1 |
You need to run etc-update (or one of it's cousins) if you are told files need |
2 |
updating. After that run the /etc/init.d/sshd restart. |
3 |
|
4 |
On Friday April 28 2006 20:55, Kevin wrote: |
5 |
> On a related note, what is the most correct procedure for restarting a |
6 |
> service after an update to a service (say named or cyrus-imapd or apache |
7 |
> or sshd)? |
8 |
> |
9 |
> I've been doing something like this: |
10 |
> |
11 |
> # emerge -v openssh |
12 |
> examine config file differences and make any adjustments that are |
13 |
> required to be done by hand. |
14 |
> # /etc/init.d/sshd stop |
15 |
> # mv -i /etc/init.d/._cfg0000_sshd /etc/init.d/sshd |
16 |
> # /etc/init.d/sshd start |
17 |
> |
18 |
> But I've noticed that there are cases when this general procedure |
19 |
> doesn't work. Perhaps because binaries change location between package |
20 |
> versions and the old /etc/init.d/service script presumes the old package |
21 |
> binary location and the new package has already been installed so the |
22 |
> new binaries are no longer with the old startup script thinks they are. |
23 |
> In those cases, I usually find the service with ps and kill it by hand, |
24 |
> then zap the service and start it again, using the new start script, but |
25 |
> maybe there's a better way. Seems to me that it might be better to do |
26 |
> something like this: |
27 |
|
28 |
-- |
29 |
|
30 |
Brett I. Holcomb |
31 |
-- |
32 |
gentoo-user@g.o mailing list |