I have documented some of update-inetd's design issues in

    http://lists.debian.org/debian-devel/2009/09/msg00611.html

and, with the help of Roger Leigh, made a proposal for a cleaner solution (the
above url has a second, much simplified version of the original proposal).

However, there has been no consensus on the proposal. In the meantime, I don't
see a clean way forward for dealing with most of update-inetd's pending bugs.

-- Serafeim Zanikolas  Sat, 13 Feb 2010 19:29:44 +0100

#24043: do not only look at service name, but also to protocal (tcp/udp)
#33797: seems to ask for replace instead of add?
#53396: debconf messages still confusing?
#64874: improve documentation?
#168847: improve documentation/add replace?
#179318: close as debconf is used since some time...
#207892: merge with #53396?
#245517: don't act if the conf file doesn't exist
#322528: investigate issue by reproducing given test case
#430849: implement hostname/address prefix mgmt
#161532: implement spawning rate support
#212540: improve documentation about update-inetd not being a sysadmin tool
#261711: merge with #53396?
#307697: merge with #53396?
#311111: improve documentation about --remove?
#344139: close as debconf is used since some time...
#374542: document support for --group combined with --add or --remove?
#123929: include po4a support for the manpage?
#45334: merge with #161532?
#63210: investigate if disabling with one kind of comment and enabling with another would have an acceptable result
#74339: implement status inquiry
#82875: merge with #74339?
#159366: merge with #33797?
#236917: close as proper debconf support is enabled?