duplicate SSL record in different TCP packets from a Google Drive client

Hello, There seems to be a possible problem with the SSL implementation used in Google Drive on MacOS 10.8.2. I seems that this SSL implementation is NSS - please let me know if you know that Google Drive uses a different SSL implementation and I should direct this question elsewhere. Packet

Re: duplicate SSL record in different TCP packets from a Google Drive client

I can provide PCAP files by e-mail, if needed. Peter -- dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

client-only support of DHE ciphers

suites are supported only on the client and not on the server. Is server support for these ciphers just not implemented yet or is there some issue that blocks such implementation? Best Regards, Peter Djalaliev -- dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https

Re: A dedicated SSL MITM box on the market

is hidden and not advertised publicly. Best Regards, Peter Djalaliev On Mar 29, 11:11 am, Jean-Marc Desperrier jmd...@gmail.com wrote: Jean-Marc Desperrier wrote: Article on Wired here : http://www.wired.com/threatlevel/2010/03/packet-forensics/ The original article is well worth reading

Re: A dedicated SSL MITM box on the market

To prevent case 2, CAs are supposed to keep their certificates safe and revoke them if a suspicion arises that the key is not safe. Or rather, CAs are supposed to keep their private keys safe, not their certificates. -- dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org

Re: A dedicated SSL MITM box on the market

In the case of Netronome's SSL Inspector, if the device replaces bankofamerica.com's server certificate in the SSL handshake, the new certificate does _not_ have CN=bankofamerica.com. It is also not signed by a root CA. Hence, Alice should be fully aware of the man-in- the-middle and could

Re: [Fwd: How to display the cause of an SSL client authentication failure]

It is our standard security nightmare.  Side A thinks it is Side B's problem.  Side B thinks it is Side A's problem.  In the meantime the user doesn't use the tech because it doesn't work, and the sides are too busy arguing to solve the problem.  So zero security is delivered. In this case,

Re: attack against AES-256 with complexity 2^119

... Best Regards, Peter Djalaliev -- dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

Re: attack against AES-256 with complexity 2^119

The weakness was discovered when we looked at AES as a hash function, and tried to find weaknesses that are specific for hash functions. We think that most cryptographers used only blockcipher-oriented techniques, against which AES was well protected by the designers. All this quote says, I

attack against AES-256 with complexity 2^119

/new_attack_on_a.html Some of the new SHA-3 algorithm candidates may be affected, too... Best Regards, Peter Djalaliev -- dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

Re: attack against AES-256 with complexity 2^119

algorithm input data may be related, which may make related-key attacks plausible against those SHA-3 candidates. The authors have not shown that the attack is effective against AES-128. However, in many real-world applications, such as TLS, AES-256 is still more secure than AES-128. Best Regards, Peter

Re: PKCS#11 Module for TPM availiable

-friendly functions, such as wrapping symmetric keys and a more secure random number generator, that Trousers' may directly use in its PKCS#11 module, but a lot of PKCS#11 functionality is not provided by the TPM and has to be implemented in software. Best Regards, Peter Djalaliev -- dev-tech-crypto

SSL module for nginx implemented using NSS

Hello, Does anybody know if there is an SSL/TLS module for nginx implemented using NSS? The module that ships with nginx uses OpenSSL. I didn't find anything on Google. Best Regards, Peter Djalaliev -- dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org

Re: nsIChannel for internal domains

due to some other controversies.    Will there be any problems in querying the nsISSLStatus interface within http-on-examine-response ? regards Arun On Jun 26, 8:08 pm, Peter Djalaliev peter.djalal...@gmail.com wrote: Arun, One way is to register a nsIStreamListener using

Re: nsIChannel for internal domains

3.0.3.. For earlier versions, I don't know if there is a way to get the nsISSLStatus. From what I can see in the Firefox source code, only the module that initiated the HTTPS request can get this information. I hope this helps. Best Regards, Peter Djalaliev On Jun 26, 2:40 am, MAK arungene

Re: nsIChannel for internal domains

the certificate. I hope this helps. Best Regards, Peter Djalaliev On Jun 25, 12:50 am, MAK arungene...@gmail.com wrote: Hi all,    My requirement is to fetch a website's SSL server certificate. For this I need the nsIChannel, so that I can use the securityInfo option.    I need to fetch nsIChannel

Re: KEYGEN tag - documentation

. Best Regards, Peter Djalaliev On Jun 1, 9:31 am, Jan Schejbal jan.schejbal_n...@gmx.de wrote: I did of course google and I did find the site you linked, but it did not help me much, as I found no information what has to happen server-side (or links to such information). I understand that the key

ways to import private keys into the Firefox database

to import private keys into the Firefox key/cert database? As far as I know, the FF database is normally protected with a master password. What operation exactly requires providing this password? Regards, Peter Djalaliev ___ dev-tech-crypto mailing list dev

Re: inserting own extended validation certificate root

that adding your own rogue EV CAs defeats the purpose of having EV certificates issued by CAs that conform to established EV policies. I would like to reiterate the question they asked you in the OpenSSL users forum: what are you trying to accomplish by adding your own EV CA cert? Regards, Peter

automatically installing new client SSL certificate into Firefox

certificate (along with the private key) into the NSS database. Regards, Peter Djalaliev ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

Re: creating DH server certificates with NSS

in order to generate a DH SSL certificate using NSS? Does NSS also have an API to generate the DH parameters? Regards, Peter Djalaliev ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

question about certificate chain from https://suppliers.intel.com

Hello, I tried connecting to http://suppliers.intel.com (which redirects to https://supplier.intel.com/supplierhub) from Firefox 3 and IE7 and saw two different certificate chains when I tried to view the server certificate. IE7 recognized the root certificate as coming from a trusted issuer,

Re: question about certificate chain from https://suppliers.intel.com

The correct initial URL is http://supplier.intel.com, redirected to https://supplier.intel.com/supplierhub ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

Re: question about certificate chain from https://suppliers.intel.com

Ah, I see. From what I can see in the RFC, this usage is not really forbidden, but not really standard either. Generalizing my question, what kind of X509v3 extensions that NSS currently support? I am aware that CA often use these extensions in less-than-standard ways :) Peter On Jul 23,

static compiling of applications using NSS

software token code. Frankly, I'm not sure how to deal with that. Does anybody have experience with trying to statically compile applications that use NSS on the Linux platform? How did you proceed? I welcome all tips :) Thanks! Regards, Peter Djalaliev

Re: On use of authenticated cryptography

of hardware that they need to use properly in order to preserve security. As we see from previous messages in this discussion thread (blindly accepting self-signed certificates), users do not properly execute security-sensitive actions. Regards, Peter Djalaliev

Re: On use of authenticated cryptography

On Aug 27, 2:41 am, Nelson B [EMAIL PROTECTED] wrote: Peter Djalaliev wrote: I don't know what you mean by full-blown TPMs. I assume that for you, full-blown TPMs = Big Brother. I don't buy into this completely, however. TPM-enabled systems are still under implementation

Re: On use of authenticated cryptography

Now, when I send my sensitives data to that party, that party could always turn around and give my data to my enemies, put it on a road-side bill board, or disseminate it in various ways of which I don't approve. Having an authenticated certificate doesn't assure me that the party won't do

Re: Debug. Log data being signed and verified.

Mariano, You can use the NSPR logging macros from within any Mozilla code in conjunction with the NSPR_LOG_MODULES and the NSPR_LOG_FILE environment variables. Please refer to the NSPR documentation on the subject: http://www.mozilla.org/projects/nspr/reference/html/prlog.html Regards, Peter

Re: dev-tech-crypto Digest, Vol 15, Issue 28

The subject of this ought to have been Re: Email certificate from TPM does not show up in Thunderbird ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

Re: dev-tech-crypto Digest, Vol 15, Issue 28

Funnily enough, one of the other applications that Infineon list as supporting their product is ... Netscape Communicator! http://www.infineon.com/cgi-bin/ifx/portal/ep/channelView.do?channelId=-84614channelPage=%2Fep%2Fchannel%2FproductOverview.jsppageTypeId=17099 Regards, Peter

Re: dev-tech-crypto Digest, Vol 15, Issue 28

Subject: Email certificate from TPM does not show up in Thunderbird (or My shy certificate revisited) From:Stephen Gryphon [EMAIL PROTECTED] Date:Fri, 30 Mar 2007 11:00:13 +1000 To: dev-tech-crypto@lists.mozilla.org G?day, I am suffering from what appears to

Re: getting cred(pkcs12) out of mozilla?

Jana, pk12util is part of the utilities that ship with a NSS release. You can find the NSS Release Directory here: ftp://ftp.mozilla.org/pub/mozilla.org/security/nss/releases/ You have a choice between downloading source code or pre-compiled binaries. If you download the source code, you can

Re: getting cred(pkcs12) out of mozilla?

Jana, If you look in the JSS Release directory, JSS 3.4 has a pre-compiled release for Linux2.4, as well as source code. JSS 4.2 has pre-compiled releases for Linux 2.4 and 2.6. I don't know why JSS 3.7 has only a built for the HP-UXB platform. Would these work for you? If you want to use

Re: Mozilla certificate database on windows?

Jana, I found mine at: C:\Documents and Settings\my-user-name\Application Data\Mozilla \Firefox\Profiles\random-profile-name Whatever the directory is, you need the following files: cert8.db, cert3.key and secmod.db (certificate db, key db, security module) Regards, Peter

Re: Mozilla certificate database on windows?

cert3.key and secmod.db (certificate db, key db, security module) I meant key3.db, cert8.db and secmod.db Regards, Peter ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

Re: Mozilla certificate database on windows?

Jana, These dynamic libraries are NSPR libraries and I'm surprised that they are not part of the NSS release. You can get a NSPR release from: ftp://ftp.mozilla.org/pub/mozilla.org/nspr/releases I guess you downloaded a pre-compiled NSS package, right? If you compile NSS from scratch, I think

Re: Implementing XML-based security protocols in Firefox

Sure Anders, Here is an additional piece of documentation: http://www.xulplanet.com/references/xpcomref/group_XMLSchema.html The implementation of this interface is in nsSchema: http://lxr.mozilla.org/mozilla1.7/source/extensions/webservices/schema/src/nsSchema.cpp Apparently, Mozilla has

Re: Implementing XML-based security protocols in Firefox

...and some documentation on the Mozilla SOAP scriptable API: http://lxr.mozilla.org/mozilla1.8/source/extensions/webservices/docs/Soap_Scripts_in_Mozilla.html Regards, Peter Djalaliev ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org

trying to verify a RSA signature

Hello, I think I am missing something here because this really should work. So, I have four buffers of unsigned bytes: a digest, the digst's signature, a public key modulus, and a public key exponent. The digest was signed using the private key corresponding to that public key. I tried to

Re: trying to verify a RSA signature

Found it. Thanks anyway. ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

Re: Debugging a SSL transaction on Firefox 2.0.x

. embedded apps) Given this, we can also write a nice documentation module on debugging NSS in the NSS documentation wiki that Wan-Teh maintains. What do you all think about that? Regards, Peter Djalaliev On Mar 8, 12:16 pm, Pedro DeKeratry [EMAIL PROTECTED] wrote: Our device implements

Re: Sftp extension for Firefox

Here are my doubts about the easiness of this: Let's say you use the FTP implementation in Firefox (I assume you want to do this). When the you attempt to initiate a FTP connection, Firefox's FTP code will attempt toe establish a socket connection using the socket objects provided by

Re: Sftp extension for Firefox

Has Secure FTP been standardized? I can't seem to find any sort of an RFC or another standard to do with Secure FTP. Is the sFTP you are talking about more than just FTP over an SSL connection? If it is not, then it might not be too hard to implement this secure FTP through an extension. If

Re: checking if a file descriptor corresponds to a TCP or TLS connection

When you call GetUniqueIdentifier(), does unique imply that the identifier will be unique for that layer name, or that the identifier will just be unique among the identities issued so far? I wasn't familiar with the semantics of this function and assumed the latter. I agree that the

Re: checking if a file descriptor corresponds to a TCP or TLS connection

Sorry for replying late on this thread... ssl_FindSocket first checks the identity of the layer, so it seems that it should fail with the PR_BAD_DESCRIPTOR_ERROR if it receives a non-TLS PRFileDesc:http://lxr.mozilla.org/seamonkey/source/security/nss/lib/ssl/sslsock 220

Re: checking if a file descriptor corresponds to a TCP or TLS connection

Well, Due to the changes I made to my code I am now able to do it. I am doing this from within the mod_nss Apache module, which actually stores a pointer to the SSL file descriptor when it layers it upon the original TCP fd. Otherwise, if I weren't to do within mod_nss, Nelson's idea about

checking if a file descriptor corresponds to a TCP or TLS connection

Hello, Given only a NSPR file descriptor, what would be the correct way to see if: 1) TLS functionality has been layered on the file descriptor stack? 2) The TLS handshake has been done with the host on the remote end of the socket connection? For 1), can I use the PR_GetDescType() from the

Re: Building NSS on Ubuntu Linux?

Yeah, I agree with David. I don't know if there is a make install feature in the NSS build system (I don't remember, I think I couldn't find one when I first started using NSS). I copy the header files and shared libraries of my NSS builds manually to /usr/include and /usr/lib. After you build

Re: Accessing NSS methods from a Firefox extension's XPCOM DLL

I think the answer to your question is that you can only sign and verify text (for signing forms), as well as generate RSA keys (for the key generation mechanism embedded in HTML). Most NSS functionality is not exported to Firefox. Any functionality that is exported is done so through the PSM

using framework for client/server hello extensions

Hello, I see that NSS 3.11 features a framework (in development) for client hello extensions, but server hello extensions are not used. Are there plans to extend the framework to allow server hello extensions, too? Regards, Peter ___ dev-tech-crypto

Re: Import PKCS#12 cert into FF

John, There is a password for the Mozilla store, which protects the private keys stored inside. The password is used to wrap a private key, which is then used to access the private keys stored in the Mozila db. I don't remember the exact details, do you need them? There might be additional

Re: NSS_SetDomesticPolicy() return 12266

Alex, mod_nss is available here : http://directory.fedora.redhat.com/wiki/Mod_nss and some additional documentation is available here: http://directory.fedora.redhat.com/docs/mod_nss.html If you are running mod_ssl and are using NSS's crypto library, you might be much better off running

Re: question about gathering data in SSL

Nelson, Thanks for the great reply! This gives me a much better idea of the behavior of the code :) I erroneously assumed the connection between SECWouldBlock and PR_WOULD_BLOCK_ERROR (I read the notes.txt file a while ago...). Also, a function in the NSS code would in some cases check if a

Re: NSS_SetDomesticPolicy() return 12266

Alex, I think this basically means that NSS_SetDomesticPolicy() or NSS_SetExportPolicy() is trying to flag an unimplemented cipher as SSL_ALLOWED. You shouldn't be getting this error because the table in sslsock.c that NSS_SetDomesticPolicy() / NSS_SetExportPolicy() get all ciphers from contains

question about gathering data in SSL

Hello, I have a question about what happens when the first SSL handshake on a SSL connection tries to gather data from a socket that would block (e.g. there is no data tobe read yet). I am using the SSL3 implementation on a Linux platform. So, in Do1stHandshake(), the next handshake function is

Re: Updated build instructions / release notes

Hello, I don't know about the release notes (I suspect these are the newest ones), but here are some build instructions that should work. They haven't really changed as far as I know: 1) Download the NSS and NSPR source tarballs from their respective release dirs:

Re: handshake, next handshake, security handshake

Nelson, I completely agree with you about this not having to be a priority. Code transparency is definitely important, ambivalent/outdated code can cause the introduction of new bugs down the road, but it's eventually up to the NSS maintainers/contributors to decide what has higher priority.

Re: handshake, next handshake, security handshake

Helson, Thanks for you reply :) What is the ss-securityHandshake used for? I think it may now be unused. I think it was part of the old SOCKS implementation that was abandoned. This probably doesn't matter as much, but ssl_SecureConnect sets the handshake function using

handshake, next handshake, security handshake

Hello, I have a question about something I don't understand in the SSL implementation of NSS. When ssl_Do1stHandshake is called, it checks three handshake function pointers in the sslSocket struct: handshake, nextHandshake and securityHandshake. What is the difference between the three? I can

Re: encrypt with firefox (and other mozilla browsers)

Hello Carlos, I don't claim to be the authority on this, but you can read the last three messages of this thread: http://groups.google.com/group/mozilla.dev.tech.crypto/browse_thread/thread/298e0723330e9697 From what Anders and Kyle say, I get the impression that signText is the only

Re: My shy certificate

Nelson Bolyard написа: I would expect that these details all go on beneath the PKCS#11 API layer, and are all hidden inside of the PKCS#11 module. I suspect that the wrapped keys (wherever they physically reside) still appear as PKCS#11 objects in the PKCS#11 slot or token, and would be

Re: My shy certificate

Oh, well, I understood that Dave used his Mozilla browser only to navigate to the CA website and click the Buy Now button, not to generate his own private key and CSR. Can Firefox generate private keys? I though that none of the NSS functionality (except for signing and verifying text) was

Re: My shy certificate

Nelson Bolyard wrote: You generated the key pair on a PC that didn't have the TPM chip. So the private key couldn't have been generated in the TPM chip, and when you generated it, mozilla (FF/TB/SM) didn't ask you which device you wanted to use to generate the keypair because, on that

Re: My shy certificate

More information on how the TPM enables protected storage can be found starting on p. 145 of the TCPA specification (v. 1.1): https://www.trustedcomputinggroup.org/specs/TPM/TCPA_Main_TCG_Architecture_v1_1b.pdf Regards, Peter ___ dev-tech-crypto

Re: My shy certificate

More information on how the TPM enables protected storage can be found starting on p. 145 of the TCPA specification (v. 1.1): https://www.trustedcomputinggroup.org/specs/TPM/TCPA_Main_TCG_Architecture_v1_1b.pdf Regards, Peter ___ dev-tech-crypto

Re: My shy certificate

ftp://ftp.compaq.com/pub/products/security/embedded_security_-_implementation.pdf ...and as the ProtectTools implementation white-paper explains, their Embeded Security Manager uses the TPM to create wrapping keys, which are then used to encrypt the private keys of the user. The wrapped keys are

Re: NSS Apache module - mod_nss

I am modifying mod_nss to implement TLS upgrades (RFC2817) to use in a special-purpose web client-server system. In fact, I think the modifications to mod_nss are done, but I am not yet done with implementing TLS upgrades in Firefox, so I haven't tested the mod_nss modifications. As we discussed

Re: dev-tech-crypto Digest, Vol 5, Issue 13

So there is patch for including the Hello extension? Where can I find it? Is it in some kind of a nightly build or in the CVS directory? Thanks :))) Peter ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org

Re: Server Name Indication and ECC

Before, I heard somebody talking on this mailing list about developing a framework for using TLS extensions within the NSS implementation of TLS. Does anybody know how this is going? Do the ECC cipher suites for TLS and the Server Name Indication mentioned about use this framework or do they use

SSL_RedoHandshake() calls SendHelloRequest(); SSL_ForceHandshake doesn't

Hello, I see in the NSS code that RedoHandshake, for example, sends a Hello request handshake message, but some handshake-initiating functions (e.g. ForceHandshake) do not. Instead, the server just starts waiting to receive data from the client (I assume a Client hello message). Why is the

SSL_ConfigMPServerSIDCache calls LaunchLockPoller on a NULL cache

Hello, I am having an issue with SSL_ConfigMPServerSIDCache on a Fedora Core 5 with 2.6.16 kernel.On a Unix machine, SSL_ConfigMPServerSIDCache calls LaunchLockPoller, which starts LockPoller() in another thread to poll the cache for expired locks. However, the cache pointer that LockPoller

Re: assertion failure in ssl3_config_match_init() when invoking ForceHandshake()

I think this deserves a new topic :) ___ dev-tech-crypto mailing list dev-tech-crypto@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-crypto

SSL/TLS upgrades - RFC2817

The goal of this topic to get a wide range of opinions about the current status/problems of SSL/TLS upgrades. I saw an old, very long discussion about TLS upgrades as specified by the much disliked RFC2817 :) I have been meaning to post something here to see if the same arguments and concerns

Re: using client SSL certificate with Firefox

Well, let me see if I can explain it better:The Apache web server is sending a certificate request handshake message. The client receives it, calls the certificate callback function (which fails) and then sends and empty certificate, which is handled on the server side. (I was wrong before when I

Re: new to mozilla nss

Hello Aditya,What problems exactly are you having?Did you see the Build documentation?http://www.mozilla.org/projects/security/pki/nss/nss-3.9/nss-3.9-build.html You can either follow this documentation or you can just download the NSPR and NSS source code tarballs

RSA signature and verification under NSS and OpenSSL

Hello,Has anybody tried to verify under NSS the signature of data signed under OpenSSL and vice versa? Assuming the same RSA public key (modulus and public exponent) and the same signature algorithm (RSA signature with PKCS#1 padding and SHA-1 message digest), we should be able to sign a buffer of