I overcame the issue!
The problem was on the server side: I forgot to initialize the
prestate(gnutls_dtls_prestate_st) structure.
But in my opinion the case when client side behave the undocumented way
if server sends garbage is a bug on the client side too.
I haven't reproduce it with exaples but reproduced it with the next code
on server side:
gnutls_dtls_prestate_st prestate;
memset(&prestate, 0, sizeof(prestate));
prestate.record_seq = 105791312;
prestate.hsk_read_seq = 0;
prestate.hsk_write_seq = 67166359;
No I have the next issue withing handshake.
Gnutls: REC[0x65ddbb0]: INVALID VERSION PACKET: (-1) 254.255
I hope, I will overcame this too.
Kind regards
Yury Shvedov
WiMark Systems
On 10/28/2015 04:12 PM, Yury Shvedov wrote:
Hi!
I'm back with my problem, described here:
(http://lists.gnutls.org/pipermail/gnutls-help/2015-April/003852.html).
The client begins handshake, with NONBLOCK exchange and after
receiving HELLO VERIFY REQUEST it stops working and seems to waiting
for other data, calling pull_timeout with non-zero timeout once and
then - zero timeout endlessly.
Here I have logs of server and client.
Such behaviour I have for both 3.3.17 and 3.4.6 versions.
Thank you!
_______________________________________________
Gnutls-help mailing list
[email protected]
http://lists.gnupg.org/mailman/listinfo/gnutls-help
_______________________________________________
Gnutls-help mailing list
[email protected]
http://lists.gnupg.org/mailman/listinfo/gnutls-help
