From: Carlos Santos <casantos@datacom.ind.br>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCH v3] radvd: improve startup script
Date: Wed, 25 Apr 2018 21:08:19 -0300 (BRT) [thread overview]
Message-ID: <1735406129.182527.1524701299831.JavaMail.zimbra@datacom.ind.br> (raw)
In-Reply-To: <20180425232512.081aa142@windsurf>
> From: "Thomas Petazzoni" <thomas.petazzoni@bootlin.com>
> To: "Carlos Santos" <casantos@datacom.ind.br>
> Cc: "buildroot" <buildroot@buildroot.org>
> Sent: Wednesday, April 25, 2018 6:25:12 PM
> Subject: Re: [Buildroot] [PATCH v3] radvd: improve startup script
> Hello,
>
> On Sun, 15 Apr 2018 23:10:37 -0300, Carlos Santos wrote:
>
>> -RADVD=/usr/sbin/radvd
>> +test -f /etc/radvd.conf || exit 0
>
> I'm still not impressed by silent exit cases. Shouldn't we let radvd
> fail to start and complain about the lack of radvd.conf ?
Hum, yes, but as already discussed in a different thread there is no
standard regarding these cases.
So far I'm just following the example existing in other start-up scripts:
$ grep 'test -f .* || exit' package/*/S[0-9]*
package/dhcp/S80dhcp-relay:test -f /usr/sbin/dhcrelay || exit 0
package/dhcp/S80dhcp-server:test -f /usr/sbin/dhcpd || exit 0
package/dhcp/S80dhcp-server:test -f /etc/dhcp/dhcpd.conf || exit 0
package/mpd/S95mpd:test -f /etc/mpd.conf || exit 0
>> +start() {
>> + printf "Starting radvd: "
>> + echo "1" > /proc/sys/net/ipv6/conf/all/forwarding
>> + start-stop-daemon -S -x /usr/sbin/radvd || {
>> + echo "FAIL"
>> + exit 1
>> + }
>
> Can we use the
>
> [ $? = 0 ] && echo "OK" || echo "FAIL"
>
> syntax that we use in almost all other init scripts ?
No, because the echo "FAIL" command succeeds, masking the error result:
$ (false; [ $? = 0 ] && echo "OK" || echo "FAIL";); echo $?
FAIL
0
instead of
$ (false || { echo "FAIL"; exit 1; }; echo "OK";); echo $?
FAIL
1
Moreover shellcheck complains that it is bad syntax:
$ shellcheck package/radvd/S50radvd
In package/radvd/S50radvd line 14:
[ $? = 0 ] && echo "OK" || echo "FAIL"
^-- SC2181: Check exit code directly with e.g. 'if mycmd;', not indirectly with $?.
In fact many Buildroot start-up scripts make shellcheck have heart
attacks. :-)
>
>> + echo "OK"
>> +}
>> +
>> +stop() {
>> + printf "Stopping radvd: "
>> + start-stop-daemon -K -q -x /usr/sbin/radvd || {
>> + echo "FAIL"
>> + exit 1
>> + }
>
> Ditto here.
>
> Also, can we use a pid file managed by start-stop-daemon, like
> S50dropbear is doing (and many other init scripts) ?
Yes, but I'd prefer make this change in a different patch if you don't
mind, since it requires additional testing.
--
Carlos Santos (Casantos) - DATACOM, P&D
?The greatest triumph that modern PR can offer is the transcendent
success of having your words and actions judged by your reputation,
rather than the other way about.? ? Christopher Hitchens
prev parent reply other threads:[~2018-04-26 0:08 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-25 9:57 [Buildroot] [PATCH] radvd: improve startup script Carlos Santos
2017-09-26 22:26 ` Arnout Vandecappelle
2017-09-27 1:15 ` Carlos Santos
2017-09-27 2:16 ` [Buildroot] [PATCH v2]] " Carlos Santos
2017-10-07 21:31 ` Thomas Petazzoni
2018-04-16 2:10 ` [Buildroot] [PATCH v3] " Carlos Santos
2018-04-25 21:25 ` Thomas Petazzoni
2018-04-26 0:08 ` Carlos Santos [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1735406129.182527.1524701299831.JavaMail.zimbra@datacom.ind.br \
--to=casantos@datacom.ind.br \
--cc=buildroot@busybox.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox