2 OpenSSL STATUS Last modified at
3 ______________ $Date: 1999/08/08 11:45:56 $
7 o OpenSSL 0.9.4: In preparation for release!
8 Proposed code freeze: August, 2nd 1999
9 Proposed release date: August, 9th 1999
10 Tarball Rolling is proposed to be done
11 by Ralf on dev.openssl.org at 1999-08-09, 14:00 CET
13 At least three OpenSSL developers should give their +1 vote
14 for OpenSSL 0.9.4 before the tarball should be actually rolled:
16 Bodo +1 (also tested Mingw32)
18 o OpenSSL 0.9.3a: Released on May 29th, 1999
19 o OpenSSL 0.9.3: Released on May 25th, 1999
20 o OpenSSL 0.9.2b: Released on March 22th, 1999
21 o OpenSSL 0.9.1c: Released on December 23th, 1998
27 o OCSP (titchenert@certco.com)
28 o getenv in ca.c and x509_def.c (jaltman@watsun.cc.columbia.edu)
32 o Steve is currently working on (in no particular order):
33 Proper (or at least usable) certificate chain verification.
34 Private key, certificate and CRL API and implementation.
35 Checking and bugfixing PKCS#7 (S/MIME code).
37 o Mark is currently working on:
38 Folding in any changes that are in the C2Net code base that were
39 not in the original SSLeay-0.9.1.b release. Plus other minor
42 o Ralf is currently working on:
43 1. Support for SSL_set_default_verify_paths(),
44 SSL_load_verify_locations(), SSL_get_cert_store() and
45 SSL_set_cert_store() functions which work like their existing
46 SSL_CTX_xxx() variants but on a per connection basis. That's needed
47 to let us provide full-featured per-URL client verification in
48 mod_ssl or Apache-SSL.
49 => It still dumps core, so I suspend this and investigate
50 again for OpenSSL 0.9.3.
51 2. The perl/ stuff to make it really work the first time ;-)
52 => I'll investigate a few more hours for OpenSSL 0.9.2
53 3. The new documentation set in POD format under doc/
54 => I'll investigate a few more hours for OpenSSL 0.9.2
55 4. More cleanups to get rid of obsolete/old/ugly files in the
56 source tree which are not really needed.
57 => Done all which were possible with my personal knowledge
59 o Ben is currently working on:
60 1. Function Prototype Thought Police issues.
61 2. Integrated documentation.
62 3. New TLS Ciphersuites.
63 4. Anything else that takes his fancy.
67 o salzr@certco.com (Rich Salz): Bug in X509_name_print
68 <29E0A6D39ABED111A36000A0C99609CA2C2BA4@macertco-srv1.ma.certco.com>
69 o $(PERL) in */Makefile.ssl
70 o "Sign the certificate?" - "n" creates empty certificate file
74 o internal_verify doesn't know about X509.v3 (basicConstraints
77 o The Makefile hierarchy and build mechanism is still not a round thing:
79 1. The config vs. Configure scripts
80 It's the same nasty situation as for Apache with APACI vs.
81 src/Configure. It confuses.
82 Suggestion: Merge Configure and config into a single configure
83 script with a Autoconf style interface ;-) and remove
84 Configure and config. Or even let us use GNU Autoconf
85 itself. Then we can avoid a lot of those platform checks
86 which are currently in Configure.
88 o Support for Shared Libraries has to be added at least
89 for the major Unix platforms. The details we can rip from the stuff
90 Ralf has done for the Apache src/Configure script. Ben wants the
91 solution to be really simple.
93 Status: Ralf will look how we can easily incorporate the
94 compiler PIC and linker DSO flags from Apache
95 into the OpenSSL Configure script.
97 o The perl/ stuff needs a major overhaul. Currently it's
98 totally obsolete. Either we clean it up and enhance it to be up-to-date
99 with the C code or we also could replace it with the really nice
100 Net::SSLeay package we can find under
101 http://www.neuronio.pt/SSLeay.pm.html. Ralf uses this package for a
102 longer time and it works fine and is a nice Perl module. Best would be
103 to convince the author to work for the OpenSSL project and create a
104 Net::OpenSSL or Crypt::OpenSSL package out of it and maintains it for
107 Status: Ralf thinks we should both contact the author of Net::SSLeay
108 and look how much effort it is to bring Eric's perl/ stuff up
112 o The EVP and ASN1 stuff is a mess. Currently you have one EVP_CIPHER
113 structure for each cipher. This may make sense for things like DES but
114 for variable length ciphers like RC2 and RC4 it is NBG. Need a way to
115 use the EVP interface and set up the cipher parameters. The ASN1 stuff
116 is also foo wrt ciphers whose AlgorithmIdentifier has more than just
117 an IV in it (e.g. RC2, RC5). This also means that EVP_Seal and EVP_Open
118 don't work unless the key length matches the fixed value (some vendors
119 use a key length decided by the size of the RSA encrypted key and expect
122 o Properly initialize the PRNG in the absence of /dev/random.
124 o ERR_error_string(..., buf) does not know how large buf is,
125 there should be ERR_error_string_n(..., buf, bufsize)