zephyr-clients init script shouldn't bail if zhm can't be restarted

[andersk-auto]

If zhm is configured to use Hesiod, but Hesiod is not configured correctly, zhm will fail to restart. Which is fine, but the postinst shouldn't fail here, particularly because this situations results from "apt-get install zephyr-clients", which pulls in an (unconfigured) Hesiod.

Adding the usual || : idiom fixes this.

{{
--- zephyr-3.0.1.orig/debian/zephyr-clients.postinst 2011-03-06 13:41:10.000000000 -0500
+++ zephyr-3.0.1/debian/zephyr-clients.postinst 2013-03-08 13:21:58.365700512 -0500
@@ -70,9 +70,9 @@
if [ -x "/etc/init.d/zhm" ]; then
# this script does not necessarily fail if these do
if [ -x "which invoke-rc.d 2>/dev/null" ]; then

•            invoke-rc.d zhm restart
  

•            invoke-rc.d zhm restart || :
   else
  

•            /etc/init.d/zhm restart
  

•            /etc/init.d/zhm restart || :
   fi
  

  fi
  }}

[andersk-auto]

Imported from trac issue 108. Created by jdreed@ATHENA.MIT.EDU on 2013-03-08T13:23:02, last modified: 2013-03-11T18:37:07

[andersk-auto]

Trac comment by jhutz@CS.CMU.EDU on 2013-03-11 18:37:07:

If Hesiod is unconfigured, zephyr should perhaps not be configured to use it?