Re: TLS 1.2 with Suite B

2014-11-17 Thread Fredrik Jansson
:6b:15: 5d:25:45:59:86:22:f6:de:06:db:0c:60:8e:30:32:45:d6:c6: 01:16:a3:94:9d:cb:4e:fd:a6:6b:a1:73 On Mon, Nov 17, 2014 at 3:06 PM, Fredrik Jansson wrote: > I actually got a bit further with a secp256r1 server certificate, I > also changed the server certificate version fro

Re: TLS 1.2 with Suite B

2014-11-17 Thread Fredrik Jansson
, 2014 at 1:19 PM, Fredrik Jansson wrote: > More tests as you suggested: > > openssl s_client -tls1_2 -connect XXX:9103 > openssl s_server -state -tls1_2 -cipher SUITEB128 -accept 9103 > > Using default temp DH parameters > ACCEPT > SSL_accept:before/accept ini

Re: TLS 1.2 with Suite B

2014-11-17 Thread Fredrik Jansson
hello C ERROR 139990990374592:error:1408A0C1:SSL routines:ssl3_get_client_hello:no shared cipher:s3_srvr.c:1398: shutting down SSL CONNECTION CLOSED Warm regards, Fredrik On Mon, Nov 17, 2014 at 1:09 PM, Fredrik Jansson wrote: > Hi! > > I have tried with s_client, and I get the same erro

Re: TLS 1.2 with Suite B

2014-11-17 Thread Fredrik Jansson
014, Fredrik Jansson wrote: > >> Some more info, >> >> SSL_get_ciphers on the server and client: >> Info2014-Nov-17 10:48:26.961112 All.TLSVerbose >> ECDHE-ECDSA-AES128-GCM-SHA256 >> Info2014-Nov-17 10:48:26.961114 All.TLSVerbose >> ECDHE-ECD

Re: TLS 1.2 with Suite B

2014-11-17 Thread Fredrik Jansson
listed (among with several others). Fredrik On Mon, Nov 17, 2014 at 9:54 AM, Fredrik Jansson wrote: > Hi Steve! > > I remade the certs as below, but I still get the same error, i.e. > 1408A0C1:SSL routines:ssl3_get_client_hello:no shared cipher. > > Anything else I can try

Re: TLS 1.2 with Suite B

2014-11-17 Thread Fredrik Jansson
:77:ba:91:4b:34:4f:94:92:b1:a3: da:5f:43:13:61:d0:02:bc:27:65:47:ac:ba:4e:79:13:84:cd: eb:c6:5e:a3:94:e9:fa:48:48:e9:78:f9:d3 On Fri, Nov 14, 2014 at 11:32 PM, Dr. Stephen Henson wrote: > On Fri, Nov 14, 2014, Fredrik Jansson wrote: > >> Hi Steve, thanks for

Re: TLS 1.2 with Suite B

2014-11-14 Thread Fredrik Jansson
Fri, Nov 14, 2014, Fredrik Jansson wrote: > >> Hi! >> >> Thanks! >> >> I am using 1.0.2b3 on both server and client, and I have the call to >> SSL_CTX_set_ecdh_auto, but still no luck. >> >> The exact code is as follows: >> >> 358

Re: TLS 1.2 with Suite B

2014-11-14 Thread Fredrik Jansson
SSL_CTX_set_options(ctx_, SSL_OP_NO_TICKET); 415 SSL_CTX_set_session_cache_mode(ctx_, SSL_SESS_CACHE_BOTH); 416 SSL_CTX_sess_set_remove_cb(ctx_, ssl_remove_session_cb); 417 CertStore::setStoreInCTX(ctx_); 418 } Warm regards, Fredrik On Fri, Nov 14, 2014 at 3:36 PM, Dr. Stephen Henson

TLS 1.2 with Suite B

2014-11-14 Thread Fredrik Jansson
Hi! I am trying to force my TLS 1.2 connection into Suite B mode, but at handshake I get an error "no shared cipher". The server code is basically: SSL_CTX_new(TLSv1_2_server_method()); //ECDSA cert is added to the ctx SSL_CTX_use_certificate(ctx_, serverCert.cert.get()) SSL_CTX_use_PrivateKey(c

Re: External client certificate signature function

2014-10-15 Thread Fredrik Jansson
? Warm regards, Fredrik On Mon, Oct 13, 2014 at 6:08 PM, Dr. Stephen Henson wrote: > On Mon, Oct 13, 2014, Fredrik Jansson wrote: > >> Hi! >> >> I have a device where I cannot access the client certificate's private >> key directly, but have access to verification

Re: External client certificate signature function

2014-10-15 Thread Fredrik Jansson
Hi Steve! I will try to take that path, thank you! //Fredrik On Mon, Oct 13, 2014 at 6:08 PM, Dr. Stephen Henson wrote: > On Mon, Oct 13, 2014, Fredrik Jansson wrote: > >> Hi! >> >> I have a device where I cannot access the client certificate's private >&g

External client certificate signature function

2014-10-13 Thread Fredrik Jansson
rt_cb function. I can convert the encoded cert to a X509 structure and return that, but I cannot provide it with a EVP_PKEY object. Is there any way I can instruct any of the SSL_CTX, SSL or EVP_PKEY objects to call a signature function (that I provide) during the handshake? Best regards, Fredrik Ja

X509_STORE_add_crl with updated CRL

2014-03-12 Thread Fredrik Jansson
I have two CRL, one empty and one with a revoked certificate. I create a X509_STORE and use X509_STORE_add_crl to add the empty CRL. When verifying the certificate using the store, it verifies alright. Then I add the CRL with the revoked cert to the same store, again using X509_STORE_add_crl. W

Re: Flushing encrypted data to file

2014-03-10 Thread Fredrik Jansson
Hi! Some example code to extract a cert from a P12 file: BIO* certFile = BIO_new_file("cert path", "r"); PKCS12* p12 = nullptr; X509* cert = nullptr; if (!certFile) { goto done; } p12 = d2i_PKCS12_bio(certFile, nullptr); if (!

Re: AES CCM in DTLS v1.2

2014-03-06 Thread Fredrik Jansson
Mar 04, 2014, Fredrik Jansson wrote: > >> I am currently using DTLS v1.1 but with the introduction of v1.2 in OpenSSL >> 1.0.1f I was hoping to be able to use AES CCM mode. >> >> We use PSK to key DTLS and the resulting algorithm is PSK-AES256-CBC-SHA. >> Is it possi

AES CCM in DTLS v1.2

2014-03-04 Thread Fredrik Jansson
I am currently using DTLS v1.1 but with the introduction of v1.2 in OpenSSL 1.0.1f I was hoping to be able to use AES CCM mode. We use PSK to key DTLS and the resulting algorithm is PSK-AES256-CBC-SHA. Is it possible to stick with PSK and migrate to AES CCM? Best regards, Fredrik

Re: DTLS handshake messages

2013-12-11 Thread Fredrik Jansson
Hi Sravanthi, I have implemented this as follows: One thread per listen socket, calling DTLSv1_listen. This thread spawns one thread per client and those treads call SSL_accept. The listening threads share a single SSL_CTX, and each call to SSL_new is protected by a mutex. Each SSL object has a

Re: DTLS PSK in FIPS mode

2013-11-04 Thread Fredrik Jansson
04, 2013, Fredrik Jansson wrote: > > > Thanks, that did it! > > > > To try to understand the implications of this, if I add SSL_FIPS > > to TLS1_TXT_PSK_WITH_AES_128_CBC_SHA and > TLS1_TXT_PSK_WITH_AES_256_CBC_SHA, > > am I violating the security policy? AES 128/256

Re: DTLS PSK in FIPS mode

2013-11-04 Thread Fredrik Jansson
2:31 PM, Dr. Stephen Henson wrote: > On Mon, Nov 04, 2013, Fredrik Jansson wrote: > > > Steve, thanks for getting back! > > > > Since I could not reproduce this using s_client and s_server I set out to > > take the code I am using into a sample project. > > >

Re: DTLS PSK in FIPS mode

2013-11-04 Thread Fredrik Jansson
Sun, Nov 3, 2013 at 4:15 PM, Dr. Stephen Henson wrote: > On Fri, Oct 25, 2013, Fredrik Jansson wrote: > > > > > I am trying to use DTLS with PSK (cipher: SSL_TXT_PSK). Everything works > > well if I don't set OpenSSL in FIPS mode (FIPS_mode_set(1)). > > > >

Re: DTLS PSK in FIPS mode

2013-10-25 Thread Fredrik Jansson
t; } before *p == SSL3_MT_CLIENT_HELLO, it works. Should I report a bug? Fredrik On Fri, Oct 25, 2013 at 2:03 PM, Fredrik Jansson < fredrik.jansson...@gmail.com> wrote: > Hi! > > > I am trying to use DTLS with PSK (cipher: SSL_TXT_PSK). Everything works > well if

DTLS PSK in FIPS mode

2013-10-25 Thread Fredrik Jansson
Hi! I am trying to use DTLS with PSK (cipher: SSL_TXT_PSK). Everything works well if I don't set OpenSSL in FIPS mode (FIPS_mode_set(1)). If I do, I get crashes as below where p =0; Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7fffddffb700 (LWP 15278)] 0x7

Re: Support for SHA-512/224 and SHA-512/256

2013-10-04 Thread Fredrik Jansson
Hi! Note, this is OpenSSL 1.0.1c. sha.h: int SHA512_Init(SHA512_CTX *c); evp.h: const EVP_MD *EVP_sha512(void); Best regards, Fredrik On Fri, Oct 4, 2013 at 9:44 AM, Gnanasekar R wrote: > Hi, > > I would like to know if OpenSSL has support for SHA-512/224 and > SHA-512/256. I see init functio

Deadlock in FIPS mode

2013-06-25 Thread Fredrik Jansson
Hi! I have managed to deadlock OpenSSL while running in FIPS mode. The locking functions are setup according to mttest.c and th-lock.c using pthread_mutex_. Please note I have NOT explicitly set a thread id function. Then env is: openssl-1.0.1e openssl-fips-2.0.5 RHEL 6 - 32 bit. Please let me

DTLS - cannot make client detect restarted server

2012-01-03 Thread Fredrik Jansson
e the client needs to take some measure. The client happily keeps sending data. Any help on how to resolve this would be greatly appreciated. Best regards, Fredrik Jansson