Does openssl include intel AES-NI-GCM?

2013-02-28 Thread Kane Huang
Hi all
I'm reading a paper from intel that mention an AES-NI-GCM implementation within 
the Linux kernel cryptographic framework using the new instructions(PCLMULQDQ 
and AES-NI).

As I know ,aes-ni is already include in 1.0.1, So How does the AES-NI-GCM?

The paper is here
http://www.intel.com/content/dam/www/public/us/en/documents/white-papers/aes-ipsec-performance-linux-paper.pdf


Regards,
Kane
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


RE: x509 certificate conforming to RFC 3739

2013-02-28 Thread Steven Madwin
Hi Walter,
I hope this helps.
Steve

id-pda-dateOfBirth  = 1.3.6.1.5.5.7.9.1

[ subjectDirAttrib_sec ]
1.3.6.1.5.5.7.9.1  = SEQUENCE:dateOfBirth_sec   #id-pda-dateOfBirth

[ dateOfBirth_sec ]
id-pda-dateOfBirth.name= OID:id-pda-dateOfBirth
id-pda-dateOfBirth.value = SET:dateOfBirth_val
[ dateOfBirth_val ]
id-pda-dateOfBirth.value = GENERALIZEDTIME:19870427092500Z

-Original Message-
From: owner-openssl-us...@openssl.org
[mailto:owner-openssl-us...@openssl.org] On Behalf Of Walter H.
Sent: Sunday, February 17, 2013 9:07 PM
To: openssl-users@openssl.org
Subject: Re: x509 certificate conforming to RFC 3739

On 15.02.2013 07:42, Walter H. wrote:
 On Fri, February 15, 2013 07:07, Matthew Hall wrote:
 On Fri, Feb 15, 2013 at 07:03:20AM +0100, Walter H. wrote:
 Hello,

 can someone, please tell me, how to generate a certificate
 that conforms to http://www.ietf.org/rfc/rfc3739.txt (RFC 3739)

 Thanks,
 Walter
 Hi Walter,

 We could help better if we understood what's not working for you and what
 you
 tried already.

 Regards,
 Matthew.
 Hello,

 Section C.1.1.1. The subjectDirectoryAttributes Extension

 shows this:

 {
 type id-pda-dateOfBirth,
 values {
GeneralizedTime : 197110141200Z
 }
 },

 How do I get an attribute as GeneralizedTime instead of PrintableString?

 Thanks,
 Walter
Nobody?

Thanks,
Walter



smime.p7s
Description: S/MIME cryptographic signature


Re: openssl_private_encrypt crashes browser after 10 seconds

2013-02-28 Thread rahmad.bkt
I had the same problem with you, I've done the same thing with you to resolve
the problem. but after I mentioned how my apache server can not run. is
there any other way to solve this problem?



--
View this message in context: 
http://openssl.6102.n7.nabble.com/openssl-private-encrypt-crashes-browser-after-10-seconds-tp28780p43967.html
Sent from the OpenSSL - User mailing list archive at Nabble.com.
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


Re: check certificate chain in a pem file

2013-02-28 Thread ashish2881
Hi Jakob ...
 My doubt is like this .

I have a .pem file say : chain.pem 
chain.pem == server certificate- intermediate CA certificate - self signed
root certificate .

Now i am writing the code in C using opensl Api's to verify this (chain.pem)
chain .
filename == chain.pem
Currently I am doing like this : 
SSL_CTX *sslctx = NULL;
STACK_OF(X509)* ca_stack = NULL ;
sslctx = SSL_CTX_new(SSLv23_server_method());

BIO_new(BIO_s_file_internal());
x =PEM_read_bio_X509
store=X509_STORE_new
vrfy_ctx = X509_STORE_CTX_new();
SSL_CTX_use_certificate_chain_file(sslctx, filename)
ca_stack = sslctx-extra_certs
X509_STORE_CTX_init(vrfy_ctx, NULL, x, ca_stack)
X509_verify_cert(vrfy_ctx)

---i am seeing error while doing this .am i doing everything right .

or please let me know ...how can i verify my chain.pem .




--
View this message in context: 
http://openssl.6102.n7.nabble.com/check-certificate-chain-in-a-pem-file-tp43871p43971.html
Sent from the OpenSSL - User mailing list archive at Nabble.com.
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


Re: Need help with Windows and FIPS enabled OpenSSL libs

2013-02-28 Thread Glenn Smith
I wondered about that.  I saw several mentions about that.



--
View this message in context: 
http://openssl.6102.n7.nabble.com/Need-help-with-Windows-and-FIPS-enabled-OpenSSL-libs-tp43956p43975.html
Sent from the OpenSSL - User mailing list archive at Nabble.com.
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


Re: Need help with Windows and FIPS enabled OpenSSL libs

2013-02-28 Thread Glenn Smith
Does the same apply to DLLs?  I'm planning on using OpenSSL+Fips in a DLL
that gets loaded either by other DLLs or directly from an application as a
support DLL (the App won't be known, as we're doing DLL to be used by other
vendors).





--
View this message in context: 
http://openssl.6102.n7.nabble.com/Need-help-with-Windows-and-FIPS-enabled-OpenSSL-libs-tp43956p43984.html
Sent from the OpenSSL - User mailing list archive at Nabble.com.
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


Syntax in extension section of openssl.cnf

2013-02-28 Thread Walter H.
Hello,

I have the following:

---
name = ASN1:SEQUENCE:section

[ section ]
value.1 = XXX:text
---

what is possible to have instead of XXX?

Thanks,
Walter



smime.p7s
Description: S/MIME Cryptographic Signature


Re: [openssl-users] Syntax in extension section of openssl.cnf

2013-02-28 Thread Erwann Abalea

man asn1parse
man ASN1_generate_nconf

That should give you some bootstrap information.

--
Erwann ABALEA
-
tridécatabulophobie: peur d'être treize à table

Le 28/02/2013 11:16, Walter H. a écrit :

Hello,

I have the following:

---
name = ASN1:SEQUENCE:section

[ section ]
value.1 = XXX:text
---

what is possible to have instead of XXX?

Thanks,
Walter





Re: Need help with Windows and FIPS enabled OpenSSL libs

2013-02-28 Thread Dr. Stephen Henson
On Wed, Feb 27, 2013, Glenn Smith wrote:

 Does the same apply to DLLs?  I'm planning on using OpenSSL+Fips in a DLL
 that gets loaded either by other DLLs or directly from an application as a
 support DLL (the App won't be known, as we're doing DLL to be used by other
 vendors).
 

Since you haven't quoted any context it's not clear what the same means. I'll
assume you mean the FIPS signature referred to in this thread.

If the DLL ends up linking in fipscanister.lib then yes it needs to include
the signature. That will happen if you link against the static versions of
OpenSSL.

If the DLL just links to the FIPS capable OpenSSL DLL (libeay32.dll) then it
wont embed a signature as fipscanister.lib is linked into libeay32.dll and the
build process automatically includes the signature.

Steve,
--
Dr Stephen N. Henson. OpenSSL project core developer.
Commercial tech support now available see: http://www.openssl.org
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


how to disable

2013-02-28 Thread Chaim | Sz Studios
BH

Hi All,
I'm trying to pass a pci scan, I'm on Ubuntu 12.04 lts server and Nginx.
I've tried everything I know and did a lot of research... apparently seems
that need to disable a setting in OpenSSL which I can't find how to do.

This is the result of the scan:

SSL/TLS Protocol Initialization Vector Implementation Information
Disclosure Vulnerability www (443/tcp)

CVSS Score: Medium 4.3

Fail

CVE-2011-3389
   and this is the suggested fix:

  Configure SSL/TLS servers to only use TLS 1.1 or TLS 1.2 if supported.
Configure SSL/TLS servers to only support cipher suites that do not use
block ciphers. Apply patches if available.

OpenSSL uses empty fragments as a countermeasure unless the
'SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS' option is specified when OpenSSL is
initialized.
  I don't know how to do this, please help.
Note: tlsv1.1 and tlsv1.2 are not supported. I upgraded to latest verions
of all (Ubuntu, nginx and openssl) which I read it takes care of the tls
being not supported by default. But I'm getting the exact same report from
the scan.

-- 

Chaimy
786.277.8760


Re: [openssl-users] Syntax in extension section of openssl.cnf

2013-02-28 Thread Walter H.

Thanks,
Walter

On 28.02.2013 12:03, Erwann Abalea wrote:

man asn1parse
man ASN1_generate_nconf

That should give you some bootstrap information.

--
Erwann ABALEA
-
tridécatabulophobie: peur d'être treize à table
Le 28/02/2013 11:16, Walter H. a écrit :

Hello,

I have the following:

---
name = ASN1:SEQUENCE:section

[ section ]
value.1 = XXX:text
---

what is possible to have instead of XXX?

Thanks,
Walter







smime.p7s
Description: S/MIME Cryptographic Signature


Re: aesni usage in C

2013-02-28 Thread Dr. Stephen Henson
On Thu, Feb 28, 2013, John-Alan Simmons wrote:

 I am trying to implement aes encryption via openssl using C, and cant
 figure out how to use the aesni hardware acceleration. I have already
 confirmed my processor can use, any help would be great.
 

If you use the EVP interface AES-NI is automatically used on supported
platforms.

Steve.
--
Dr Stephen N. Henson. OpenSSL project core developer.
Commercial tech support now available see: http://www.openssl.org
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


TLS ClientHello v3.3 not supported by all servers

2013-02-28 Thread LANGLOIS Olivier PIS -EXT
They just reset the connection upon the reception of the message.

This is a new behavior from version 1.0.1-e 1.0.1-d was which sending
ClientHello v3.2 are fine.

 How to reproduce:

openssl s_client -connect aur.archlinux.org:443



CONFIDENTIALITY : This e-mail and any attachments are confidential and may be 
privileged. If you are not a named recipient, please notify the sender 
immediately and do not disclose the contents to another person, use it for any 
purpose or store or copy the information in any medium.
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


TLS ClientHello v3.3 not supported by all servers

2013-02-28 Thread LANGLOIS Olivier PIS -EXT
They just reset the connection upon the reception of the message.

This is a new behavior from version 1.0.1-e
1.0.1-d was which sending ClientHello v3.2 are fine.

How to reproduce:

openssl s_client -connect aur.archlinux.org:443



CONFIDENTIALITY : This e-mail and any attachments are confidential and may be 
privileged. If you are not a named recipient, please notify the sender 
immediately and do not disclose the contents to another person, use it for any 
purpose or store or copy the information in any medium.
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


ongoing fingerprint does not match issues

2013-02-28 Thread T J


Comparing printed debug values that I put into the incore script, it 
looks like our compileris putting the signature somewhere between 
FIPS_rodata_start and FIPS_rodata_end resulting in thefingerprint does 
not match segment aliasing error. I can get around this by removing the 
-fPIC option from the Configure files (both fips and base), but then I 
get the straight fingerprint does not match error. Is this because I 
changed the Configure file in openssl-fips-2.0? It also seems incore is 
incorrectly calculating the location of FIPS_signature - it's out by 
65536. Compensating for it in the incore script (I simply add 65536) 
however, doesn't appear to make any difference. Any advice anyone?

TJ.



Re: TLS ClientHello v3.3 not supported by all servers

2013-02-28 Thread Le Huang
Olivier,

The command line you provided just worked fine in my box, in which a fresh
openssl-1.0.1e installed.

On Fri, Mar 1, 2013 at 7:31 AM, LANGLOIS Olivier PIS -EXT 
olivier.pis.langl...@transport.alstom.com wrote:

 They just reset the connection upon the reception of the message.

 This is a new behavior from version 1.0.1-e 1.0.1-d was which sending
 ClientHello v3.2 are fine.

  How to reproduce:

 openssl s_client -connect aur.archlinux.org:443


 
 CONFIDENTIALITY : This e-mail and any attachments are confidential and may
 be privileged. If you are not a named recipient, please notify the sender
 immediately and do not disclose the contents to another person, use it for
 any purpose or store or copy the information in any medium.
 __
 OpenSSL Project http://www.openssl.org
 User Support Mailing Listopenssl-users@openssl.org
 Automated List Manager   majord...@openssl.org




-- 
Regards,
Huang Le (Eric, Alibaba DevOps)
Email: 4tarhl AT gmail.com, le.hl AT alibaba-inc.com


Identifying length of decrypted data.

2013-02-28 Thread Tayade, Nilesh
Hi,

I am using the AES_cbc_encrypt() API to perform symmetric decryption.
But in one of the scenarios I see the data encrypted is of 16bytes but just 
1byte is useful and 15bytes is for padding.

Is there some way I can know the actual encrypted data length?

I came across some pointers like using EVP APIs- but I need to avoid using EVP 
APIs as there are locking callbacks and mallocs() involved- which I have to 
avoid, as my application is performance critical. Also the data I am decrypting 
is HTTPS packet, so I cannot manipulate the encrypted data (to mention the 
length in data, and then encrypt it).

Any advice on this will be appreciated.

--
Thanks,
Nilesh
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org


Re: Identifying length of decrypted data.

2013-02-28 Thread Jakob Bohm

On 3/1/2013 7:24 AM, Tayade, Nilesh wrote:

Hi,

I am using the AES_cbc_encrypt() API to perform symmetric decryption.
But in one of the scenarios I see the data encrypted is of 16bytes but just 
1byte is useful and 15bytes is for padding.

Is there some way I can know the actual encrypted data length?

I came across some pointers like using EVP APIs- but I need to avoid using EVP 
APIs as there are locking callbacks and mallocs() involved- which I have to 
avoid, as my application is performance critical. Also the data I am decrypting 
is HTTPS packet, so I cannot manipulate the encrypted data (to mention the 
length in data, and then encrypt it).

Any advice on this will be appreciated.



Please refer to the TLS RFCs (there are different RFCs depending on 
protocol version and encryption algorithm).  Each one specifies in 
detail how to recover the length from the encrypted data.


You also need to read the research document linked from the latest
OpenSSL security advisory, as it describes security critical changes
to the information in the RFCs about how to recover the length without
creating a security hole (fundamentally, you need to make sure you
take exactly the same number of CPU cycles to process a given number
of encrypted bytes, regardless of the recovered length or decryption
failures).



Enjoy

Jakob
--
Jakob Bohm, CIO, Partner, WiseMo A/S.  http://www.wisemo.com
Transformervej 29, 2730 Herlev, Denmark.  Direct +45 31 13 16 10
This public discussion message is non-binding and may contain errors.
WiseMo - Remote Service Management for PCs, Phones and Embedded
__
OpenSSL Project http://www.openssl.org
User Support Mailing Listopenssl-users@openssl.org
Automated List Manager   majord...@openssl.org