Hello,

Just to update the status for anyone who would stumble upon this in the future.

This issue has been addressed by https://github.com/wolfSSL/wolfssl/pull/1785 and it can be marked as solved.

Best regards,
Martin

Hi John,

Thank you for the explanation. I can see that doing it this way simplifies things, and that if reasonable settings are used, all should be working fine thanks to the cache miss/corruption tolerance.

However, now I ran into wolfSSL_i2d_SSL_SESSION which should mimic OpenSSL's i2d_SSL_SESSION (https://www.openssl.org/docs/man1.0.2/s … SSION.html), I assume.
According to the OpenSSL documentation, I would write something like this if I wanted to export the session into a buffer:

    unsigned char buffer[BUF_SIZE];
    unsigned char* buf = buffer;
    int sizeOfExportedSession = wolfSSL_i2d_SSL_SESSION(session, NULL);
    if (sizeOfExportedSession <= BUF_SIZE) {
        // what if session data got changed in the meantime and more than BUF_SIZE is to be written now?
        int sizeOfExportedSession2 = wolfSSL_i2d_SSL_SESSION(session, &buf);
        assert(sizeOfExportedSession == sizeOfExportedSession2); // will it always pass?
    }

As I indicate in the comments in the code snippet, I am worried about the race condition potentially leading to an out-of-bounds write.

Do you have any suggestions how to mitigate this on the user's side (if there really is this race condition)? Or would it require some change in wolfSSL, maybe adding some reference counting to the session cache?

Best regards,
Martin

Hi Kaleb,

The test client I attached to https://www.wolfssl.com/forums/post4061.html#p4061 can also be used for reproducing this issue, just replace runTests.sh with runTests2.sh attached here (now the OpenSSL server is not restarted after each connection, so that resumption can really happen).
Edit: I attached the whole archive with fixed test client here, see below.

When running
runTests2.sh ticket
I get

INFO: UseSessionTicket requested
INFO: going to attempt connection
Session ID not reused; Successful resume.
Connected to server.
wolfSSL_get_chain_count returned: 1
wolfSSL_get1_session returned 0x7fc7dfcfb868
sizeOfExportedSession is 1284
INFO: going to attempt connection
INFO: session for resumption set to ssl
Session ID reused; Successful resume.
Connected to server.
wolfSSL_get_chain_count returned: 1
wolfSSL_get1_session returned (nil)
sizeOfExportedSession is -173
INFO: going to attempt connection
INFO: session for resumption set to ssl
Session ID reused; Successful resume.
Connected to server.
wolfSSL_get_chain_count returned: 1
wolfSSL_get1_session returned (nil)
sizeOfExportedSession is -173
...

So it looks like that when I use the previously stored pointer to the session, the session gets resumed (confirmed by observing the traffic in Wireshark), but something gets broken, and wolfSSL_i2d_SSL_SESSION stops working.

Edit: The negative values returned by wolfSSL_i2d_SSL_SESSION were due to an error in my code, I was calling wolfSSL_i2d_SSL_SESSION on the returned NULL pointer instead of on the previously strored pointer. After fixing this, the correct value is returned.
I updated the attachment with the fixed code. To reproduce this " wolfSSL_get1_session returns NULL" issue, run demo2.

Best regards,
Martin

Hi Kaleb,

I might be able to send you a simple test application and script tomorrow or next week.

Currently, I am not using anything special for the testing. Just

openssl s_server -status_verbose -key testKey.pem -cert testCert.pem -accept <port_number>

and restarting manually by Ctrl+C and launching again.
OpenSSL version is 1.1.0h-fips (Fedora 27).

I don't think that the issue depends on ciphersuites used (maybe except for some PSK), I tested with TLS_DHE_RSA_WITH_AES_128_CBC_SHA and with TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256.

Best regards,
Martin