Future of Cyrus-SASL

[Quanah Gibson-Mount]

Hi all,

I started a small discussion on this in cyrus-sasl issue #433, but the 
discussion better belongs here. :)

While I know that Fastmail is taking on the Cyrus-IMAPD project now that 
CMU is divesting itself of it, there does not seem to be a strong plan 
around Cyrus-SASL.  My general understanding is that it is not something 
that Fastmail generally uses.  However, there are many other projects where 
cyrus-sasl is a critical part of the software.

Those include:
OpenLDAP
Postfix
Apache2 (Sasl module)
SSSD
autofs (ldap integration)
Squid proxy
qemu
memcached
Exim
PHP
mutt
389 directory server
subversion

It has been 5 years since the last release of Cyrus-SASL, and unfortunately 
it had a number of bugs.  There have been a large number of patches sent to 
the list in that time, as well as some 142 issues on the Github site (hard 
to know full correlation at this time).  There has been some work towards a 
2.1.27, but it seem stalled at this time.

I'm not sure if keeping SASL tied to the IMAPD project makes particular 
sense anymore, given the overall changes that have occurred.  Regardless of 
that particular question, the SASL project needs a dedicated development 
team to start churning through what has come in via the list in the last 
few years, as well as what's in github.

Generally, for the path going forward, perhaps an inquiry to the cyrus-sasl 
list to see if there are individuals willing to commit/join the SASL 
project?  I certainly would be willing to help with the release engineering 
portion, as I do for the OpenLDAP and Heimdal projects.

I look forward to hearing people's thoughts.

Warm regards,
Quanah

--

Quanah Gibson-Mount
Product Architect
Symas Corporation
Packaged, certified, and supported LDAP solutions powered by OpenLDAP:
<http://www.symas.com>