Jump to content


Photo

full ci+ support


  • Please log in to reply
342 replies to this topic

Re: full ci+ support #141 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 28 March 2017 - 20:05

There is already a bug: https://bugzilla.gno...g.cgi?id=775721

I guess next time we/I should check bugs before investigating :P
Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #142 Taapat

  • PLi® Core member
  • 2,343 posts

+120
Excellent

Posted 28 March 2017 - 20:16

Before check bugs need to know exactly what needs to check ;) .
In any case thank you, I can confirm that this patch fix my problem with 112 live stream.


Re: full ci+ support #143 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 28 March 2017 - 20:24

Also @samsamsam hit the bug:

https://bugzilla.gno...g.cgi?id=775774

He would probably warn us, but he left OpenPLi :(
Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #144 samsamsam

  • Senior Member
  • 2,024 posts

+146
Excellent

Posted 28 March 2017 - 20:54

@athoik

 

Yes I also prepared patch for this. It is not hard to do.


Edited by samsamsam, 28 March 2017 - 20:56.


Re: full ci+ support #145 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 28 March 2017 - 21:17

Nice to see you here.

Once you realize that ca-certificates is not used, things are getting easy!

Edited by athoik, 28 March 2017 - 21:17.

Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #146 littlesat

  • PLi® Core member
  • 56,273 posts

+691
Excellent

Posted 28 March 2017 - 21:24

Do we need to do something or do we just need to wait?

WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: full ci+ support #147 samsamsam

  • Senior Member
  • 2,024 posts

+146
Excellent

Posted 28 March 2017 - 21:26

Hello @athoik,

 

I just left IPTVPlayer because of you know who.

 

There is one more problem with  TLS backend based on OpenSSL.

You need to add another patch to call function

SSL_set_tlsext_host_name(ssl, servername)

to set server domain otherwise SNI will not be works.

 

This is also simple patch:

You can check were gnutls_server_name_set  (this is GnuTLS equivalent) is called in GnuTLS backend.

 

Regards,

SSS


Edited by samsamsam, 28 March 2017 - 21:29.


Re: full ci+ support #148 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 28 March 2017 - 22:28

@samsamsam, humans what a strange animal....

Regarding SNI, I guess the following will do the job ;)

It seems that nobody is really looking on their bugtracker, but I will post the patch a see if they going to review it.

Till then, most probably patches can be added to our bitbake.

BTW, I followed your advice and totally get rid of gnutls to master-next, by switch to OpenSSL other packages as well. Images reduced few MB!

diff --git a/tls/openssl/gtlsclientconnection-openssl.c b/tls/openssl/gtlsclientconnection-openssl.c
index cb9e647..26f4449 100644
--- a/tls/openssl/gtlsclientconnection-openssl.c
+++ b/tls/openssl/gtlsclientconnection-openssl.c
@@ -422,6 +422,7 @@ g_tls_client_connection_openssl_initable_init (GInitable       *initable,
   GTlsClientConnectionOpenssl *client = G_TLS_CLIENT_CONNECTION_OPENSSL (initable);
   GTlsClientConnectionOpensslPrivate *priv;
   long options;
+  const char *hostname;

   priv = g_tls_client_connection_openssl_get_instance_private (client);

@@ -437,25 +438,21 @@ g_tls_client_connection_openssl_initable_init (GInitable       *initable,
     }

   options = SSL_OP_NO_TICKET;
+  hostname = get_server_identity (client);

   /* Only TLS 1.2 or higher */
   SSL_CTX_set_options (priv->ssl_ctx, options);

 #if OPENSSL_VERSION_NUMBER >= 0x10200000L
-  {
-    const char *hostname;
-
-    hostname = get_server_identity (client);
-    if (hostname)
-      {
-        X509_VERIFY_PARAM *param;
-
-        param = X509_VERIFY_PARAM_new ();
-        X509_VERIFY_PARAM_set1_host (param, hostname);
-        SSL_CTX_set1_param (priv->ssl_ctx, param);
-        X509_VERIFY_PARAM_free (param);
-      }
-  }
+  if (hostname)
+    {
+      X509_VERIFY_PARAM *param;
+
+      param = X509_VERIFY_PARAM_new ();
+      X509_VERIFY_PARAM_set1_host (param, hostname);
+      SSL_CTX_set1_param (priv->ssl_ctx, param);
+      X509_VERIFY_PARAM_free (param);
+    }
 #endif

   SSL_CTX_set_generate_session_id (priv->ssl_ctx, generate_session_id);
@@ -477,6 +474,11 @@ g_tls_client_connection_openssl_initable_init (GInitable       *initable,
   data_index = SSL_get_ex_new_index (0, "gtlsclientconnection", NULL, NULL, NULL);
   SSL_set_ex_data (priv->ssl, data_index, client);

+#ifdef SSL_CTRL_SET_TLSEXT_HOSTNAME
+  if (hostname)
+    SSL_set_tlsext_host_name (priv->ssl, hostname);
+#endif
+
   SSL_set_connect_state (priv->ssl);

   if (!g_tls_client_connection_openssl_parent_initable_iface->

Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #149 samsamsam

  • Senior Member
  • 2,024 posts

+146
Excellent

Posted 28 March 2017 - 22:40

@athoik

 

Yes, this patch looks exactly as it should.

 

I use this TLS backed with this two patches since at least three months. 

And everything works just like it should. I do not have any further problems.

But, first of all OpenSSL is much, much faster than GnuTLS.

 

Regards,

SSS


Edited by samsamsam, 28 March 2017 - 22:44.


Re: full ci+ support #150 WanWizard

  • PLi® Core member
  • 68,614 posts

+1,739
Excellent

Posted 29 March 2017 - 11:06

Guys, may I just say

9e8cc8a0f37b1adefd98dd56cd0b3630_good-jo


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: full ci+ support #151 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 29 March 2017 - 20:22

PR: https://github.com/O...e-core/pull/194

Both patches reported upstream, so others user will benefit as well. Until then patches added in recipe in master-next.
Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #152 WanWizard

  • PLi® Core member
  • 68,614 posts

+1,739
Excellent

Posted 29 March 2017 - 20:32

Merged. I assume you keep an eye on upstream?


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: full ci+ support #153 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 29 March 2017 - 20:33

Sure, I will receive an email from gnome bugzilla once tickets change status ;)
Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #154 Orkf

  • Member
  • 7 posts

0
Neutral

Posted 7 April 2017 - 16:25

Does the slot ci+ work on beta OpenPLI 5 ?



Re: full ci+ support #155 WanWizard

  • PLi® Core member
  • 68,614 posts

+1,739
Excellent

Posted 7 April 2017 - 16:43

The slot works.

 

If you mean if a CI+ module works in CI+ mode in that slot, the answer is no. Not until someone creates a generic solution that doesn't violate any legal constraints (i.e. doesn't contain a CI+ certificate, either as file or obscured in a binary).


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: full ci+ support #156 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 7 April 2017 - 17:16

Merged. I assume you keep an eye on upstream?


Upstream patches commited, will update to new tag later today.
Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #157 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 7 April 2017 - 20:52

glib-openssl updated to 2.50.2 that includes the patches added previously in order to solve SNI and CA certificates issues.

PR: https://github.com/O...e-core/pull/199

Edited by athoik, 7 April 2017 - 20:53.

Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #158 WanWizard

  • PLi® Core member
  • 68,614 posts

+1,739
Excellent

Posted 7 April 2017 - 21:05

Merged.


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: full ci+ support #159 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 7 April 2017 - 21:11

Great, lets hope that faster libsouphttp will make gstreamer experience better :)
Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: full ci+ support #160 WanWizard

  • PLi® Core member
  • 68,614 posts

+1,739
Excellent

Posted 7 April 2017 - 21:38

Now if we could go back to the original topic and fix that too? ;)  :ph34r:


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.



0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users