hi, i have been playing around with ser now for around a month and am getting nowhere.
i have two sites running ser on redhat (7.3 and 8). the ser machines are each behind a redhat (7.3) firewall with port 5060(udp) forwarded to the ser machines.
i cannot see presence of any users at the other site and vice versa.
please can someone guide me through how they would set up such an operation.
this really is a hand-holding job i'm afraid.
very grateful for a quick response.
cheers
Hello,
Do you have record routing enabled ? (You must have rewriteFromRoute() and addRecordRoute() in your script). Please send us as much information as possible (ser configs, ser logs, message dumps etc..)
Jan.
On 16-04 14:40, Mat Harris wrote:
hi, i have been playing around with ser now for around a month and am getting nowhere.
i have two sites running ser on redhat (7.3 and 8). the ser machines are each behind a redhat (7.3) firewall with port 5060(udp) forwarded to the ser machines.
i cannot see presence of any users at the other site and vice versa.
please can someone guide me through how they would set up such an operation.
this really is a hand-holding job i'm afraid.
very grateful for a quick response.
cheers
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
On Wed, Apr 16, 2003 at 03:58:37PM +0200, Jan Janak wrote:
Hello,
Do you have record routing enabled ? (You must have rewriteFromRoute() and addRecordRoute() in your script). Please send us as much information as possible (ser configs, ser logs, message dumps etc..)
Jan.
hi, thanks for the quick response, its great of you to help out :)
ok here's a summary of all the probs.
using msn messenger i can connect to the ser server at my house from my home lan. i cannot connect to the server at work nor can i connect from work to my house.
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops 1: 10.194.217.5 (7.596 ms) SIP/2.0 483 Too Many Hops 2: 10.194.217.5 (11.760 ms) SIP/2.0 483 Too Many Hops 3: 10.194.217.5 (18.771 ms) SIP/2.0 483 Too Many Hops 4: 10.194.217.5 (26.441 ms) SIP/2.0 483 Too Many Hops 5: 10.194.217.5 (33.709 ms) SIP/2.0 483 Too Many Hops 6: 10.194.217.5 (53.847 ms) SIP/2.0 483 Too Many Hops 7: 10.194.217.5 (66.244 ms) SIP/2.0 483 Too Many Hops 8: 10.194.217.5 (76.296 ms) SIP/2.0 483 Too Many Hops 9: 10.194.217.5 (87.163 ms) SIP/2.0 483 Too Many Hops
and that goes on forever. it just times out if i do the -vv option.
all the while this is going on, i get messages like these in the message log:
Apr 16 14:33:56 maiden ./ser[2342]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2342]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2342]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:56 maiden ./ser[2345]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2345]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2345]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:58 maiden ./ser[2343]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:58 maiden ./ser[2343]: ERROR: bad header field Apr 16 14:33:58 maiden ./ser[2343]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed
i have posted my config to my site (http://www.genestate.com/debug/sip/ser.cfg)
i have not taken any special steps as far as firewall routing or anything like that other than forwarding port 5060 as i said previously.
I do not know anything about the config directives you just spoke of so I do not know how to answer you.
i think the beast thing would be for me to start from scratch and do it properly with help from someone who knows more that me as I have probably screwed this attemp :)
thanks again
On Wednesday 16 April 2003 16:09, Mat Harris wrote:
using msn messenger i can connect to the ser server at my house from my home lan. i cannot connect to the server at work nor can i connect from work to my house.
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops 1: 10.194.217.5 (7.596 ms) SIP/2.0 483 Too Many Hops 2: 10.194.217.5 (11.760 ms) SIP/2.0 483 Too Many Hops 3: 10.194.217.5 (18.771 ms) SIP/2.0 483 Too Many Hops 4: 10.194.217.5 (26.441 ms) SIP/2.0 483 Too Many Hops 5: 10.194.217.5 (33.709 ms) SIP/2.0 483 Too Many Hops 6: 10.194.217.5 (53.847 ms) SIP/2.0 483 Too Many Hops 7: 10.194.217.5 (66.244 ms) SIP/2.0 483 Too Many Hops 8: 10.194.217.5 (76.296 ms) SIP/2.0 483 Too Many Hops 9: 10.194.217.5 (87.163 ms) SIP/2.0 483 Too Many Hops
This is an endless loop. Which means their is a problem with the configuration. If i lookup the IP for genestate.com i get 62.78.46.80. If understand it right you just forwards the port 5060 from the public IP to the private IP 10.194.217.5, right? In this case the uri==myself will not work, because Ser will just listen on the private IP (which DNS name this have ever). Because of this every request hits the t_relay, which looks up the destination IP (62.78.46.80) and send it their again. You can try to add 'alias=genstate.com', but generaly you will encounter probably MANY problems with Network Address Tranlastion (NAT). And if you use NAT on both sides of the setup this will probably not work without additional work.
Regards Nils Ohlmeier
and that goes on forever. it just times out if i do the -vv option.
all the while this is going on, i get messages like these in the message log:
Apr 16 14:33:56 maiden ./ser[2342]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2342]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2342]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:56 maiden ./ser[2345]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2345]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2345]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:58 maiden ./ser[2343]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:58 maiden ./ser[2343]: ERROR: bad header field Apr 16 14:33:58 maiden ./ser[2343]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed
i have posted my config to my site (http://www.genestate.com/debug/sip/ser.cfg)
i have not taken any special steps as far as firewall routing or anything like that other than forwarding port 5060 as i said previously.
I do not know anything about the config directives you just spoke of so I do not know how to answer you.
i think the beast thing would be for me to start from scratch and do it properly with help from someone who knows more that me as I have probably screwed this attemp :)
thanks again
Hi All
I am new to Linux as a whole so forgive be if some of this should be obvious
1. I have set my domain in the cfg file but I have not set up the relevant DNS server entries and a reverse DNS on my IP will not yield the domain I have set. I reference the SER using the IP in my SIP device is this OK.
2. Does the above have any consequence for "uri==myself"
3. Setting the Environment variable SIP DOMAIN - is this something I should be doing every time I use serctl or should I be putting it is a start up file somewhere.
4. Lastly what is digest authentication - and when would I need it. (I have set up mysql and persistence)
Thanks, Dinesh
On 16-04 23:19, Dinesh wrote:
Hi All
I am new to Linux as a whole so forgive be if some of this should be obvious
- I have set my domain in the cfg file but I have not set up the
relevant DNS server entries and a reverse DNS on my IP will not yield the domain I have set. I reference the SER using the IP in my SIP device is this OK.
It should work. But I recommend you to use real domain that can be resolved using DNS.
- Does the above have any consequence for "uri==myself"
No, it will handle the IP address.
- Setting the Environment variable SIP DOMAIN - is this something I
should be doing every time I use serctl or should I be putting it is a start up file somewhere.
You can put it anywhere you want.
- Lastly what is digest authentication - and when would I need it. (I
have set up mysql and persistence)
Digest authentication is an authentication mechanism that allows you to verify that a user knows his password and thus you can authorize him -- for example to access a PSTN gateway, make calls using your server and so on.
Jan.
Hi All
I would like to use groups to restrict calling such that a person may only call another person that is in a group that they belong to. A person may however belong to more than one group.
Is there anyway to use the existing groups feature to implement such a restriction.
Thanks, Dinesh
Hello,
yes, you can use existing group membership checks for that.
Jan.
On 23-04 15:50, Dinesh wrote:
Hi All
I would like to use groups to restrict calling such that a person may only call another person that is in a group that they belong to. A person may however belong to more than one group.
Is there anyway to use the existing groups feature to implement such a restriction.
Thanks, Dinesh
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
Hi All
I find that when I start SER from the a terminal, it starts fine and returns to the command prompt.
When I start SER from initd then the system does not continue after starting SER.
Any help Please.
Thanks, Dinesh
======================================================================== ======= STARTED FROM INITD
Apr 23 14:30:14 localhost httpd: httpd startup succeeded Apr 23 14:30:16 localhost ser: Listening on Apr 23 14:30:16 localhost ser: 127.0.0.1 [127.0.0.1]::5060 Apr 23 14:30:16 localhost ser: Aliases: localhost:5060 localhost.localdomain:5060 Apr 23 14:30:16 localhost /usr/sbin/ser[1758]: mod_init(): Database connection opened successfuly Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 14:30:17 localhost /usr/sbin/ser[1775]: INFO: fifo process starting: 1775 Apr 23 14:30:17 localhost /usr/sbin/ser[1775]: SER: open_uac_fifo: fifo server up at /tmp/ser_fifo... Apr 23 14:33:25 localhost ntpd[1692]: kernel time discipline status change 41
SYSTEM DOES NOT COMPLETE THE BOOT
======================================================================== ========
STARTED FROM TERMINAL
Apr 23 15:56:46 localhost ser[1971]: mod_init(): Database connection opened successfuly Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 15:56:46 localhost ser[1989]: INFO: fifo process starting: 1989 Apr 23 15:56:46 localhost ser[1989]: SER: open_uac_fifo: fifo server up at /tmp/ser_fifo...
RETURN TO COMMAND PROMPT - NO PROBLEM ======================================================================== ========
Thanks, Dinesh
Hello,
do you have fork=yes in your config script ?
Jan.
On 23-04 16:02, Dinesh wrote:
Hi All
I find that when I start SER from the a terminal, it starts fine and returns to the command prompt.
When I start SER from initd then the system does not continue after starting SER.
Any help Please.
Thanks, Dinesh
========================================================================
STARTED FROM INITD
Apr 23 14:30:14 localhost httpd: httpd startup succeeded Apr 23 14:30:16 localhost ser: Listening on Apr 23 14:30:16 localhost ser: 127.0.0.1 [127.0.0.1]::5060 Apr 23 14:30:16 localhost ser: Aliases: localhost:5060 localhost.localdomain:5060 Apr 23 14:30:16 localhost /usr/sbin/ser[1758]: mod_init(): Database connection opened successfuly Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 14:30:17 localhost /usr/sbin/ser[1775]: INFO: fifo process starting: 1775 Apr 23 14:30:17 localhost /usr/sbin/ser[1775]: SER: open_uac_fifo: fifo server up at /tmp/ser_fifo... Apr 23 14:33:25 localhost ntpd[1692]: kernel time discipline status change 41
SYSTEM DOES NOT COMPLETE THE BOOT
========================================================================
STARTED FROM TERMINAL
Apr 23 15:56:46 localhost ser[1971]: mod_init(): Database connection opened successfuly Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 15:56:46 localhost ser[1989]: INFO: fifo process starting: 1989 Apr 23 15:56:46 localhost ser[1989]: SER: open_uac_fifo: fifo server up at /tmp/ser_fifo...
RETURN TO COMMAND PROMPT - NO PROBLEM
========
Thanks, Dinesh
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
I have seen the same thing happen on Solaris with other scripts. If you type "exit" on the terminal, it should continue booting. The problem has to do with the startup script (not the ser.cfg). It somehow is unable to exit after it starts the process (in this case SER).
-- Mensaje original -- From: "Dinesh" feedbak@imelhk.com To: serusers@lists.iptel.org Subject: [Serusers] Startup Date: Wed, 23 Apr 2003 16:02:06 +0800
Hi All
I find that when I start SER from the a terminal, it starts fine and returns to the command prompt.
When I start SER from initd then the system does not continue after starting SER.
Any help Please.
Thanks, Dinesh
========================================================================
STARTED FROM INITD
Apr 23 14:30:14 localhost httpd: httpd startup succeeded Apr 23 14:30:16 localhost ser: Listening on Apr 23 14:30:16 localhost ser: 127.0.0.1 [127.0.0.1]::5060 Apr 23 14:30:16 localhost ser: Aliases: localhost:5060 localhost.localdomain:5060 Apr 23 14:30:16 localhost /usr/sbin/ser[1758]: mod_init(): Database connection opened successfuly Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 14:30:17 localhost /usr/sbin/ser[1758]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 14:30:17 localhost /usr/sbin/ser[1775]: INFO: fifo process starting: 1775 Apr 23 14:30:17 localhost /usr/sbin/ser[1775]: SER: open_uac_fifo: fifo server up at /tmp/ser_fifo... Apr 23 14:33:25 localhost ntpd[1692]: kernel time discipline status change 41
SYSTEM DOES NOT COMPLETE THE BOOT
========================================================================
STARTED FROM TERMINAL
Apr 23 15:56:46 localhost ser[1971]: mod_init(): Database connection opened successfuly Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is initially 65535 Apr 23 15:56:46 localhost ser[1971]: INFO: udp_init: SO_RCVBUF is finally 131070 Apr 23 15:56:46 localhost ser[1989]: INFO: fifo process starting: 1989
Apr 23 15:56:46 localhost ser[1989]: SER: open_uac_fifo: fifo server up at /tmp/ser_fifo...
RETURN TO COMMAND PROMPT - NO PROBLEM
========
Thanks, Dinesh
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
________________________________________ Acceso rápido a Internet con Epm.Net http://www.epm.net.co
OK, there is no addRecordRoute in your config script so you will have to add it in config files of BOTH SIP servers. Simply add addRecordRoute() just before line containing if(!t_relay()) {...
Without addRecordRoute() NOTIFY messages will be send directly from one UA to the other and your firewall will block it.
But it will still not handle responses properly. To do that you will have to put public IP of NAT box into Via and into Record-Route headers and this is currently under development, I recommend you to wait for the next release.
Jan.
On 16-04 15:09, Mat Harris wrote:
On Wed, Apr 16, 2003 at 03:58:37PM +0200, Jan Janak wrote:
Hello,
Do you have record routing enabled ? (You must have rewriteFromRoute() and addRecordRoute() in your script). Please send us as much information as possible (ser configs, ser logs, message dumps etc..)
Jan.
hi, thanks for the quick response, its great of you to help out :)
ok here's a summary of all the probs.
using msn messenger i can connect to the ser server at my house from my home lan. i cannot connect to the server at work nor can i connect from work to my house.
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops 1: 10.194.217.5 (7.596 ms) SIP/2.0 483 Too Many Hops 2: 10.194.217.5 (11.760 ms) SIP/2.0 483 Too Many Hops 3: 10.194.217.5 (18.771 ms) SIP/2.0 483 Too Many Hops 4: 10.194.217.5 (26.441 ms) SIP/2.0 483 Too Many Hops 5: 10.194.217.5 (33.709 ms) SIP/2.0 483 Too Many Hops 6: 10.194.217.5 (53.847 ms) SIP/2.0 483 Too Many Hops 7: 10.194.217.5 (66.244 ms) SIP/2.0 483 Too Many Hops 8: 10.194.217.5 (76.296 ms) SIP/2.0 483 Too Many Hops 9: 10.194.217.5 (87.163 ms) SIP/2.0 483 Too Many Hops
and that goes on forever. it just times out if i do the -vv option.
all the while this is going on, i get messages like these in the message log:
Apr 16 14:33:56 maiden ./ser[2342]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2342]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2342]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:56 maiden ./ser[2345]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2345]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2345]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:58 maiden ./ser[2343]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:58 maiden ./ser[2343]: ERROR: bad header field Apr 16 14:33:58 maiden ./ser[2343]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed
i have posted my config to my site (http://www.genestate.com/debug/sip/ser.cfg)
i have not taken any special steps as far as firewall routing or anything like that other than forwarding port 5060 as i said previously.
I do not know anything about the config directives you just spoke of so I do not know how to answer you.
i think the beast thing would be for me to start from scratch and do it properly with help from someone who knows more that me as I have probably screwed this attemp :)
thanks again
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
what about running some sort of sip proxy on the firewalls?
i have seen a few on freshmeat etc, but have no idea yes or no.
On Wed, Apr 16, 2003 at 05:21:32PM +0200, Jan Janak wrote:
OK, there is no addRecordRoute in your config script so you will have to add it in config files of BOTH SIP servers. Simply add addRecordRoute() just before line containing if(!t_relay()) {...
Without addRecordRoute() NOTIFY messages will be send directly from one UA to the other and your firewall will block it.
But it will still not handle responses properly. To do that you will have to put public IP of NAT box into Via and into Record-Route headers and this is currently under development, I recommend you to wait for the next release.
Jan.
On 16-04 15:09, Mat Harris wrote:
On Wed, Apr 16, 2003 at 03:58:37PM +0200, Jan Janak wrote:
Hello,
Do you have record routing enabled ? (You must have rewriteFromRoute() and addRecordRoute() in your script). Please send us as much information as possible (ser configs, ser logs, message dumps etc..)
Jan.
hi, thanks for the quick response, its great of you to help out :)
ok here's a summary of all the probs.
using msn messenger i can connect to the ser server at my house from my home lan. i cannot connect to the server at work nor can i connect from work to my house.
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops 1: 10.194.217.5 (7.596 ms) SIP/2.0 483 Too Many Hops 2: 10.194.217.5 (11.760 ms) SIP/2.0 483 Too Many Hops 3: 10.194.217.5 (18.771 ms) SIP/2.0 483 Too Many Hops 4: 10.194.217.5 (26.441 ms) SIP/2.0 483 Too Many Hops 5: 10.194.217.5 (33.709 ms) SIP/2.0 483 Too Many Hops 6: 10.194.217.5 (53.847 ms) SIP/2.0 483 Too Many Hops 7: 10.194.217.5 (66.244 ms) SIP/2.0 483 Too Many Hops 8: 10.194.217.5 (76.296 ms) SIP/2.0 483 Too Many Hops 9: 10.194.217.5 (87.163 ms) SIP/2.0 483 Too Many Hops
and that goes on forever. it just times out if i do the -vv option.
all the while this is going on, i get messages like these in the message log:
Apr 16 14:33:56 maiden ./ser[2342]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2342]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2342]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:56 maiden ./ser[2345]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2345]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2345]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:58 maiden ./ser[2343]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:58 maiden ./ser[2343]: ERROR: bad header field Apr 16 14:33:58 maiden ./ser[2343]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed
i have posted my config to my site (http://www.genestate.com/debug/sip/ser.cfg)
i have not taken any special steps as far as firewall routing or anything like that other than forwarding port 5060 as i said previously.
I do not know anything about the config directives you just spoke of so I do not know how to answer you.
i think the beast thing would be for me to start from scratch and do it properly with help from someone who knows more that me as I have probably screwed this attemp :)
thanks again
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
If the proxy you mention can put public IP of the NAT int Record-Route and received parameter into Via then it could work, but I am not aware of such a proxy.
Jan.
On 16-04 16:28, Mat Harris wrote:
what about running some sort of sip proxy on the firewalls?
i have seen a few on freshmeat etc, but have no idea yes or no.
On Wed, Apr 16, 2003 at 05:21:32PM +0200, Jan Janak wrote:
OK, there is no addRecordRoute in your config script so you will have to add it in config files of BOTH SIP servers. Simply add addRecordRoute() just before line containing if(!t_relay()) {...
Without addRecordRoute() NOTIFY messages will be send directly from one UA to the other and your firewall will block it.
But it will still not handle responses properly. To do that you will have to put public IP of NAT box into Via and into Record-Route headers and this is currently under development, I recommend you to wait for the next release.
Jan.
On 16-04 15:09, Mat Harris wrote:
On Wed, Apr 16, 2003 at 03:58:37PM +0200, Jan Janak wrote:
Hello,
Do you have record routing enabled ? (You must have rewriteFromRoute() and addRecordRoute() in your script). Please send us as much information as possible (ser configs, ser logs, message dumps etc..)
Jan.
hi, thanks for the quick response, its great of you to help out :)
ok here's a summary of all the probs.
using msn messenger i can connect to the ser server at my house from my home lan. i cannot connect to the server at work nor can i connect from work to my house.
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops 1: 10.194.217.5 (7.596 ms) SIP/2.0 483 Too Many Hops 2: 10.194.217.5 (11.760 ms) SIP/2.0 483 Too Many Hops 3: 10.194.217.5 (18.771 ms) SIP/2.0 483 Too Many Hops 4: 10.194.217.5 (26.441 ms) SIP/2.0 483 Too Many Hops 5: 10.194.217.5 (33.709 ms) SIP/2.0 483 Too Many Hops 6: 10.194.217.5 (53.847 ms) SIP/2.0 483 Too Many Hops 7: 10.194.217.5 (66.244 ms) SIP/2.0 483 Too Many Hops 8: 10.194.217.5 (76.296 ms) SIP/2.0 483 Too Many Hops 9: 10.194.217.5 (87.163 ms) SIP/2.0 483 Too Many Hops
and that goes on forever. it just times out if i do the -vv option.
all the while this is going on, i get messages like these in the message log:
Apr 16 14:33:56 maiden ./ser[2342]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2342]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2342]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:56 maiden ./ser[2345]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2345]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2345]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:58 maiden ./ser[2343]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:58 maiden ./ser[2343]: ERROR: bad header field Apr 16 14:33:58 maiden ./ser[2343]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed
i have posted my config to my site (http://www.genestate.com/debug/sip/ser.cfg)
i have not taken any special steps as far as firewall routing or anything like that other than forwarding port 5060 as i said previously.
I do not know anything about the config directives you just spoke of so I do not know how to answer you.
i think the beast thing would be for me to start from scratch and do it properly with help from someone who knows more that me as I have probably screwed this attemp :)
thanks again
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
fair enough. ok, any idea how long until ser supports this feature?
On Wed, Apr 16, 2003 at 05:38:44PM +0200, Jan Janak wrote:
If the proxy you mention can put public IP of the NAT int Record-Route and received parameter into Via then it could work, but I am not aware of such a proxy.
Jan.
On 16-04 16:28, Mat Harris wrote:
what about running some sort of sip proxy on the firewalls?
i have seen a few on freshmeat etc, but have no idea yes or no.
On Wed, Apr 16, 2003 at 05:21:32PM +0200, Jan Janak wrote:
OK, there is no addRecordRoute in your config script so you will have to add it in config files of BOTH SIP servers. Simply add addRecordRoute() just before line containing if(!t_relay()) {...
Without addRecordRoute() NOTIFY messages will be send directly from one UA to the other and your firewall will block it.
But it will still not handle responses properly. To do that you will have to put public IP of NAT box into Via and into Record-Route headers and this is currently under development, I recommend you to wait for the next release.
Jan.
On 16-04 15:09, Mat Harris wrote:
On Wed, Apr 16, 2003 at 03:58:37PM +0200, Jan Janak wrote:
Hello,
Do you have record routing enabled ? (You must have rewriteFromRoute() and addRecordRoute() in your script). Please send us as much information as possible (ser configs, ser logs, message dumps etc..)
Jan.
hi, thanks for the quick response, its great of you to help out :)
ok here's a summary of all the probs.
using msn messenger i can connect to the ser server at my house from my home lan. i cannot connect to the server at work nor can i connect from work to my house.
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops 1: 10.194.217.5 (7.596 ms) SIP/2.0 483 Too Many Hops 2: 10.194.217.5 (11.760 ms) SIP/2.0 483 Too Many Hops 3: 10.194.217.5 (18.771 ms) SIP/2.0 483 Too Many Hops 4: 10.194.217.5 (26.441 ms) SIP/2.0 483 Too Many Hops 5: 10.194.217.5 (33.709 ms) SIP/2.0 483 Too Many Hops 6: 10.194.217.5 (53.847 ms) SIP/2.0 483 Too Many Hops 7: 10.194.217.5 (66.244 ms) SIP/2.0 483 Too Many Hops 8: 10.194.217.5 (76.296 ms) SIP/2.0 483 Too Many Hops 9: 10.194.217.5 (87.163 ms) SIP/2.0 483 Too Many Hops
and that goes on forever. it just times out if i do the -vv option.
all the while this is going on, i get messages like these in the message log:
Apr 16 14:33:56 maiden ./ser[2342]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2342]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2342]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:56 maiden ./ser[2345]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2345]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2345]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:58 maiden ./ser[2343]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:58 maiden ./ser[2343]: ERROR: bad header field Apr 16 14:33:58 maiden ./ser[2343]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed
i have posted my config to my site (http://www.genestate.com/debug/sip/ser.cfg)
i have not taken any special steps as far as firewall routing or anything like that other than forwarding port 5060 as i said previously.
I do not know anything about the config directives you just spoke of so I do not know how to answer you.
i think the beast thing would be for me to start from scratch and do it properly with help from someone who knows more that me as I have probably screwed this attemp :)
thanks again
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
It is already in the CVS but only partially tested.
Jan.
On 16-04 16:43, Mat Harris wrote:
fair enough. ok, any idea how long until ser supports this feature?
On Wed, Apr 16, 2003 at 05:38:44PM +0200, Jan Janak wrote:
If the proxy you mention can put public IP of the NAT int Record-Route and received parameter into Via then it could work, but I am not aware of such a proxy.
Jan.
On 16-04 16:28, Mat Harris wrote:
what about running some sort of sip proxy on the firewalls?
i have seen a few on freshmeat etc, but have no idea yes or no.
On Wed, Apr 16, 2003 at 05:21:32PM +0200, Jan Janak wrote:
OK, there is no addRecordRoute in your config script so you will have to add it in config files of BOTH SIP servers. Simply add addRecordRoute() just before line containing if(!t_relay()) {...
Without addRecordRoute() NOTIFY messages will be send directly from one UA to the other and your firewall will block it.
But it will still not handle responses properly. To do that you will have to put public IP of NAT box into Via and into Record-Route headers and this is currently under development, I recommend you to wait for the next release.
Jan.
On 16-04 15:09, Mat Harris wrote:
On Wed, Apr 16, 2003 at 03:58:37PM +0200, Jan Janak wrote:
Hello,
Do you have record routing enabled ? (You must have rewriteFromRoute() and addRecordRoute() in your script). Please send us as much information as possible (ser configs, ser logs, message dumps etc..)
Jan.
hi, thanks for the quick response, its great of you to help out :)
ok here's a summary of all the probs.
using msn messenger i can connect to the ser server at my house from my home lan. i cannot connect to the server at work nor can i connect from work to my house.
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops 1: 10.194.217.5 (7.596 ms) SIP/2.0 483 Too Many Hops 2: 10.194.217.5 (11.760 ms) SIP/2.0 483 Too Many Hops 3: 10.194.217.5 (18.771 ms) SIP/2.0 483 Too Many Hops 4: 10.194.217.5 (26.441 ms) SIP/2.0 483 Too Many Hops 5: 10.194.217.5 (33.709 ms) SIP/2.0 483 Too Many Hops 6: 10.194.217.5 (53.847 ms) SIP/2.0 483 Too Many Hops 7: 10.194.217.5 (66.244 ms) SIP/2.0 483 Too Many Hops 8: 10.194.217.5 (76.296 ms) SIP/2.0 483 Too Many Hops 9: 10.194.217.5 (87.163 ms) SIP/2.0 483 Too Many Hops
and that goes on forever. it just times out if i do the -vv option.
all the while this is going on, i get messages like these in the message log:
Apr 16 14:33:56 maiden ./ser[2342]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2342]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2342]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:56 maiden ./ser[2345]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:56 maiden ./ser[2345]: ERROR: bad header field Apr 16 14:33:56 maiden ./ser[2345]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed Apr 16 14:33:58 maiden ./ser[2343]: ERROR: get_hdr_field: bad body for <User-Agent>(4194304) Apr 16 14:33:58 maiden ./ser[2343]: ERROR: bad header field Apr 16 14:33:58 maiden ./ser[2343]: ERROR: build_res_buf_from_sip_req: alas, parse_headers failed
i have posted my config to my site (http://www.genestate.com/debug/sip/ser.cfg)
i have not taken any special steps as far as firewall routing or anything like that other than forwarding port 5060 as i said previously.
I do not know anything about the config directives you just spoke of so I do not know how to answer you.
i think the beast thing would be for me to start from scratch and do it properly with help from someone who knows more that me as I have probably screwed this attemp :)
thanks again
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
-- Mat Harris OpenGPG Public Key ID: C37D57D9 mat.harris@genestate.com www.genestate.com
At 04:09 PM 4/16/2003, Mat Harris wrote:
if i use sipsack from the server itself i get
[3:03pm]maiden:~% /usr/local/bin/sipsak-0.8.1/sipsak -T -s \ sip:matthewh@genestate.com warning: IP extract from warning activated to be more informational 0: 10.194.217.5 (0.972 ms) SIP/2.0 483 Too Many Hops
http://www.iptel.org/ser/doc/seruser-html/x878.html#AEN883
-Jiri
Hello, usually the NOTIFYs are sent directly to contact address which might advertise a different port than standard 5060. If the request has to pass the firewall and the port is blocked then the message is dropped. I hope this helps.
Best regards, Daniel
On 4/16/2003 3:40 PM, Mat Harris wrote:
hi, i have been playing around with ser now for around a month and am getting nowhere.
i have two sites running ser on redhat (7.3 and 8). the ser machines are each behind a redhat (7.3) firewall with port 5060(udp) forwarded to the ser machines.
i cannot see presence of any users at the other site and vice versa.
please can someone guide me through how they would set up such an operation.
this really is a hand-holding job i'm afraid.
very grateful for a quick response.
cheers
On Wed, Apr 16, 2003 at 04:06:21 +0200, Daniel-Constantin MIERLA wrote:
Hello, usually the NOTIFYs are sent directly to contact address which might advertise a different port than standard 5060. If the request has to pass the firewall and the port is blocked then the message is dropped. I hope this helps.
Best regards, Daniel
hi, i thought this may have been a firewall issue as i am not using the FCP(?) or anything so any requests using other reports can get out, but not back in
Hello, On 4/16/2003 4:10 PM, Mat Harris wrote:
On Wed, Apr 16, 2003 at 04:06:21 +0200, Daniel-Constantin MIERLA wrote:
Hello, usually the NOTIFYs are sent directly to contact address which might advertise a different port than standard 5060. If the request has to pass the firewall and the port is blocked then the message is dropped. I hope this helps.
Best regards, Daniel
hi, i thought this may have been a firewall issue as i am not using the FCP(?) or anything so any requests using other reports can get out, but not back in
What about the second firewall? It allows requests on other ports to go behind?
Best regards, Daniel
-
Daniel-Constantin MIERLA -
Dinesh -
Jan Janak -
Jiri Kuthan -
Mat Harris -
Nils Ohlmeier -
ricvil＠epm.net.co