Having mixed issues with configuring ZMD settings through ZCC.

Managing 5k devices on:
Code:
cat /etc/opt/novell/zenworks/zenworks-release
major-version=7
minor-version=2
micro-version=2
For device123, I overrode ZMD setting via ZCC:

This is sourced in from Global settings
Inherited ZMD Settings
Code:
Name	                Value
device-locked	false
rollback	        true
The effective settings view
Code:
Effective ZMD Settings on this Device
blackout-interval	05:01-23:59
refresh-interval	3600
My overrides
Code:
ZENworks Management Daemon Settings
blackout-interval	05:01-23:59
refresh-interval	3600
Running this on the agent system will refresh settings for some systems, but not all. If running this command does not work then I exec a 'rug ref'. This was the case for this this particular device:
Did not work:
Code:
# rczmd restart ; sleep 20; rug get | egrep "black|refresh"
Did work, settings reflected what ZCC showed for this device.
Code:
# rug ref; rug get | egrep "black|refresh"
After the manual override I wanted to see where things ended up if I reverted back:

This sourced in from Global settings
Code:
Inherited ZMD Settings
Name	                Value
device-locked	false
rollback	        true
The effective settings view
Code:
Effective ZMD Settings on this Device
blackout-interval	05:01-23:59
refresh-interval	3600
No overrides
Code:
ZENworks Management Daemon Settings
Name	Value
No items available.
I had to destroy /etc/zmd/zmd.conf. Where in the world is 'refresh-interval 7200 ' coming from?
Code:
# rczmd restart; rug ref; rug get | egrep "black|refresh"
Shutting down ZENworks Management Daemon                              done
Starting ZENworks Management Daemon                                   done
blackout-interval          |                    | Time interval for which the device will be locked, format is in 24hr clock. For eg: 09:00-18:00
refresh-interval           | 7200               | How long between refreshes, in seconds
refresh-interval-hardware  | 86400              | How long between hardware inventory refreshes, in seconds
refresh-interval-software  | 86400              | How long between software inventory refreshes, in seconds
refresh-interval-system    | 86400              | How long between system inventory refreshes, in seconds
AND why is ZCC not reflecting this?
Code:
Effective ZMD Settings on this Device
blackout-interval	05:01-23:59
refresh-interval	3600
I did just about everything imaginable to bring Agent/ZCC in alignment. Blew away cache (figured what the heck), removed agent DB (rm /var/lib/zmd/zmd.db), restarted server DB, restarted server procs.

Not sure if there are dangling objects or refs in DB. Perhaps things are in a bad state? Maybe a set of aggressive Object store/DB maintenance UTILS exist that need to be executed?

What I've outlined in this post is a representation of a handful of agent quirks I've encountered. Others include: agents taking forever to refresh, agents hanging on refreshes, ZMD RPOC going rogue and requiring a kill-9 (why base enterprise solution on mono, was this a cross platform shortcut decision?).

After 3 weeks of waiting for replies to 3 unanswered emails, a Novell support engineer tells me they just experienced a round of layoffs after a merger (not something you really want to mention in a support email). After 2 more weeks of not receiving a reply, I am told to upgrade the product. Yeah, I'll get right on that and have it done in 24 hours while putting a PROD environment of 5k devices at risk.

Sorry for the sarcastic tones, but something's not right between the integrity of this product and the support coming from this company. Anyone with any insight into this would be greatly appreciated.

Other (related questions) to follow if anyone bytes ;) Rainer?