[gentoo-dev] rc-scripts/init: new service dependency

[gentoo-dev] rc-scripts/init: new service dependency

[Wolfram Schlich]

Hi,

I just switched my internet gateway from a heavily handpatched SuSE 7.3 
to Gentoo 1.4rc4. So far it *works* :o)
My setup is:

	/etc/init.d/net.eth1 (PPPoE-device):
		-> provide net.eth-dsl (virtual provide)

	/etc/init.d/net.ppp0 (DSL-iface):
		-> need net.eth-dsl (virtual provide from net.eth1)
		-> provide net.internet (virtual provide)

	/etc/init.d/ddclient (DynDNS.org updater):
		-> need net.internet (virtual provide from net.ppp0)

This works *so far*:
--8<--
[root@chiron(pts/2):tmp]$ rcddclient start
 * Bringing eth1 up... [ ok ]
 * Starting pppd (ppp0)... [ ok ]
 * Starting DDClient... [ ok ]
[root@chiron(pts/2):tmp]$ rcnet.eth1 stop
 * Stopping DDClient... [ ok ]
 * Stopping pppd (ppp0)... [ ok ]
 * Bringing eth1 down... [ ok ]
[root@chiron(pts/2):tmp]$
--8<--

What I tried to do now is the following: when net.ppp0 is started,
ddclient should be started automatically as well. I've found no
dependency-way to do this (except making net.ppp0 need ddclient, which
turns out to be kinda logical loop, despite the fact it seems to work in
practice... still feels bad).

What about a new dependency like "triggeredby"?
Another name could be "stickto"...
This would/should/could both affect "start" and "stop" operations. E.g.:

	/etc/init.d/ddclient:
		-> triggeredby net.ppp0
		-> after net.ppp0

This would cause ddclient to be started when net.ppp0 is started, but
due to the "after" dep, it would be started *after* net.ppp0.
Kinda reverse/twisted need dependency... or something.

This "triggeredby" could be applied to e.g. squid, iptables, postfix... all the
stuff you *might* choose to need when connecting to the internet but
don't want to start before you *really* connect.

Comments please.
-- 
Wolfram Schlich; Friedhofstr. 8, D-88069 Tettnang; +49-(0)178-SCHLICH

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Martin Schlemmer]

[-- Attachment #1: Type: text/plain, Size: 913 bytes --]

On Mon, 2003-05-05 at 22:23, Wolfram Schlich wrote:

> What about a new dependency like "triggeredby"?
> Another name could be "stickto"...
> This would/should/could both affect "start" and "stop" operations. E.g.:
> 
> 	/etc/init.d/ddclient:
> 		-> triggeredby net.ppp0
> 		-> after net.ppp0
> 
> This would cause ddclient to be started when net.ppp0 is started, but
> due to the "after" dep, it would be started *after* net.ppp0.
> Kinda reverse/twisted need dependency... or something.
> 
> This "triggeredby" could be applied to e.g. squid, iptables, postfix... all the
> stuff you *might* choose to need when connecting to the internet but
> don't want to start before you *really* connect.
> 

Might be a good feature to support.  I just do not like the name =)


Regards,

-- 

Martin Schlemmer
Gentoo Linux Developer, Desktop/System Team Developer
Cape Town, South Africa



[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Wolfram Schlich]

[-- Attachment #1: Type: text/plain, Size: 1035 bytes --]

* Martin Schlemmer <azarah@gentoo.org> [2003-05-05 22:55]:
> On Mon, 2003-05-05 at 22:23, Wolfram Schlich wrote:
> 
> > What about a new dependency like "triggeredby"?
> > Another name could be "stickto"...
> > This would/should/could both affect "start" and "stop" operations. E.g.:
> > 
> > 	/etc/init.d/ddclient:
> > 		-> triggeredby net.ppp0
> > 		-> after net.ppp0
> > 
> > This would cause ddclient to be started when net.ppp0 is started, but
> > due to the "after" dep, it would be started *after* net.ppp0.
> > Kinda reverse/twisted need dependency... or something.
> > 
> > This "triggeredby" could be applied to e.g. squid, iptables, postfix... all the
> > stuff you *might* choose to need when connecting to the internet but
> > don't want to start before you *really* connect.
> > 
> 
> Might be a good feature to support.  I just do not like the name =)

Me neither :-P I spent ~5 minutes thinking about a dep. name... :-/
-- 
Wolfram Schlich; Friedhofstr. 8, D-88069 Tettnang; +49-(0)178-SCHLICH

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Wout Mertens]

On Mon, 5 May 2003, Martin Schlemmer wrote:

> On Mon, 2003-05-05 at 22:23, Wolfram Schlich wrote:
>
> > What about a new dependency like "triggeredby"?
> > Another name could be "stickto"...
> > This would/should/could both affect "start" and "stop" operations. E.g.:
> >
> > 	/etc/init.d/ddclient:
> > 		-> triggeredby net.ppp0
> > 		-> after net.ppp0
> >
> > This would cause ddclient to be started when net.ppp0 is started, but
> > due to the "after" dep, it would be started *after* net.ppp0.
> > Kinda reverse/twisted need dependency... or something.
> >
> > This "triggeredby" could be applied to e.g. squid, iptables, postfix... all the
> > stuff you *might* choose to need when connecting to the internet but
> > don't want to start before you *really* connect.
> >
>
> Might be a good feature to support.  I just do not like the name =)

How about subscribe? And you can ask to be called before and/or after the
script you subscribe to:

subscribe after net.ppp0

event() {
	case $2 in
		start|restart)
			start
			;;
		stop)
			stop
			;;
	esac
}

This would make runscript call ddclients event() after net.ppp0 was run,
with $1 set to the script, $2 set to the action and $3 set to before or
after.

Of course, this still doesn't fix having multiple ways of bringing up your
interface, like I proposed for gentoo-net (need to work on that :-/). My
proposal also took care of calling ddclient after the interface was up...

Wout.

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Martin Schlemmer]

[-- Attachment #1: Type: text/plain, Size: 819 bytes --]

On Tue, 2003-05-06 at 11:40, Wout Mertens wrote:

> How about subscribe? And you can ask to be called before and/or after the
> script you subscribe to:
> 
> subscribe after net.ppp0
> 
> event() {
> 	case $2 in
> 		start|restart)
> 			start
> 			;;
> 		stop)
> 			stop
> 			;;
> 	esac
> }
> 
> This would make runscript call ddclients event() after net.ppp0 was run,
> with $1 set to the script, $2 set to the action and $3 set to before or
> after.
> 

Well, if you need to start something before, you can just use 'use', or
'before/after'.

I think what he wants, and what will be a unique addition, is being able
to tell a script:  If you start successfully, start 'foobar'


-- 

Martin Schlemmer
Gentoo Linux Developer, Desktop/System Team Developer
Cape Town, South Africa



[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Wout Mertens]

On Tue, 6 May 2003, Martin Schlemmer wrote:

> On Tue, 2003-05-06 at 11:40, Wout Mertens wrote:
>
> > How about subscribe? And you can ask to be called before and/or after the
> > script you subscribe to:
> >
> > subscribe after net.ppp0
> >
> > event() {
> > 	case $2 in
> > 		start|restart)
> > 			start
> > 			;;
> > 		stop)
> > 			stop
> > 			;;
> > 	esac
> > }
> >
> > This would make runscript call ddclients event() after net.ppp0 was run,
> > with $1 set to the script, $2 set to the action and $3 set to before or
> > after.
> >
>
> Well, if you need to start something before, you can just use 'use', or
> 'before/after'.
>
> I think what he wants, and what will be a unique addition, is being able
> to tell a script:  If you start successfully, start 'foobar'

That's what I meant: when the service is called for an action (start,
stop, restart, reload, custom things), runscript will run the event()
function for all the scripts that are subscribed to that script, and then
those scripts can figure out what to do with it themselves...

And the gentoo-net scripts that I am working on will go down a
user-configurable list when certain things happen. (interface up/down, ip
address changed, etc.

The design pattern we're looking at here is events and interested parties,
and it's just a matter of defining what events we handle and how we notify
the parties.

Wout.

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Martin Schlemmer]

[-- Attachment #1: Type: text/plain, Size: 2088 bytes --]

On Tue, 2003-05-06 at 21:59, Wout Mertens wrote:

> That's what I meant: when the service is called for an action (start,
> stop, restart, reload, custom things), runscript will run the event()
> function for all the scripts that are subscribed to that script, and then
> those scripts can figure out what to do with it themselves...
> 
> And the gentoo-net scripts that I am working on will go down a
> user-configurable list when certain things happen. (interface up/down, ip
> address changed, etc.
> 
> The design pattern we're looking at here is events and interested parties,
> and it's just a matter of defining what events we handle and how we notify
> the parties.
> 

So in the network context:

-------- add to /etc/conf.d/net ---------
modules_eth0="static route firewall"
-----------------------------------------

-------- add to /etc/init.d/net.eth? ----
depend() {
    subscribe after ${modules_eth0}
}
-----------------------------------------

and if net.eth0 starts (meaning the interface is present for example),
it will run the 'static' module to set the static ip, then do the
routes, and lastly add firewall/whatever rules.  Same in reverse if
its stopped.

Note that above might be too simplified, and out of context.

This brings up a question or two:

1)  should this work for all rc-scripts, or should we maybe
    add a /etc/init.d/modules/ that have very job specific
    scripts that are related to net.* (net.eth?, net.ppp?, net.foo)

2)  if yes to 1), will we need it for anything else than net.* stuff ?

3)  you mentioned a few events ... do we really need them all, or could
    we get away with a 'start/stop' only ?

To get it running, is not the issue.  The bigger issue is getting it
functional, without too much bloat/speed decrease/complexity, with
a stable API that do not have to change every few months.  Basically
we will need to think what exactly is needed, and will be needed.


-- 

Martin Schlemmer
Gentoo Linux Developer, Desktop/System Team Developer
Cape Town, South Africa



[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Wout Mertens]

On Tue, 6 May 2003, Martin Schlemmer wrote:

> On Tue, 2003-05-06 at 21:59, Wout Mertens wrote:
>
> > That's what I meant: when the service is called for an action (start,
> > stop, restart, reload, custom things), runscript will run the event()
> > function for all the scripts that are subscribed to that script, and then
> > those scripts can figure out what to do with it themselves...
> >
> > And the gentoo-net scripts that I am working on will go down a
> > user-configurable list when certain things happen. (interface up/down, ip
> > address changed, etc.
> >
> > The design pattern we're looking at here is events and interested parties,
> > and it's just a matter of defining what events we handle and how we notify
> > the parties.
> >
>
> So in the network context:
>
> -------- add to /etc/conf.d/net ---------
> modules_eth0="static route firewall"
> -----------------------------------------
>
> -------- add to /etc/init.d/net.eth? ----
> depend() {
>     subscribe after ${modules_eth0}
> }
> -----------------------------------------
>
> and if net.eth0 starts (meaning the interface is present for example),
> it will run the 'static' module to set the static ip, then do the
> routes, and lastly add firewall/whatever rules.  Same in reverse if
> its stopped.

Actually, his is what I want to implement in gentoo-net, but it isn't what
I described :)

I described the reverse:

In your example, interpreted by what I meant, if you start
/etc/init.d/static, it will, after running the start function in
/etc/init.d/static, run the event function in /etc/init.d/net.eth?, with
arguments "static" "start" "after".

So the net.eth? script gets notified about the static script being
started, without the static script having to do anything.

And it calls the event() function because you don't necessarily want to
start the service when another one started, so that way you can choose
what you do.

> Note that above might be too simplified, and out of context.
>
> This brings up a question or two:
>
> 1)  should this work for all rc-scripts, or should we maybe
>     add a /etc/init.d/modules/ that have very job specific
>     scripts that are related to net.* (net.eth?, net.ppp?, net.foo)
> 2)  if yes to 1), will we need it for anything else than net.* stuff ?

Very generally speaking, we want to be able to run scripts when some
device or service changes state. In gentoo-net, that is the network
interface device and the IP "device" (e.g. when the IP address changes you
want to update your dynamic dns entry).

But what about hotplugged devices? Cardbus, usb, pci? These, too, are
interesting to run scripts against, for instance to automatically mount a
usb storage device.

Right now, Mandrake has something in place that handles some hotplugging
events, but networks aren't handled. In 2.6, the kernel will have a more
generic devices tree (sysfs), which might or might not consider bringing a
network device up a hotplug event, easy to handle in a generic way.

So what I'm saying is that it could be useful to set up a generic
framework that handles all these things, and reroute the packages that do
their own thing to use that interface: hotplug, pcmcia_cs, dhcpcd, ppp and
so on all have hooks that can be executed when something happens, and they
could thus be written to send an event instead of doing something.

So what I would like to see is that you can subscribe to generic
user-definable events, and that triggering an event will cause all the
subscribed initscripts to be notified.

> 3)  you mentioned a few events ... do we really need them all, or could
>     we get away with a 'start/stop' only ?

I think that the overhead required to implement start/stop is roughly the
same as doing it for a number of events, so I wouldn't want us to cut
ourselves short...

> To get it running, is not the issue.  The bigger issue is getting it
> functional, without too much bloat/speed decrease/complexity, with
> a stable API that do not have to change every few months.  Basically
> we will need to think what exactly is needed, and will be needed.

Here's two possibilities for doing things after a network interface came
up:

1) semi-hardcoded:
- put a bunch of SysV init style symlinks to initscripts in
  /etc/conf.d/net.d/ifup.d/
- the interface comes up
- start/stop each of them in order, with $INTERFACE set to the interface,

Pro: Easy to write, the user has full control over what gets started or
stopped

Con: hard to maintain the SysV init style directory, each event needs a
directory, solves only one problem.

2) event based:
- the initscripts put subscribe statements in their depend() code
- the interface comes up
- the event dispatcher runs the subscribed initscripts event() functions,
  with details about the event (who, what and args).

Pro: Easy to write the initscripts, solves many problems. The event
dispatcher shouldn't be difficult.

Con: Need to find a nice way for a user to disable an initscript, so it
doesn't get events. Example: the net may be up, but the user doesn't want
wwwoffle to be brought online.

Thoughts?

Wout.

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Daniel]

On Wed, 7 May 2003 09:26 am, Wout Mertens wrote:
>
> 1) semi-hardcoded:
> - put a bunch of SysV init style symlinks to initscripts in
>   /etc/conf.d/net.d/ifup.d/
> - the interface comes up
> - start/stop each of them in order, with $INTERFACE set to the interface,
>
> Pro: Easy to write, the user has full control over what gets started or
> stopped
>
> Con: hard to maintain the SysV init style directory, each event needs a
> directory, solves only one problem.
>
> 2) event based:
> - the initscripts put subscribe statements in their depend() code
> - the interface comes up
> - the event dispatcher runs the subscribed initscripts event() functions,
>   with details about the event (who, what and args).
>
> Pro: Easy to write the initscripts, solves many problems. The event
> dispatcher shouldn't be difficult.
>
> Con: Need to find a nice way for a user to disable an initscript, so it
> doesn't get events. Example: the net may be up, but the user doesn't want
> wwwoffle to be brought online.
>
>
> Wout.
>

Expanding (combining?, bastardising (hopefully not)) your ideas:

SysV init style scripts in /etc/netinit.d

Have /etc/netlevel directory (akin to /etc/runlevels) with the subdirectories 
representing the interface e.g /etc/netlevel/ppp0 /etc/netlevel/eth0 and 
possibly a sepecial /etc/netlevel/all (called for any interface going up or 
down).

Could have a user interface in the form of:
"rc-netupdate {add|del} firewall ppp0"

Aguement to these scripts could be "start"."stop" as usual. Additionaly 
"reconfig" for example if the network interface parmeter change (IP change) 
with the details of the change in environment variables (IP_OLD,IP_NEW, etc). 
I think it would be hard to make all permutations of network change in a 
generic command line format. Are there any other generic  events that need to 
occur?

These would be called upon the hotplug events or /etc/ppp/ip-{up-down} since 
as you mentioned a generic interface is getting hard. I mentioned 
ip-{up|down} because of the case of dial-on-demand and 
/etc/init.d/net.{interface} does not necessarly bring up the interface. If 
its a simple interface the could be called from the /etc/init.d/net.* script 
through some function.

This solves your con of having all events is a different direcory. Scripts are 
now subscription based and triggered by "events".

Taking into account the hotplug usb device mount idea - a few find and 
replaces "s/net/event/g" and replace net interface directories with 
"/etc/eventlevel.d/net.{interface}" and the predominate mechanism is the 
same. I'm not too familiar with hotplug so appoligies if this is unfesable.

Ok "evenlevel" sounds sucky but how does the framework sound.


Daniel
-- 
Key fingerprint = 2300 3292 3351 9665 D41B DC20 4C34 9BB6 32A6 4DC8


--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] online/offline rc-scripts - was: rc-scripts/init: new service dependency

[Daniel]

[-- Attachment #1: signed data --]
[-- Type: text/plain, Size: 2122 bytes --]


Another thought - http://www.gentoo.org/doc/en/rc-scripts.xml refers to 
"online" and "offline" runlevels as suggestion for offline machines. Can we 
make this part of the standard? I imagine a lot of users (and developers too) 
do run offline machines and it makes sence to have some guidance to this. A 
machine attempting to make connections to the internet when there isn't one 
is frustrating. Rather that everyone developing their own solutions I see 
standardisation as a mechanism around this. I noticed Redhat/Mandrake (please 
don't flame me for mentioning them here) have the directories 
/etc/sysconfig/network-scripts/ifup.d /etc/sysconfig/network-scripts/ifdown.d 
for such a purpose. I have faith that Gentoo can also meet this requirement 
in its own way.

Scripts that depend on online/offline status can have requirements to start up 
at a "default" runlevel and wwwoffled, pdnsd, sendmail are example of this.
To make these scripts and others work properly in an offline/online manner 
they need to be told about the connection status. As these scripts are often 
in the /etc/init.d directory (having started at the default runlevel) may I 
humble propose an "online|offline" arguement to these scripts? This would 
consolidate the functionality of the script in one place rather that having 
duplicates (e.g. wwwoffled and wwoffled-offle). Additional functionality such 
as IP changes could also be incorportated here too. I propose these be 
started with something like "rc-online|rc-offline" in the same way that 
/sbin/rc would start other scripts (exept with a different arguement).

I'm willing to submit changes to all the packages that I work with (and others 
for the fun of it) that depend on offline/status indicators but what I realy 
require is some guidnce on how online/offline wants to be handled by the 
Gentoo community in general.

Daniel Black
--
Proudly a Gentoo User.
GnuPG/PGP signed and encrypted email preferred
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x32A64DC8
Key fingerprint = 2300 3292 3351 9665 D41B DC20 4C34 9BB6 32A6 4DC8

[-- Attachment #2: signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Frank Zschockelt]

[-- Attachment #1: Type: text/plain, Size: 276 bytes --]

Es schrieb Wolfram Schlich (lists@schlich.org):
> 
> What I tried to do now is the following: when net.ppp0 is started,
> ddclient should be started automatically as well.

That's no problem. Just add the line for ddclient to /etc/ppp/ip-up.            

Bye,
Franky

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Wolfram Schlich]

Any news out here regarding the issue? :-)
-- 
Wolfram Schlich; Friedhofstr. 8, D-88069 Tettnang; +49-(0)178-SCHLICH

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] rc-scripts/init: new service dependency

[Daniel]

On Tue, 6 May 2003 02:36 pm, Frank Zschockelt wrote:
> That's no problem. Just add the line for ddclient to /etc/ppp/ip-up.

and change AUTOCFGFILES = "no" in /etc/conf.d/net.ppp0 otherwise
/etc/init.d/net.ppp0 will override it.

Maybe there could be an init.d structure in /etc/ppp/ip-up.d/ and get ip-up
 to call them? There as Wolfram mentioned are a few items that could go here.
 /etc/rc.firewall (for guarddog) iptables call maybe, fetchmail, sendmail,
 dyndns, "/usr/bin/wwwoffle -online and /usr/bin/wwwoffle -fetch" if using
 wwwoffled, /usr/sbin/gentoo-stats --update, /usr/sbin/pdnsd-ctl server 0 up
 (if using pdnsd - ref bug #20450), some icq online activiation, etc...

Those are most of the ones I've got in ip-up. The order realy has to be user
defined, or maybe some indication in priority? Ok I haven't thought that much
in detail.

Maybe this is too ppp specific. Does ISDN use /etc/ppp/ip-up? A per interface
is a more generic solution? too generic maybe. Ok I'm babling now - I'll stop
;-)

The "triggered by" solution may have applicablility elsewhere but say for
instance you use dial-on-demand in /etc/init.d/net.ppp0?


Daniel

GnuPG/PGP signed and encrypted email preferred
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x32A64DC8
Key fingerprint = 2300 3292 3351 9665 D41B DC20 4C34 9BB6 32A6 4DC8


--
gentoo-dev@gentoo.org mailing list