SOLVED Hardware Scan/Inventory Randomly Disappearing

[m0lochwalker]

Hello,

As title states, I have around 1/10 of my clients which have a full Hardware Inventory one day, along with a Hardware Scan date showing the console - and the next morning (not every morning) both have disappeared. No scan date showing in the console, no Hardware Inventory available.

I can identify this easily because I created a Device Collection which contains all Clients with no Hardware Inventory...

select
RSYS.ResourceID,
RSYS.ResourceType,
RSYS.Name,
RSYS.SMSUniqueIdentifier,
RSYS.ResourceDomainORWorkgroup,
RSYS.Client
from SMS_R_System as rsys
LEFT JOIN SMS_G_SYSTEM_COMPUTER_SYSTEM as gsys ON rsys.ResourceID=gsys.ResourceID
WHERE gsys.Model IS NULL

I can produce any log you would like, but when I check InventoryAgent.log, Dataldr.log, CCMMessaging.log... I do not see any significant errors or anything that other logs are not showing.

My question is: Why would a client that successfully sends a full Hardware Inventory one day, and for days in a row, on a 4 hour schedule in the Client Settings...suddenly lose it's Hardware Inventory and Hardware Scan date? This happens perhaps ~twice a week. When I re-run the Hardware Cycle on these machines, it all comes back like there is no problem at all.

Please advise on where to start and I can provide more information as needed.

I do see this in the MP_Status log...

Mp Status: processing event: CLIMSG_HINV_ERROR_COLLECTIONFAILURE, for machine: x

 

[m0lochwalker]

Ok. Well I think the issue was related to Delete Inactive Client Discovery Data maintenance task. It was running more frequent than Heartbeat Discovery. This resulted in losing client discovery data.

 

[Garth]

it is generally recommend that heartbeat discovery be set to daily same thing with HW inventory too.

 

[m0lochwalker]

Thank you for that info, Garth. What if Site-Wide Client Push is enabled? In relation to running Heart Beat every day...

 

[Garth]

There is no connection between the two. so...

 

[m0lochwalker]

I guess my question is - when you configure Heartbeat Discovery frequency, it says to make sure the Clear Install Flag maintenance task rediscovery period is run more frequently, when Client Push is enabled. If Heartbeat is run every day, how do we set the Client Install Flag rediscovery to more frequent than 1 day?

 

[Garth]

You don't and you shouldn't. It doesn't say that, it says less frequently. The clear flag should be at least 3 times the heartbeat time. But honestly the 21 days which is the default should be fine.

 

[m0lochwalker]

Ok. I will configure as you suggest. When I look at Heartbeat Discovery Properties, it says, 'If automatic site-wide client push installation is enabled, configure the heartbeat schedule to run less frequently than the client rediscovery period for the Clear Install Flag maintenance task.'

If Heartbeat is run every day, it cannot possibly run less frequently than client rediscovery on the Clear Install Flag task? I seem to be missing something obvious here.

 

[Garth]

Ok. I will configure as you suggest. When I look at Heartbeat Discovery Properties, it says, 'If automatic site-wide client push installation is enabled, configure the heartbeat schedule to run less frequently than the client rediscovery period for the Clear Install Flag maintenance task.'

If Heartbeat is run every day, it cannot possibly run less frequently than client rediscovery on the Clear Install Flag task? I seem to be missing something obvious here.

You are not reading they way they indent. if Clear Install Flag is set to 7 days then running heartbeat less frequently would be daily.

I know it is almost a double negative but that is what they mean.

Again I would leave the clear flag at its default of 21 days.

 

[m0lochwalker]

Right, I see. To me, 'frequency' means how often something runs. Every day = most frequent. I digress.

Thanks for the input!