Charles,
That seems to solve the problem. Usrloc is now synced at startup. I don't have any duplicate dmq peers anymore as well. I will do some testing but for now the issue seems to be solved.
Many thanks for your quick help.
Jan
Charles Chance schreef op 2019-06-18 11:33:
Hi Jan,
What happens if you set the dmq server address to the advertised address?
listen=udp:172.17.0.2:5063 [1] advertise 10.20.8.88:5063 [2] modparam("dmq", "server_address", "SIP:10.20.8.88:5063 [2]")
Cheers,
Charles
On Tue, 18 Jun 2019 at 07:53, jenus@cyberchaos.nl wrote:
Julien,
I have the following config:
Host ip: 10.20.8.88 Internal container ip: 172.17.0.2
listen=udp:172.17.0.2:5063 [1] advertise 10.20.8.88:5063 [2]
# ----- dmq params ----- modparam("dmq", "server_address", "sip:172.17.0.2:5063 [1]") modparam("dmq", "num_workers", 4) modparam("dmq", "ping_interval", 90) modparam("dmq", "notification_address", "sip:sip12.voip.domain.test:5063") (on the other node this is sip11.domain.test) modparam("dmq", "multi_notify", 1)
# ----- dmq_usrloc params ----- modparam("dmq_usrloc", "enable", 1) modparam("dmq_usrloc", "sync", 1) modparam("dmq_usrloc", "batch_msg_contacts", 50) # 50 contacts / message modparam("dmq_usrloc", "batch_size", 10000) # 10000 contacts /
batch modparam("dmq_usrloc", "batch_usleep", 500000) # one batch every 500ms
Here is the output of the kamcmd dmq.list_nodes
root@9d81434d491f:/# kamcmd dmq.list_nodes { host: 10.20.8.89 port: 5063 resolved_ip: 10.20.8.89 status: active last_notification: 0 local: 0 } { host: 10.20.8.88 port: 5063 resolved_ip: 10.20.8.88 status: active last_notification: 0 local: 0 } { host: 172.17.0.2 port: 5063 resolved_ip: 172.17.0.2 status: active last_notification: 0 local: 1 }
I will collect some traces as well of the sync request as requested by Charles.
Thanks,
Jan
Julien Chavanton schreef op 2019-06-17 21:00:
Can you share the settings you are using :
modparam("dmq", "server_address", "?") modparam("dmq", "notification_address", "?")
On Mon, Jun 17, 2019 at 11:30 AM jenus@cyberchaos.nl wrote:
Hi all,
Is there any other way to filter DMQ peers? I now have a issue
with
dmq_usrloc when running in a docker container. When the node
comes
online the peer is sending the KDMQ usrloc packet (with the registered contacts) to the internal container ip and not the actual
external
ip. Both ip's show up as DMQ peers, after a few keepalive failures
the
dmq peer with the internal container ip becomes inactive and no
updates
are sent to that peer. But the usrloc sync seems to happen before the peer is inactive.
I tried to strip the peers with the internal container ip's, this works for the requests, but i'm not able to fix this for the responses.
Thanks,
Jan
jenus@cyberchaos.nl schreef op 2019-05-02 08:59:
Hi all,
I'm trying to run a kamailio DMQ node inside a docker container.
All
works fine but i have one issue with DMQ. Since the kamailio
process
inside the container is behind a nat i have set the advertise
address
to the external ip, this works fine for the via headers but DMQ
keeps
sending the internal interface in the KDMQ notification_peer
messages.
The external ip is in there as well. Example:
SIP/2.0 200 OK Via: SIP/2.0/UDP
10.10.10.13;branch=z9hG4bK2dcb.c6326de2000000000000000000000000.0
To: <sip:notification_peer@10.10.10.94:5060 [3]
[1]>;tag=616b7acb681ed5e9cbc485b3faa09d62.0a3d
From: <sip:notification_peer@10.10.10.13:5060 [4]
[2]>;tag=e8f29c06c4b85a188f6533f08d60feb6-ff29
CSeq: 10 KDMQ Call-ID: 17042656602ca532-23367@x.x.x.x Content-Type: text/plain Server: kamailio (5.1.4 (x86_64/linux)) Content-Length: 222
sip:10.10.10.13:5060;status=active sip:10.10.10.12:5060;status=active sip:10.10.10.92:5060;status=active sip:10.10.10.93:5060;status=active sip:10.10.10.94:5060;status=active <-- entry for container
external
ip sip:172.17.0.2:5060;status=disabled <-- entry for container
internal
ip
I tried to strip this entry using the replace_body_atonce in the tm:local-request route. This works fine for the KDMQ request,
but
when
i try to remove the same line from 200OK responses that are send
out
as a response to incomming KDMQ messages this seems to fail. I
use
the
following code:
# Working code (for outgoing KDMQ requests): event_route [tm:local-request] { if(is_method("KDMQ") && $rU == "notification_peer") replace_body_atonce("sip:172.17.0.*:5060;status=.*", ""); }
# Failing code (for 200OK responses): reply_route { xlog("L_ALERT", "REPLY: Sending out reply, tU is $tU rs is
$rs\n");
if ($rs == "200" && $tU == "notification_peer") { replace_body_atonce("sip:172.17.0.*:5060;status=.*", ""); } }
I'm running kamailio 5.1.4. Is there a way to strip a line from
the
body of a local generated 200OK response? Or is there a way to
avoid
that DMQ inserts the internal container ip in the first place.
Thanks,
Jan Hazenberg
Kamailio (SER) - Users Mailing List sr-users@lists.kamailio.org https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
Kamailio (SER) - Users Mailing List sr-users@lists.kamailio.org https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
Links:
[1] http://sip:notification_peer@10.10.10.94:5060 [2] http://sip:notification_peer@10.10.10.13:5060 _______________________________________________ Kamailio (SER) - Users Mailing List sr-users@lists.kamailio.org https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
Kamailio (SER) - Users Mailing List sr-users@lists.kamailio.org https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
--
CHARLES CHANCE Managing Director
t. 0330 120 1200 m. 07932 063 891 Sipcentric Ltd. Company registered in England & Wales no. 7365592. Registered office: Faraday Wharf, Innovation Birmingham Campus, Holt Street, Birmingham Science Park, Birmingham B7 4BB.
Links:
[1] http://172.17.0.2:5063 [2] http://10.20.8.88:5063 [3] http://sip:notification_peer@10.10.10.94:5060 [4] http://sip:notification_peer@10.10.10.13:5060 _______________________________________________ Kamailio (SER) - Users Mailing List sr-users@lists.kamailio.org https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users