[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#531207: marked as done (initscripts: modes of execution for services)

Your message dated Mon, 19 Sep 2022 21:10:58 -0700
with message-id <877d1yit2l.fsf@hope.eyrie.org>
and subject line Re: Bug#531207: initscripts: modes of execution for services
has caused the Debian Bug report #531207,
regarding initscripts: modes of execution for services
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
531207: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=531207
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: initscripts
Version: 2.86.ds1-61
Severity: wishlist

Hi,
This is a wishlist request which I think might require an amendment in
the Debian Policy.

Currently, many of the services in a Debian installation are single
mode. By "single mode", I mean they are always running daemons. There do
are some (like exim), which are offered at install time, if a user wants to run them
from inted.

My proposal is that we have the init system with a machine type
identifier.
Example:
MACHINE=laptop

Currently, on my machine, I have many services running stand-alone as a
daemon. Example: cupsd. In a laptop/desktop environment, we don't do
regular printing. There is room for efficiency by providing the user
with an option to let as many daemons as possible, to be run from inetd.

This has a benefit. Especially for memory resources.
The dictionary server, dictd, is another example. Previously, it was
available as a standalone server. With multiple dictionary databases
installed, I noticed that memory consumption could rise to around
100Megs. So, in standalone mode, ditcd alone would be consuming a 100Meg
resource.

d-i currently does detect hardware type during installation. We should
have a global system variable defining the type of machine. And then, we
should have a policy stating, "Every application, if possible to
inet-ize, should have that option."
And then that could be triggered with the global system variable.

crond
dictd
many more daemons....

For laptop/destkop, we can save a lot of memory resource this way.


Thanks,
Ritesh

-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'unstable'), (1, 'experimental')
Architecture: i386 (i686)

Kernel: Linux 2.6.29-custom (SMP w/1 CPU core; PREEMPT)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages initscripts depends on:
ii  debianutils                  3.1.3       Miscellaneous utilities specific t
ii  e2fsprogs                    1.41.3-1    ext2/ext3/ext4 file system utiliti
ii  libc6                        2.9-12      GNU C Library: Shared libraries
ii  lsb-base                     3.2-22      Linux Standard Base 3.2 init scrip
ii  mount                        2.13.1.1-1  Tools for mounting and manipulatin
ii  sysvinit-utils               2.86.ds1-61 System-V-like utilities

Versions of packages initscripts recommends:
ii  psmisc                        22.7-1     utilities that use the proc file s

initscripts suggests no packages.

-- no debconf information

--- End Message ---
--- Begin Message --- 
Sean Whitton <spwhitton@spwhitton.name> writes:
> On Thu 21 Feb 2019 at 05:55PM +00, Dmitry Bogatov wrote:

>> I support your proposal, but would not definitely volonteer to drive
>> this initiative to get included into Policy. Either way, reassigning to
>> debian-policy.

> We would want to see this implemented in the archive before
> standardising it in Policy.  Please consider reassigning the bug back to
> the package in which it might be implemented, unless you have some
> reason why the Policy change absolutely has to happen first.

This was a fairly old bug (originally opened in 2009) requesting some sort
of global Debian setting to run as many daemons as possible via inetd or
the equivalent instead of starting them as separate daemons.

While this is an interesting idea, it's not the sort of thing that Policy
can tackle.  It's more of a debian-devel discussion that would need to be
followed by a design if it seemed popular and then some implementation
work, and I can think of a few challenges that would need to be overcome.

If someone wants to drive this forward, they are most welcome, but it's
not really actionable in Policy until that's been done.  Since the bug has
been tagged moreinfo for more than three years now, I'm going to go ahead
and close it as wontfix.

-- 
Russ Allbery (rra@debian.org)              <https://www.eyrie.org/~eagle/>

--- End Message ---
Reply to: