, this is also what happens with
> max_contacts.
> --
> Dana
>
> On Wed, 23 Mar 2005 16:54:21 GMT, Iqbal <iqbal(a)gigo.co.uk> wrote:
> >
> > Hi
> >
> > I never tried max_contacts, but would it not be possible to logoff the
> > previous contact b4 the new one registers, that way you wont have a
> > locking issue.
> >
> > This is the kind of thing I guess yahoo IM etc all do...kind of
> >
> > Iqbal
> >
> > PS is max contacts in CVS head, or in 0.9 also
> >
> > On 3/23/2005, "Java Rockx" <javarockx(a)gmail.com> wrote:
> >
> > >We're shipping UTstarcom iAN-02EX ATAs and these are fully remote
> > >configurable as they query our Apache farm for their configuration
> > >upon bootup (based on their MAC address).
> > >
> > >So we have full control of the customer IAD, and the customer doesn't.
> > >
> > >For softphones --- this is bad. Xten (eyebeam) AFAIK, do not have a
> > >way to remotely configure them and therefore you have to give the SIP
> > >proxy UID and PWD to the customer and we decided under no
> > >circumstances will this be done on our network.
> > >
> > >The solution for a softphone is to find one that will remotely
> > >configure like the UTstarcom IAD. We're not shipping softphones so I
> > >can't really help you there.
> > >
> > >Regards,
> > >Paul
> > >
> > >
> > >On Wed, 23 Mar 2005 10:05:26 -0500, Dana Olson <rickaster(a)gmail.com>
wrote:
> > >> On Wed, 23 Mar 2005 09:14:42 -0500, Java Rockx
<javarockx(a)gmail.com> wrote:
> > >> > IMHO setting max_contacts is very dangerous because if you reboot
your
> > >> > SIP UA and your NAT device gives you a new port assignment, the
you
> > >> > will __fail__ to register if your previous contact AOR has not
> > >> > expired.
> > >> >
> > >> > I think this is the hardest part of implementing a
"per-seat"
> > >> > configuration. You just don't know when a NAT will assign a
new port
> > >> > to a SIP UA.
> > >> >
> > >> > I gave up on this for this very reason. Our solution is to lock
down
> > >> > the SIP UAs and never let a customer get in to the settings
(except
> > >> > for the LAN/WAN settings).
> > >> >
> > >> > Regards,
> > >> > Paul
> > >>
> > >> Hey Paul, I hear you. I'm just saying that max_contacts is
basically
> > >> what we're attempting to accomplish, only in the stable banch of
SER.
> > >> And I've basically got it down to the point where the only time
it
> > >> locks me out is if A) someone is already logged in with that account,
> > >> or B) my IP address changes. I also wrote a real quick CGI that will
> > >> list all the users logged in, and if you click on any of them,
it'll
> > >> force them out of the location table, allowing someone else to login
> > >> (or re-login, if the IP changed).
> > >>
> > >> However, if my superiors find this to not be an acceptable solution,
> > >> I'd like to know what you are using, and how did you lock it down?
We
> > >> tried with SJphone and had some success, but the quality just
doesn't
> > >> compare to eyeBeam. We managed to lockdown eyeBeam a bit, but users
> > >> can still get into the settings. I'd appreciate your insight.
> > >> --
> > >> Dana
24 Mar
24 Mar
2:48 p.m.
that would make life alot easier and more billable :-)
Iqbal
Java Rockx wrote:
It would be really nice if somehow the SIP UA's MAC
address could be
discovered. If that were possible you could really do per-seat
connections.
Regards,
Paul
On Wed, 23 Mar 2005 14:29:37 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
As far as I knew, 0.9.0 wasn't officially
released as stable yet. Am I wrong?
And yes, what you said to prevent the locking issue is exactly
correct. But my scripting does this on 0.8.14 as well. The only time
it will lock is if an IP address changes and then a reregister request
is sent without unregistering first. I don't even know if that is a
possibility, but I assume it is.
Ideally, we would be able to know if it's the same phone
re-registering, but you can't tell the difference if it's the same
person with a new IP, or, for example, if you open up task manager,
kill the softphone (which means its registration doesn't force
expire/unregister), change IP addresses, and then log in again.
My script does this in its current state. It will only reject if the
registration hasn't expired and a new IP address is trying to
register. I'm not sure, but I assume, this is also what happens with
max_contacts.
--
Dana
On Wed, 23 Mar 2005 16:54:21 GMT, Iqbal <iqbal(a)gigo.co.uk> wrote:
>Hi
>
>I never tried max_contacts, but would it not be possible to logoff the
>previous contact b4 the new one registers, that way you wont have a
>locking issue.
>
>This is the kind of thing I guess yahoo IM etc all do...kind of
>
>Iqbal
>
>PS is max contacts in CVS head, or in 0.9 also
>
>On 3/23/2005, "Java Rockx" <javarockx(a)gmail.com> wrote:
>
>
>
>>We're shipping UTstarcom iAN-02EX ATAs and these are fully remote
>>configurable as they query our Apache farm for their configuration
>>upon bootup (based on their MAC address).
>>
>>So we have full control of the customer IAD, and the customer doesn't.
>>
>>For softphones --- this is bad. Xten (eyebeam) AFAIK, do not have a
>>way to remotely configure them and therefore you have to give the SIP
>>proxy UID and PWD to the customer and we decided under no
>>circumstances will this be done on our network.
>>
>>The solution for a softphone is to find one that will remotely
>>configure like the UTstarcom IAD. We're not shipping softphones so I
>>can't really help you there.
>>
>>Regards,
>>Paul
>>
>>
>>On Wed, 23 Mar 2005 10:05:26 -0500, Dana Olson <rickaster(a)gmail.com> wrote:
>>
>>
>>>On Wed, 23 Mar 2005 09:14:42 -0500, Java Rockx <javarockx(a)gmail.com>
wrote:
>>>
>>>
>>>>IMHO setting max_contacts is very dangerous because if you reboot your
>>>>SIP UA and your NAT device gives you a new port assignment, the you
>>>>will __fail__ to register if your previous contact AOR has not
>>>>expired.
>>>>
>>>>I think this is the hardest part of implementing a "per-seat"
>>>>configuration. You just don't know when a NAT will assign a new port
>>>>to a SIP UA.
>>>>
>>>>I gave up on this for this very reason. Our solution is to lock down
>>>>the SIP UAs and never let a customer get in to the settings (except
>>>>for the LAN/WAN settings).
>>>>
>>>>Regards,
>>>>Paul
>>>>
>>>>
>>>Hey Paul, I hear you. I'm just saying that max_contacts is basically
>>>what we're attempting to accomplish, only in the stable banch of SER.
>>>And I've basically got it down to the point where the only time it
>>>locks me out is if A) someone is already logged in with that account,
>>>or B) my IP address changes. I also wrote a real quick CGI that will
>>>list all the users logged in, and if you click on any of them, it'll
>>>force them out of the location table, allowing someone else to login
>>>(or re-login, if the IP changed).
>>>
>>>However, if my superiors find this to not be an acceptable solution,
>>>I'd like to know what you are using, and how did you lock it down? We
>>>tried with SJphone and had some success, but the quality just doesn't
>>>compare to eyeBeam. We managed to lockdown eyeBeam a bit, but users
>>>can still get into the settings. I'd appreciate your insight.
>>>--
>>>Dana
>>>
>>>
.
22 Mar
22 Mar
2:40 p.m.
Hi there
isn't there a way on which a new registration would
override the previous one? this way, the cases of
re-registering (changed IPs) would be solved and
anyone that wish to force two simultaenous registerings
would have some hard time?
Cheers
!3runo
Jffdsaklj dsjk wrote:
Finally I found a solution. I've attached the
block I included in the
SER cfg file.
The problem (big problem) is that after a day (more or less), there
are defunct processes. I think it's a problem of the exec_msg
sentence, but I'm not sure. With these defunct processes, SER crashed.
:(
[...]
3:20 p.m.
I'm pretty sure that the default is this, and if not, you can
certainly do it with stateless forwarding, which I am using. However,
this is kinda defeating the goal I'm trying to achieve. If I'm logged
into my account, and someone else logs in with my account after me, I
stop receiving calls, and they go to the other person until they log
out or I re-register. That's not good. I need to give the priority to
the person who logged in first, and reject further registrations. I'm
sure that someone can use your methodology though.
--
Dana
On Tue, 22 Mar 2005 10:40:41 -0300, Bruno Lopes F. Cabral
<bruno(a)openline.com.br> wrote:
Hi there
isn't there a way on which a new registration would
override the previous one? this way, the cases of
re-registering (changed IPs) would be solved and
anyone that wish to force two simultaenous registerings
would have some hard time?
Cheers
!3runo
3:34 p.m.
Hi
if you are 1 person only and your telefone is registered,
supposedely no one else will have your id/password to try
to register a second time. if they do so, you will complain
(as you described) and the admin will know your id/password
was stolen (and perhaps provide you a new one)
I suppose this is the desire of any admin wishing to limit
SIP registrations to 1 user only, to avoid having two customers
sharing a single account.
please note that, in second case, if you change IP and
re-register, as you are only one person/account, the old
register should expire imediately
my excuses if I didn't myself clearer early
Cheers
!3runo
Dana Olson wrote:
I'm pretty sure that the default is this, and if
not, you can
certainly do it with stateless forwarding, which I am using. However,
this is kinda defeating the goal I'm trying to achieve. If I'm logged
into my account, and someone else logs in with my account after me, I
stop receiving calls, and they go to the other person until they log
out or I re-register. That's not good. I need to give the priority to
the person who logged in first, and reject further registrations. I'm
sure that someone can use your methodology though.
--
Dana
On Tue, 22 Mar 2005 10:40:41 -0300, Bruno Lopes F. Cabral wrote
>isn't there a way on which a new registration would
>override the previous one? this way, the cases of
>re-registering (changed IPs) would be solved and
>anyone that wish to force two simultaenous registerings
>would have some hard time?
4:37 p.m.
hi everybody,
I am using ser with dbtext database
The code is working,
the authentification users is ok
but when i add news users, i have a pb, they are not loaded in the suscriber.
My question is : can i add users in the suscriber table (dbtext)
without restart ser
Thank you.
#
# $Id: ser.cfg,v 1.21.4.1 2003/11/10 15:35:15 andrei Exp $
#
# simple quick-start config script
#
# ----------- global configuration parameters ------------------------
debug=7 # debug level (cmd line: -dddddddddd)
fork=yes
log_stderror=yes
check_via=no # (cmd. line: -v)
dns=no # (cmd. line: -r)
rev_dns=no # (cmd. line: -R)
#port=5060
#children=4
fifo="/tmp/ser_fifo"
# ------------------ module loading ----------------------------------
loadmodule "/usr/local/lib/ser/modules/dbtext.so"
loadmodule "/usr/local/lib/ser/modules/sl.so"
loadmodule "/usr/local/lib/ser/modules/tm.so"
loadmodule "/usr/local/lib/ser/modules/rr.so"
loadmodule "/usr/local/lib/ser/modules/maxfwd.so"
loadmodule "/usr/local/lib/ser/modules/usrloc.so"
loadmodule "/usr/local/lib/ser/modules/registrar.so"
# Uncomment this if you want digest authentication
loadmodule "/usr/local/lib/ser/modules/auth.so"
loadmodule "/usr/local/lib/ser/modules/auth_db.so"
# ----------------- setting module-specific parameters ---------------
# -- usrloc params --
#################################################################
modparam("usrloc", "db_mode",2)
modparam("usrloc", "db_url", "/usr/local/etc/ser/dbtext")
# -- auth params --
#
modparam("auth_db", "db_url", "/usr/local/etc/ser/dbtext")
modparam("auth_db", "calculate_ha1", 1)
modparam("auth_db", "password_column", "password")
################################################################
# -- registrar params --
##########################################################
#modparam("registrar", "nat_flag", 6)
#modparam("registrar", "max_expires", 1200)
#modparam("registrar", "min_expires", 300)
#modparam("registrar", "default_expires", 1200)
#modparam("registrar", "use_domain", 1)
#modparam("registrar","desc_time_order",1)
# -- rr params --
# add value to ;lr param to make some broken UAs happy
modparam("rr", "enable_full_lr", 1)
# ------------------------- request routing logic -------------------
# main routing logic
route{
# initial sanity checks -- messages with
# max_forwards==0, or excessively long requests
if (!mf_process_maxfwd_header("10")) {
sl_send_reply("483","Too Many Hops");
break;
};
if ( msg:len > max_len ) {
sl_send_reply("513", "Message too big");
break;
};
# we record-route all messages -- to make sure that
# subsequent messages will go through our proxy; that's
# particularly good if upstream and downstream entities
# use different transport protocol
record_route();
# loose-route processing
if (loose_route()) {
t_relay();
break;
};
# if the request is for other domain use UsrLoc
# (in case, it does not work, use the following command
# with proper names and addresses in it)
if (uri==myself) {
if (method=="REGISTER") {
# Uncomment this if you want to use digest authentication
# if (!www_authorize("softiphone.net", "subscriber")) {
# www_challenge("softiphone.net", "0");
# break;
# };
if (!proxy_authorize("", "subscriber")) {
proxy_challenge("", "0");
break;
}
save("location");
break;
};
# native SIP destinations are handled using our USRLOC DB
if (!lookup("location")) {
sl_send_reply("404", "Not Found");
break;
};
};
# forward to current uri now; use stateful forwarding; that
# works reliably even if we forward from TCP to UDP
if (!t_relay()) {
sl_reply_error();
};
}
On Tue, 22 Mar 2005 11:34:30 -0300, Bruno Lopes F. Cabral
<bruno(a)openline.com.br> wrote:
Hi
if you are 1 person only and your telefone is registered,
supposedely no one else will have your id/password to try
to register a second time. if they do so, you will complain
(as you described) and the admin will know your id/password
was stolen (and perhaps provide you a new one)
I suppose this is the desire of any admin wishing to limit
SIP registrations to 1 user only, to avoid having two customers
sharing a single account.
please note that, in second case, if you change IP and
re-register, as you are only one person/account, the old
register should expire imediately
my excuses if I didn't myself clearer early
Cheers
!3runo
Dana Olson wrote:
--
Karim Oumalou
I'm pretty sure that the default is this, and
if not, you can
certainly do it with stateless forwarding, which I am using. However,
this is kinda defeating the goal I'm trying to achieve. If I'm logged
into my account, and someone else logs in with my account after me, I
stop receiving calls, and they go to the other person until they log
out or I re-register. That's not good. I need to give the priority to
the person who logged in first, and reject further registrations. I'm
sure that someone can use your methodology though.
--
Dana
On Tue, 22 Mar 2005 10:40:41 -0300, Bruno Lopes F. Cabral wrote
>isn't there a way on which a new registration would
>override the previous one? this way, the cases of
>re-registering (changed IPs) would be solved and
>anyone that wish to force two simultaenous registerings
>would have some hard time?
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
16 Mar
16 Mar
5:55 p.m.
I was asking this same sort of thing last week...
I have set this up using the stable version of SER, but yes, I am
running into the problem that Klaus has mentioned. I haven't had time
to debug any further, but the basic concept works... Just not good
enough for production use as of yet.
--
Dana
On Wed, 16 Mar 2005 17:04:30 +0100, Jffdsaklj dsjk <jjjaime(a)gmail.com> wrote:
How can we restrict the registration to only one user
per login?.
Currently, you can register with the same login/password in different
machines at the same time. But what we require is that only the first
registration is valid and the rest ones are invalidated by the SER.
Thanks
7192
days inactive
7200
days old
21 comments
9 participants
participants (9)
-
Bruno Lopes F. Cabral -
Dana Olson -
Iqbal -
Java Rockx -
Jffdsaklj dsjk -
Jiri Kuthan -
karim Oumalou -
Klaus Darilion -
Marian Dumitru