Hi,

I hope I managed to get your report here. If so, take a look at the "no-dns-failover" option when doing the t_relay(). So you can instruct OpenSIPS not to do the automatic DNS based failover and give you full control via failure route.

Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
  https://www.opensips-solutions.com
  https://www.siphub.com

On 03.06.2025 03:41, nz deals wrote:
Is there anyone who has seen this issue? Seems like a bug to me.

Thanks.

Regards,
Jason


On Sun, 1 Jun 2025 at 06:06, Ben Newlin <ben.new...@genesys.com> wrote:

    Oh sorry I missed that in your email. I thought you were trying to
    avoid the failover.

    Dropping the auth info on the DNS failover I don’t think is
    expected, since a DNS failover doesn’t trigger failure_route so
    you can’t add it back.

    I’d recommend opening a bug for this on the Github, but maybe
    someone else has ideas.

    Ben Newlin

    *From: *Users <users-boun...@lists.opensips.org> on behalf of nz
    deals <nzdealsh...@gmail.com>
    *Date: *Friday, May 30, 2025 at 10:49 PM
    *To: *OpenSIPS users mailling list <users@lists.opensips.org>
    *Subject: *Re: [OpenSIPS-Users] Issue with proxy failover and
    uac_auth()

    * EXTERNAL EMAIL - Please use caution with links and attachments *

    ------------------------------------------------------------------------

    Thank you for your response.

    The problem is, opensips sends the INVITE to secondary srv (failed
    over) without Authorization. It makes sense that the dns failover
    is not managed by opensips but atleast the same INVITE should be
    failover to the secondary. Why the Authorization is removed when
    it goes to the secondary.

    Thanks

    On Sat, 31 May 2025 at 03:52, Ben Newlin <ben.new...@genesys.com>
    wrote:

        The issue here is not really with the uac_auth module, as that
        module isn’t sending the message only updating it with the
        correct authentication info.

        This is normal and correct behavior. When you send the message
        the second time using the same DNS, it will follow the same
        process as the first, trying A then timing out and failing
        over to B. Standard DNS SRV doesn’t include any behavior to
        try to avoid non-responding nodes.

        Ultimately what you need is to know the actual IP that
        elicited the 401 so the next INVITE with the authentication
        can be sent to the same one, using $du or $dd(:$dp). Have you
        tried to get the remote IP in onreply_route and store it is an
        AVP using $si [1] or $socket_in [2]? I don’t think I’ve ever
        used one of these in a reply route. The documentation doesn’t
        specify whether it is valid and they will contain the source
        of the reply, not the request.

        [1] - https://www.opensips.org/Documentation/Script-CoreVar-3-6#si

        [2] -
        https://www.opensips.org/Documentation/Script-CoreVar-3-6#socket_in

        Ben Newlin

        *From: *Users <users-boun...@lists.opensips.org> on behalf of
        nz deals <nzdealsh...@gmail.com>
        *Date: *Thursday, May 29, 2025 at 9:32 AM
        *To: *OpenSIPS users mailling list <users@lists.opensips.org>
        *Subject: *[OpenSIPS-Users] Issue with proxy failover and
        uac_auth()

        * EXTERNAL EMAIL - Please use caution with links and attachments *

        ------------------------------------------------------------------------

        Hi All,

        I'm using OpenSIPS 3.4 and managing carrier trunks via the
        registrant table. In the table, I'm using a proxy value like
        sips:mysip.xx.x

        When the primary carrier A sbc SRV record becomes unreachable,
        OpenSIPS correctly times out INVITE and attempts to fail over
        to the secondary A record (via SRV).

        The secondary endpoint responds with a 401 Unauthorized and
        includes a WWW-Authenticate header. At this point, I assume
        that opensips should not try on the primary carrier A SRV
        record otherwise it will also timeout. but it is trying to
        send another INVITE with Authorization to the primary. this
        timeout because primary A SRV record is not responding.
        opensips sends another INVITE to secondary and this time its
        without Authorization.

        Is there any way to fix this or work around it? Has anyone
        faced a similar problem when using |uac_auth()| in combination
        with failover and the same proxy domain?

        Any advice or suggestions would be greatly appreciated.

        Thank you

        Regards,

        Jason

        _______________________________________________
        Users mailing list
        Users@lists.opensips.org
        http://lists.opensips.org/cgi-bin/mailman/listinfo/users

    _______________________________________________
    Users mailing list
    Users@lists.opensips.org
    http://lists.opensips.org/cgi-bin/mailman/listinfo/users


_______________________________________________
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
_______________________________________________
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users

Reply via email to