| 1 |
Maintainer notes for sysvinit |
| 2 |
|
| 3 |
News |
| 4 |
---- |
| 5 |
Last updated 16 May 2009 |
| 6 |
|
| 7 |
Stuff to implement |
| 8 |
------------------ |
| 9 |
* Do not allow system to continue booting if both fsck and sulogin fail (#337444)? |
| 10 |
* Make sulogin try /bin/sh and /bin/sash too (#40144, #43317)? |
| 11 |
* Port bootlogd to use posix_openpt(3), since BSD PTs are now deprecated and |
| 12 |
/dev/pts _is_ now available at the time that bootlogd starts (...in Debian)? |
| 13 |
|
| 14 |
Stuff to do after etch |
| 15 |
---------------------- |
| 16 |
* mountnfs.sh: Don't start portmap |
| 17 |
|
| 18 |
Ubuntu changes to backport |
| 19 |
-------------------------- |
| 20 |
(none known ATM) |
| 21 |
|
| 22 |
Stuff to send upstream |
| 23 |
---------------------- |
| 24 |
debian/patches/fix-minor-warnings should be applied upstream. |
| 25 |
So should a lot of the other patches. |
| 26 |
|
| 27 |
src/Makefile should not set CC. |
| 28 |
|
| 29 |
set -e |
| 30 |
------ |
| 31 |
Note that many scripts do not use "set -e". This is deliberate. |
| 32 |
Instead, return statuses are checked. |
| 33 |
|
| 34 |
PATH |
| 35 |
---- |
| 36 |
* Current policy: |
| 37 |
+ Leave unset if absolutely no pathless commands are run; otherwise... |
| 38 |
+ set to "/sbin:/bin" in /etc/init.d/rc and in scripts that run prior |
| 39 |
to mountnfs.sh; otherwise... |
| 40 |
+ set to "/usr/sbin:/usr/bin:/sbin:/bin". |
| 41 |
+ Add "/lib/init" to beginning of PATH if readlink command or domount |
| 42 |
function used. This will disappear once coreutils 5.94 makes it |
| 43 |
into testing, and we Depend on that. |
| 44 |
* Question: Should we include /usr/local/sbin and /usr/local/bin too? |
| 45 |
I don't see this question addressed in Policy. Ref: #227540 |
| 46 |
|
| 47 |
Dependencies |
| 48 |
------------ |
| 49 |
To go away: |
| 50 |
* sysvinit Pre-Depends on initscripts, sysv-rc | file-rc so that the |
| 51 |
conffiles in the old monolithic sysvinit package are taken over by |
| 52 |
the initscripts package. (Only a Pre-Depends is strong enough to |
| 53 |
ensure that the conffiles are taken over without an orphanic hiatus.) |
| 54 |
Strictly speaking we can drop this dependency now since everyone is |
| 55 |
either installing anew or else has upgraded to sarge; and Debian does |
| 56 |
not support skip upgrades. However, I don't see any harm in keeping |
| 57 |
this dependency around until after etch releases. But then it should |
| 58 |
go away. |
| 59 |
* initscripts Depends on mount >= 2.11x-1 because that's when swapon's |
| 60 |
"-e" option was added. Sarge satisfies this requirement so this |
| 61 |
dependency could be dropped. We can drop it after etch releases; |
| 62 |
then both stable and oldstable will satisfy the requirement. |
| 63 |
* initscripts needs to Depend on debianutils >= 2.12.0 until etch has |
| 64 |
been released so that which can be assumed to be in /bin (rather than |
| 65 |
in /usr/bin, as it was before, including in 2.8.4 in sarge) |
| 66 |
* initscripts Conflicts with and Replaces mdutils which was last seen in |
| 67 |
potato. Can disappear after etch. |
| 68 |
* initscripts Conflicts with sysvinit (<< 2.86.ds1-12) because earlier |
| 69 |
versions of bootlogd didn't have the -c option which the bootlogd |
| 70 |
initscripts now uses |
| 71 |
|
| 72 |
Questionable: |
| 73 |
* initscripts Replaces various libc packages for reasons described here: |
| 74 |
http://lists.debian.org/debian-hurd/2004/06/msg00048.html. |
| 75 |
Anyone know how long this dependency needs to be carried? |
| 76 |
* sysv-rc (current) Conflicts with and Replaces file-rc. file-rc (0.8.7) |
| 77 |
Replaces: sysv-rc but does not Conflict with it. I am not sure that |
| 78 |
sysv-rc needs to Replace: file-rc. Isn't Conflicting sufficient? |
| 79 |
And why doesn't file-rc Conflict with sysv-rc? Strange. |
| 80 |
|
| 81 |
Possible future: |
| 82 |
* If the mtab initscript is ever changed so that it updates the mtab file |
| 83 |
without rerunning mountdevsubfs.sh then initscripts should be made to |
| 84 |
Conflict with udev << 0.080-1. (The latter versions of udev put the |
| 85 |
udev initscript at S04, _after_ S04mountdevsubfs.sh, and the script |
| 86 |
unmounts /dev/pts and /dev/shm.) |
| 87 |
|
| 88 |
|
| 89 |
Speed and compatibility |
| 90 |
----------------------- |
| 91 |
People running Debian on low-memory systems say: |
| 92 |
* Use built-ins |
| 93 |
* Avoid fork-and-exec, pipelines, $() |
| 94 |
* Reduce the number of commands within pipelines |
| 95 |
* Avoid unnecessary code |
| 96 |
* Avoid commands that don't work the same way in BusyBox: |
| 97 |
'nice', 'find', 'mount', 'umount', 'init', 'halt', 'shutdown', |
| 98 |
'syslogd', 'klogd', 'hwclock', 'cron', 'anacron', 'crontab'. |
| 99 |
|
| 100 |
Regenerate /etc/default/rcS? |
| 101 |
---------------------------- |
| 102 |
The current /etc/default/rcS template just refers the admin to rcS(5). |
| 103 |
This way we have only one set of docs to maintain, and we don't have |
| 104 |
to worry about out-of-date comments in /etc/default/rcS. |
| 105 |
|
| 106 |
I wonder if we should strip comments out of existing /etc/default/rcS |
| 107 |
files, leaving behind only the reference to rcS(5). |
| 108 |
|