[OpenSIPS-Users] Intermittent TLS Problem -

Liviu Chircu liviu at opensips.org
Tue Jun 19 02:56:02 EDT 2018


Hi,

Looking at both successful and failed attempts, the TLS connection seems 
to be properly established in both cases. However, for some unknown 
reason, writing the effective data (from the client side) may either 
take a quick 400 us, or it may time out after 60 ms!

As a first way of troubleshooting this, I suggest we increase the 
tls_mgm's module "tls_send_timeout" to some ridiculous value (e.g. 2000 
ms) and see if the client write timeouts still happen.

Best regards,

Liviu Chircu
OpenSIPS Developer
http://www.opensips-solutions.com

On 18.06.2018 18:05, Callum Guy wrote:
> Hi All,
>
> Further to this issue I have established that the fault occurs when 
> the server does not provide a finish "Encrypted Handshake Message" 
> following key/cert exchange.
>
> Based on my test patterns it seems to indicate that this improves 
> /slightly/ over time meaning that its gets a bit more reliable after a 
> few calls have been processed by the remote. This doesn't mean that 
> the calls don't ever fail though, it goes from 50% failure rate to 20% 
> perhaps.
>
> Does anyone have confidence that this indicates a fault with the 
> remote system rather than the OpenSIPs implementation?
>
> Thanks
>
> On Mon, Jun 18, 2018 at 1:15 PM Callum Guy <callum.guy at x-on.co.uk 
> <mailto:callum.guy at x-on.co.uk>> wrote:
>
>     Hi All,
>
>     I'm running a TLS protected service using OpenSIPs 2.3.3
>     and openssl-1.0.2k-12.el7.x86_64 on a CentOS 7.5 server. RTPProxy
>     is in place to forward all the media on every call.
>
>     OpenSIPs is acting as a proxy to an internal FreeSWITCH instance
>     and handling all inbound and outbound TLS negotiation.
>
>     The call scenario is:
>     Leg A: carrier -> opensips -> freeswitch (answer)
>     Leg B: freeswitch -> opensips ->carrier
>
>     I have a situation where calls sometimes fail during TLS
>     negotiation on the B leg. The exact same call will work only
>     seconds before and then the subsequent call will fail. This is
>     still a test service so there is no traffic on the platform which
>     would effect the calls.
>
>     The destination carrier is Twilio where we are targeting a custom
>     domain which provides three separate fallback IP's. The logs
>     (below) show each being attempted in sequence and OpenSIPs is
>     reporting failure.
>
>     I have a packet capture (SIP encrypted) which shows that the
>     carrier (Twilio) are setting up the connection and attempting to
>     change the cipher spec, from my understanding it looks like this
>     is what triggers the errors in OpenSIPs which then returns a SIP
>     packet to Twilio before trying the next IP in the list (see screen
>     shot below).
>
>     When viewing the packet capture on a call which was not affected I
>     don't see wireshark reporting "Change Cipher Spec" from the
>     carrier network so at present this is the main suspicion.
>     Unfortunately this is as far as my current understanding can take
>     me so while i continue to research i wanted to reach out to see is
>     anyone in the community can help! Let me know if I can provide any
>     further information.
>
>     Thanks!
>
>     *OpenSIPs Log*
>
>     Jun 18 11:20:33 opensips[6373]: INFO: [INVITE] (outbound) REST
>     response: RU: sip:+35361234567
>     <tel:+353%2061%20234%20567>@custom.pstn.ie1.twilio.com:5061;transport=tls
>     DU: sip:custom.pstn.ie1.twilio.com:5061
>     <http://custom.pstn.ie1.twilio.com:5061/> FU: "+4940261234567
>     <tel:+49%2040%20261234567>" <si
>     ps:+4940261234567 at 172.20.0.101
>     <mailto:ps%3A%2B4940261234567 at 172.20.0.101>>
>     To(sip:35361234567 at 172.18.0.110:5061;transport=tls)
>     From(sip:+40261234567 at 172.20.0.101
>     <mailto:sip%3A%2B40261234567 at 172.20.0.101>)
>     ID:137a5be9-ed84-1236-10aa-d0431e9b59cf
>     Jun 18 11:20:33 opensips[6373]: INFO:core:probe_max_sock_buff:
>     using snd buffer of 416 kb
>     Jun 18 11:20:33 opensips[6373]: INFO:core:init_sock_keepalive: TCP
>     keepalive enabled on socket 29
>     Jun 18 11:20:33 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 2
>     Jun 18 11:20:33 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     Jun 18 11:20:33 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 1
>     Jun 18 11:20:33 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     Jun 18 11:20:33 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 0
>     Jun 18 11:20:33 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     *Jun 18 11:20:34 opensips[6373]:
>     ERROR:proto_tls:tls_blocking_write: TLS send timeout (60)*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:proto_tls:proto_tls_send:
>     failed to send*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:tm:msg_send: send() to
>     54.171.127.194:5061 <http://54.171.127.194:5061/> for proto tls/3
>     failed*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:tm:t_forward_nonack:
>     sending request failed*
>     Jun 18 11:20:34 opensips[6373]: INFO:core:probe_max_sock_buff:
>     using snd buffer of 416 kb
>     Jun 18 11:20:34 opensips[6373]: INFO:core:init_sock_keepalive: TCP
>     keepalive enabled on socket 29
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 2
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 1
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 0
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     *Jun 18 11:20:34 opensips[6373]:
>     ERROR:proto_tls:tls_blocking_write: TLS send timeout (60)*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:proto_tls:proto_tls_send:
>     failed to send*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:tm:msg_send: send() to
>     54.171.127.193:5061 <http://54.171.127.193:5061/> for proto tls/3
>     failed*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:tm:t_forward_nonack:
>     sending request failed*
>     Jun 18 11:20:34 opensips[6373]: INFO:core:probe_max_sock_buff:
>     using snd buffer of 416 kb
>     Jun 18 11:20:34 opensips[6373]: INFO:core:init_sock_keepalive: TCP
>     keepalive enabled on socket 29
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 2
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 1
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     depth = 0
>     Jun 18 11:20:34 opensips[6373]: NOTICE:tls_mgm:verify_callback:
>     preverify is good: verify return: 1
>     *Jun 18 11:20:34 opensips[6373]:
>     ERROR:proto_tls:tls_blocking_write: TLS send timeout (60)*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:proto_tls:proto_tls_send:
>     failed to send*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:tm:msg_send: send() to
>     54.171.127.192:5061 <http://54.171.127.192:5061/> for proto tls/3
>     failed*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:tm:t_forward_nonack:
>     sending request failed*
>     *Jun 18 11:20:34 opensips[6373]: ERROR:tm:w_t_relay:
>     t_forward_nonack failed*
>     *Jun 18 11:20:34 opensips[6373]: ERROR: Failed to relay call -
>     DU:<null> [INVITE]
>     To(sip:35361234567 at 172.18.0.110:5061;transport=tls)
>     From(sip:+40261234567 at 172.20.0.101
>     <mailto:sip%3A%2B40261234567 at 172.20.0.101>)
>     ID:137a5be9-ed84-1236-10aa-d0431e9b59*
>
>     *Wireshark Failure:*
>     https://ibb.co/ici8Ny
>
>     *Wireshark Working:*
>     https://ibb.co/kePF2y
>
>     -- 
>     Callum Guy
>     Head of Information Security
>     X-on
>
> -- 
> Callum Guy
> Head of Information Security
> X-on
>
>
> *^0333 332 0000  | www.x-on.co.uk <http://www.x-on.co.uk>  | 
> _**_^<https://www.linkedin.com/company/x-on> 
> <https://www.facebook.com/XonTel> <https://twitter.com/xonuk> *
> X-on is a trading name of Storacall Technology Ltd a limited company 
> registered in England and Wales.
> Registered Office : Avaland House, 110 London Road, Apsley, Hemel 
> Hempstead, Herts, HP3 9SD. Company Registration No. 2578478.
> The information in this e-mail is confidential and for use by the 
> addressee(s) only. If you are not the intended recipient, please 
> notify X-on immediately on +44(0)333 332 0000 and delete the
> message from your computer. If you are not a named addressee you must 
> not use, disclose, disseminate, distribute, copy, print or reply to 
> this email. Views or opinions expressed by an individual
> within this email may not necessarily reflect the views of X-on or its 
> associated companies. Although X-on routinely screens for viruses, 
> addressees should scan this email and any attachments
> for viruses. X-on makes no representation or warranty as to the 
> absence of viruses in this email or any attachments.
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20180619/e66d5864/attachment-0001.html>


More information about the Users mailing list