web/deps/dep9.mdwn

[[!meta title="DEP-9: inet-superserver configuration by maintainer scripts"]]

    Title: inet-superserver configuration by maintainer scripts
    DEP: 9
    State: DRAFT
    Date: 2012-02-05
    Drivers: Serafeim Zanikolas <sez@debian.org>
    URL: http://dep.debian.net/deps/dep9
    License: http://www.jclark.com/xml/copying.txt
    Abstract:
     Motivation, requirements and functional overview of a successor
     configuration tool for inetd superservers.

[[!toc ]]

# Introduction

The inet-superserver facility (typically provided by the openbsd-inetd
package) listens to certain sockets and invokes the right server upon the
arrival of an incoming request. Servers that are meant to be invoked by inetd
must add a service entry to /etc/inetd.conf upon package installation (and
remove the entry upon package purge). Maintainer scripts that modify
inetd.conf must currently do so using a utility called update-inetd, as per
Policy 11.2.

However, update-inetd has a problematic interface that leads to several kinds
of bugs, including cross-package ones. Fixing update-inetd is largely a matter
of fixing its interface, which would break backwards-compatibility. With that
cost as a given (ie. the cost of revising all maintainer scripts of
update-inetd's reverse-depends), one might as well create a successor tool
with a cleaner interface.

This DEP proposes such a successor inetd configuration tool, hereafter called
reconf-inetd. reconf-inetd, in addition to providing the existing functionality,
must meet the following requirements:

* the standard configuration files of inetd and xinetd must remain the
  authoritative files;
* the solution must not change the way system administrators configure inetd,
  and must have no impact on maintainers of "Provides: inet-superserver"
  packages;
* reconf-inetd must be capable of co-existing with update-inetd, so as
  to allow a gradual transition.

Note that the first requirement above implies that inetd.conf entries that
have been (i) added by reconf-inetd and (ii) were subsequently modified by the
user, shall not be removed even on package purge.

# Motivation

The main problem of update-inetd is that the service entry to be enabled,
disabled or removed, is selected in terms of a service name, such as "ftp".
This limitation typically leads to cross-package bugs because of the
difficulty to distinguish between independent implementations (eg. ftpd-ssl
versus proftpd), or an ipv4 versus an ipv6 implementation of the same kind of
service. As an example, in #168847, ftpd-ssl's invocation of update-inetd
enables the service entry of (the previously uninstalled) proftpd because both
packages' entries have an "ftp" service name.

As of now, one can try to avoid the above problem as follows. A maintainer
script may (i) further specify the acted-upon service entry in terms of a
regular expression (which is matched against the whole service entry, instead
of just the service name); (ii) override the default comment-prefix
("#&lt;off&gt;# "), to distinguish service entries of different packages that provide
the same kind of service (eg. "#&lt;off-ftpd-ssl-ipv4&gt;# " vs
"#&lt;off-proftpd-ipv4&gt;# "). These features work when used correctly, but at the
cost of fragile logic in maintainer scripts.

To summarise, the interface of update-inetd is inadequate in that it does not
require that the following three elements are specified to select an entry:
service name, protocol type, and path to server program.

A secondary issue, but nevertheless one that would be nice to solve, is
support for configuration updates of xinetd (#8927). xinetd has a relatively
low popcon but one could argue that may be the case because it is not well
integrated in Debian.


# Outline of reconf-inetd operation

reconf-inetd will operate similarly to update-inetd, ie. add, remove, enable and
disable service entries in /etc/inetd.conf. The main difference is that where
update-inetd uses command-line arguments, reconf-inetd will use xinetd.conf(5)
configuration fragments under /usr/share/reconf-inetd (hereafter called
reconf-inetd fragments, for brevity). Moreover, reconf-inetd will be invoked via a
dpkg trigger, as opposed to from maintainer scripts.

With reconf-inetd, inet-superserver configuration will occur as follows:

* reconf-inetd will provide the directory /usr/share/reconf-inetd, and declare a
  dpkg trigger for that directory
* server packages that can use inetd, will depend on reconf-inetd and install
  a reconf-inetd fragment in /usr/share/reconf-inetd
* reconf-inetd will be invoked via a dpkg trigger to update /etc/inetd.conf
  using the fragments in /usr/share/reconf-inetd

An inetd.conf service entry will be considered to be "matching" a reconf-inetd
fragment when the following fields are equal: service name, protocol, and
server program. In reconf-inetd fragments with "flags = NAMEINARGS" (eg. a tcpd
service entry), the actual server path will be extracted from the server_args
field, as per xinetd.conf(5).

Upon adding an entry to inetd.conf, reconf-inetd will make a shadow copy of the
related reconf-inetd fragment under /var/lib/reconf-inetd. This would allow to
determine the following after the uninstallation of the related package:

* to identify inetd.conf entries that have been added by reconf-inetd (as
  opposed to update-inetd or a user)
* to determine whether an inetd.conf entry, that has been previously added by
  reconf-inetd, is intact or has had local modifications

All reconf-inetd fragments will be considered enabled, regardless of the value
of the "disable" field, if such field exists. Servers that support inetd mode
but default to standalone operation must therefore not install a reconf-inetd
fragment (and should instead provide a sample inetd.conf entry that the user
has to add manually to /etc/inetd.conf).

/etc/inetd.conf (and /etc/xinetd.d) will remain the authoritative
configuration file for inetd (and xinetd). reconf-inetd fragments are only meant
for updating the standard configuration files.

An invocation of reconf-inetd (because of a fragment being installed or removed
from /usr/share/reconf-inetd) may result in a modification of /etc/inetd.conf
as summarised in the table below.

<pre>

   | server  | status of  | matching     | shadow     | reconf-inetd
   | program | inetd.conf | reconf-inetd | fragment   | action
   | exists  | entry      | fragment     | status     |
---+---------+------------+--------------+------------+-----------
  0|      no |   disabled |         no   |  identical |    remove
  1|      no |    enabled |         no   |  identical |    remove
---+---------+------------+--------------+------------+-----------
  2|     yes |   disabled |        yes   |  different |    enable
  3|     yes |   disabled |        yes   |  identical |    enable
---+---------+------------+--------------+------------+-----------
  4|     yes |    missing |        yes   |        n/a |       add
---+---------+------------+--------------+------------+-----------
  5| commented-out inetd.conf entry                   |      none
---+---------+------------+--------------+------------+-----------
  6| any other combination                            |      none

</pre>

An inetd.conf entry is considered disabled when it starts with "#&lt;off&gt;# ",
 and disabled by a user when commented-out simply with '#'.

A shadow fragment status (ie. the fragment under /var/lib/reconf-inetd) is
considered identical to a matching inetd.conf entry, by comparing the
server arguments, if any.

Follows a detailed description of the aforementioned scenarios.

* an non-commented-out inetd.conf entry will be removed, regardless of whether
  it is enabled or disabled, when it:
    - refers to a non-existing server file
    - has no matching reconf-inetd fragment
    - is identical to a matching shadow fragment

* a disabled inetd.conf entry will be enabled when it
    - refers to an existing server file
    - it has a matching reconf-inetd fragment
    - and a matching shadow fragment (identical or not)

* a new inetd.conf entry will be added when there exists a reconf-inetd fragment
  that:
    - refers to an existing server file
    - has no matching (enabled, disabled, commented-out or not) inetd.conf
      entry
    - has a matching reconf-inetd fragment

Meaning of the above listed actions:

- add: add inetd.conf entry and a matching shadow fragment
- remove: remove inetd.conf entry and matching shadow fragment
- enable: enable inetd.conf entry

# Transition of "Depends: update-inetd" packages

A time-limited transition is not a strict requirement, since reconf-inetd and
update-inetd can co-exist without problems.

Server packages that depend on update-inetd can be converted as follows:

* remove any references of update-inetd in post{inst,rm} scripts;
* install a reconf-inetd fragment in /usr/share/reconf-inetd, as a regular
  file (ie. not as a conffile);
* replace the dependency on update-inetd with one on reconf-inetd (but see
  notes below)

reconf-inetd will not touch any entries that have not been added by itself,
including entries added by update-inetd. Thus, a server package that is meant
to transition from update-inetd to reconf-inetd, must remove any entries that
it previously added using update-inetd.

Below is an example postinst snippet for an ftp server package that is meant
to be transitioned from update-inetd to reconf-inetd in release 2.0:

<pre>
    case $1 in
        configure)
            if dpkg --compare-versions "$2" le 2.0; then
                if which update-inetd; then
                    update-inetd --remove ftp
                fi
            fi
        ;;
    esac
</pre>

# Requirements for xinetd/reconf-inetd fragments

According to xinetd.conf(5), xinetd fragments must have the following fields:

<pre>
  socket_type       (mandatory)
  wait              (mandatory)
  user              (non-internal services only)
  server            (non-internal services only)
  protocol          (RPC and unlisted services only)
  port              (unlisted non-RPC services only)
</pre>

If the protocol field is omitted and the service is listed, reconf-inetd will
assume the protocol of the first matching entry from /etc/services. That will
be tcp or udp, which currently implies IPv4, so if the intention is IPv6, then
tcp6 or udp6 should be explicitly specified in the protocol field.


# xinetd support

xinetd configuration with reconf-inetd will become trivial because server
packages will have to ship reconf-inetd (ie. xinetd.conf(5) compatible)
fragments anyway.

Synchronisation between inetd.conf and reconf-inetd fragments is outside this DEP's
scope. The fact that inetd.conf (and /etc/xinetd.d) remains the authoritative
configuration file implies that any synchronisation requires a user-initiated
action, and is thus best implemented as a separate tool.

# Clarification of behaviour on package removal versus package purge

According to rules 0 and 1 in the previously listed table, reconf-inetd does
not distinguish between package removal and purge. In other words,
/etc/inetd.conf entries that were added by reconf-inetd will be removed upon
package removal, as long as they have not been locally modified. Similarly, a
locally-modified inetd.conf entry will not be removed even if the associated
package is purged. This is in line with Debian policy, on the basis that
inetd.conf is a configuration file, but not a conffile.

# Current status

20120206: upload of reconf-inetd to experimental

# Source code repo

http://git.debian.org/?p=collab-maint/reconf-inetd.git;a=summary
1	sez	171	[[!meta title="DEP-9: inet-superserver configuration by maintainer scripts"]]
2	sez	170
3			Title: inet-superserver configuration by maintainer scripts
4	sez	171	DEP: 9
5	sez	170	State: DRAFT
6	sez	263	Date: 2012-02-05
7	sez	170	Drivers: Serafeim Zanikolas <sez@debian.org>
8	sez	172	URL: http://dep.debian.net/deps/dep9
9	sez	170	License: http://www.jclark.com/xml/copying.txt
10			Abstract:
11			Motivation, requirements and functional overview of a successor
12			configuration tool for inetd superservers.
13
14			[[!toc ]]
15
16			# Introduction
17
18			The inet-superserver facility (typically provided by the openbsd-inetd
19			package) listens to certain sockets and invokes the right server upon the
20			arrival of an incoming request. Servers that are meant to be invoked by inetd
21			must add a service entry to /etc/inetd.conf upon package installation (and
22			remove the entry upon package purge). Maintainer scripts that modify
23			inetd.conf must currently do so using a utility called update-inetd, as per
24			Policy 11.2.
25
26			However, update-inetd has a problematic interface that leads to several kinds
27			of bugs, including cross-package ones. Fixing update-inetd is largely a matter
28			of fixing its interface, which would break backwards-compatibility. With that
29			cost as a given (ie. the cost of revising all maintainer scripts of
30			update-inetd's reverse-depends), one might as well create a successor tool
31			with a cleaner interface.
32
33			This DEP proposes such a successor inetd configuration tool, hereafter called
34			reconf-inetd. reconf-inetd, in addition to providing the existing functionality,
35			must meet the following requirements:
36
37			* the standard configuration files of inetd and xinetd must remain the
38			authoritative files;
39			* the solution must not change the way system administrators configure inetd,
40			and must have no impact on maintainers of "Provides: inet-superserver"
41			packages;
42			* reconf-inetd must be capable of co-existing with update-inetd, so as
43			to allow a gradual transition.
44
45			Note that the first requirement above implies that inetd.conf entries that
46			have been (i) added by reconf-inetd and (ii) were subsequently modified by the
47			user, shall not be removed even on package purge.
48
49			# Motivation
50
51			The main problem of update-inetd is that the service entry to be enabled,
52			disabled or removed, is selected in terms of a service name, such as "ftp".
53			This limitation typically leads to cross-package bugs because of the
54			difficulty to distinguish between independent implementations (eg. ftpd-ssl
55			versus proftpd), or an ipv4 versus an ipv6 implementation of the same kind of
56			service. As an example, in #168847, ftpd-ssl's invocation of update-inetd
57			enables the service entry of (the previously uninstalled) proftpd because both
58			packages' entries have an "ftp" service name.
59
60			As of now, one can try to avoid the above problem as follows. A maintainer
61			script may (i) further specify the acted-upon service entry in terms of a
62			regular expression (which is matched against the whole service entry, instead
63			of just the service name); (ii) override the default comment-prefix
64			("#<off># "), to distinguish service entries of different packages that provide
65			the same kind of service (eg. "#<off-ftpd-ssl-ipv4># " vs
66			"#<off-proftpd-ipv4># "). These features work when used correctly, but at the
67			cost of fragile logic in maintainer scripts.
68
69			To summarise, the interface of update-inetd is inadequate in that it does not
70			require that the following three elements are specified to select an entry:
71			service name, protocol type, and path to server program.
72
73			A secondary issue, but nevertheless one that would be nice to solve, is
74			support for configuration updates of xinetd (#8927). xinetd has a relatively
75	sez	242	low popcon but one could argue that may be the case because it is not well
76			integrated in Debian.
77	sez	170
78
79			# Outline of reconf-inetd operation
80
81			reconf-inetd will operate similarly to update-inetd, ie. add, remove, enable and
82			disable service entries in /etc/inetd.conf. The main difference is that where
83			update-inetd uses command-line arguments, reconf-inetd will use xinetd.conf(5)
84			configuration fragments under /usr/share/reconf-inetd (hereafter called
85			reconf-inetd fragments, for brevity). Moreover, reconf-inetd will be invoked via a
86			dpkg trigger, as opposed to from maintainer scripts.
87
88			With reconf-inetd, inet-superserver configuration will occur as follows:
89
90			* reconf-inetd will provide the directory /usr/share/reconf-inetd, and declare a
91			dpkg trigger for that directory
92			* server packages that can use inetd, will depend on reconf-inetd and install
93			a reconf-inetd fragment in /usr/share/reconf-inetd
94			* reconf-inetd will be invoked via a dpkg trigger to update /etc/inetd.conf
95			using the fragments in /usr/share/reconf-inetd
96
97			An inetd.conf service entry will be considered to be "matching" a reconf-inetd
98			fragment when the following fields are equal: service name, protocol, and
99			server program. In reconf-inetd fragments with "flags = NAMEINARGS" (eg. a tcpd
100			service entry), the actual server path will be extracted from the server_args
101			field, as per xinetd.conf(5).
102
103			Upon adding an entry to inetd.conf, reconf-inetd will make a shadow copy of the
104			related reconf-inetd fragment under /var/lib/reconf-inetd. This would allow to
105			determine the following after the uninstallation of the related package:
106
107			* to identify inetd.conf entries that have been added by reconf-inetd (as
108			opposed to update-inetd or a user)
109			* to determine whether an inetd.conf entry, that has been previously added by
110			reconf-inetd, is intact or has had local modifications
111
112			All reconf-inetd fragments will be considered enabled, regardless of the value
113			of the "disable" field, if such field exists. Servers that support inetd mode
114			but default to standalone operation must therefore not install a reconf-inetd
115			fragment (and should instead provide a sample inetd.conf entry that the user
116			has to add manually to /etc/inetd.conf).
117
118			/etc/inetd.conf (and /etc/xinetd.d) will remain the authoritative
119			configuration file for inetd (and xinetd). reconf-inetd fragments are only meant
120			for updating the standard configuration files.
121
122			An invocation of reconf-inetd (because of a fragment being installed or removed
123			from /usr/share/reconf-inetd) may result in a modification of /etc/inetd.conf
124			as summarised in the table below.
125
126			<pre>
127
128			\| server \| status of \| matching \| shadow \| reconf-inetd
129			\| program \| inetd.conf \| reconf-inetd \| fragment \| action
130			\| exists \| entry \| fragment \| status \|
131			---+---------+------------+--------------+------------+-----------
132			0\| no \| disabled \| no \| identical \| remove
133			1\| no \| enabled \| no \| identical \| remove
134			---+---------+------------+--------------+------------+-----------
135			2\| yes \| disabled \| yes \| different \| enable
136			3\| yes \| disabled \| yes \| identical \| enable
137			---+---------+------------+--------------+------------+-----------
138			4\| yes \| missing \| yes \| n/a \| add
139			---+---------+------------+--------------+------------+-----------
140			5\| commented-out inetd.conf entry \| none
141			---+---------+------------+--------------+------------+-----------
142			6\| any other combination \| none
143
144			</pre>
145
146			An inetd.conf entry is considered disabled when it starts with "#<off># ",
147			and disabled by a user when commented-out simply with '#'.
148
149			A shadow fragment status (ie. the fragment under /var/lib/reconf-inetd) is
150			considered identical to a matching inetd.conf entry, by comparing the
151			server arguments, if any.
152
153			Follows a detailed description of the aforementioned scenarios.
154
155			* an non-commented-out inetd.conf entry will be removed, regardless of whether
156			it is enabled or disabled, when it:
157			- refers to a non-existing server file
158			- has no matching reconf-inetd fragment
159			- is identical to a matching shadow fragment
160
161			* a disabled inetd.conf entry will be enabled when it
162			- refers to an existing server file
163			- it has a matching reconf-inetd fragment
164			- and a matching shadow fragment (identical or not)
165
166	sez	243	* a new inetd.conf entry will be added when there exists a reconf-inetd fragment
167	sez	170	that:
168			- refers to an existing server file
169			- has no matching (enabled, disabled, commented-out or not) inetd.conf
170			entry
171			- has a matching reconf-inetd fragment
172
173	sez	262	Meaning of the above listed actions:
174	sez	170
175	sez	262	- add: add inetd.conf entry and a matching shadow fragment
176			- remove: remove inetd.conf entry and matching shadow fragment
177			- enable: enable inetd.conf entry
178
179	sez	170	# Transition of "Depends: update-inetd" packages
180
181			A time-limited transition is not a strict requirement, since reconf-inetd and
182			update-inetd can co-exist without problems.
183
184			Server packages that depend on update-inetd can be converted as follows:
185
186			* remove any references of update-inetd in post{inst,rm} scripts;
187			* install a reconf-inetd fragment in /usr/share/reconf-inetd, as a regular
188			file (ie. not as a conffile);
189	sez	264	* replace the dependency on update-inetd with one on reconf-inetd (but see
190			notes below)
191	sez	170
192	sez	264	reconf-inetd will not touch any entries that have not been added by itself,
193			including entries added by update-inetd. Thus, a server package that is meant
194			to transition from update-inetd to reconf-inetd, must remove any entries that
195			it previously added using update-inetd.
196	sez	170
197	sez	264	Below is an example postinst snippet for an ftp server package that is meant
198			to be transitioned from update-inetd to reconf-inetd in release 2.0:
199
200	sez	266	<pre>
201	sez	264	case $1 in
202			configure)
203			if dpkg --compare-versions "$2" le 2.0; then
204			if which update-inetd; then
205			update-inetd --remove ftp
206			fi
207			fi
208			;;
209			esac
210	sez	266	</pre>
211	sez	264
212	sez	265	# Requirements for xinetd/reconf-inetd fragments
213	sez	264
214	sez	265	According to xinetd.conf(5), xinetd fragments must have the following fields:
215
216	sez	266	<pre>
217	sez	265	socket_type (mandatory)
218			wait (mandatory)
219			user (non-internal services only)
220			server (non-internal services only)
221			protocol (RPC and unlisted services only)
222			port (unlisted non-RPC services only)
223	sez	266	</pre>
224	sez	265
225			If the protocol field is omitted and the service is listed, reconf-inetd will
226			assume the protocol of the first matching entry from /etc/services. That will
227			be tcp or udp, which currently implies IPv4, so if the intention is IPv6, then
228			tcp6 or udp6 should be explicitly specified in the protocol field.
229
230
231	sez	170	# xinetd support
232
233			xinetd configuration with reconf-inetd will become trivial because server
234			packages will have to ship reconf-inetd (ie. xinetd.conf(5) compatible)
235			fragments anyway.
236
237			Synchronisation between inetd.conf and reconf-inetd fragments is outside this DEP's
238			scope. The fact that inetd.conf (and /etc/xinetd.d) remains the authoritative
239			configuration file implies that any synchronisation requires a user-initiated
240			action, and is thus best implemented as a separate tool.
241
242	sez	264	# Clarification of behaviour on package removal versus package purge
243	sez	170
244			According to rules 0 and 1 in the previously listed table, reconf-inetd does
245			not distinguish between package removal and purge. In other words,
246			/etc/inetd.conf entries that were added by reconf-inetd will be removed upon
247	sez	264	package removal, as long as they have not been locally modified. Similarly, a
248			locally-modified inetd.conf entry will not be removed even if the associated
249			package is purged. This is in line with Debian policy, on the basis that
250			inetd.conf is a configuration file, but not a conffile.
251	sez	170
252	sez	264	# Current status
253	sez	241
254	sez	264	20120206: upload of reconf-inetd to experimental
255
256	sez	241	# Source code repo
257
258			http://git.debian.org/?p=collab-maint/reconf-inetd.git;a=summary