10 Key Inventory Machine
Armanda Kicks
The Keytracker
The ISS Keytracker is an inventory/stocktaking data collection unit, specifically developed for financial (dollar-value) inventories.
The physical characteristics and software operations of the Keytracker were designed by a team of inventory service professionals. A team with over twenty years experience at four inventory auditing services. The design goals were strict: create a powerful but easy to use inventory machine. There was one more very important requirement; It had to be affordable.
Hello, @naktyro Thank you for your post on NAKIVO Forum! Your information was received and forwarded to our Support Team. We will follow up with you shortly. Thank you for choosing NAKIVO Backup & Replication as your data protection solution.
@naktyro I appreciate your patience during our investigation. Make sure that the job related to the inventory object is removed. Try to retry the removal later (probably the inventory object was busy with the refresh task).
@naktyro Thank you for the information provided. Please generate and send us a new support bundle with the main database for further investigation. Do not forget to specify your ticket number #167256 in the description. Check a guide: +Bundles
In case anyone else is searching around and comes across this, I had a support call from Nakivo this AM. Problem was the Windows agent wasn't starting. As soon as I manually started it everything went fine.
One of my ideas is to have the local reps fill in an excel form within their directory on the network. So a template with headers: Machine #, Project Name, Start Date, End Date, Notes.
A process is then created which collects all of these files and outputs the raw data which I can then feed into Tableau, create a dashboard like the above and then onto the server for the local reps to be able to use. When reps want to edit or delete an entry they simply delete the file off their network directory. I could then run a process once a day or whenever needed. I'd also like a way for our reps to be able to click on a booking and be able to view what dates the collisions occur so they can easily relay this back to the customers.
is there any reason why the reps can't all access the same template? It seems like this would short circuit any possibility of double booking machines if the rep can already see that a machine has been reserved for the required time.
We have a vast number of machines and bookings can be very long and complex, such as a few days here a few days there etc. The same booking can also cover multiple machines. Combined with the number of bookings a day and amount of reps, this becomes unmanageable.
I can imagine that avoidable clashes will arise where reps have 'chosen' the same machine in their template but there are other available machines. I guess that this dashboard is meant to be the way in which reps can see what's available but there will probably be a timing difference between when the rep makes a booking in their template and when the dashboard gets updated.
- are you suggesting a new file for each booking or one booking file per rep with all bookings within it. Deleting the file means that you won't be able to look backwards, perhaps another field for active/inactive might give you more flexibility?
Can be solved in a couple different ways, but it depends on how the input file would work. Can you mock the input in an excel workbook and attach? Also if you have a screen shot or two of how the results would look in Tableau to make sure the data file is shaped correctly.
I like the use of the summarize tool and and reservation count field to identify collisions. I especially like the solution to the third data format. The use of the generate rows tool to produce the end result is ingenious.
2) Is the calendar view output structured in the best way to be able to be fed into a viz software like Tableau? I'd like the view to be similar to the one in my original post, with a list of all of our machines and projects on the side and on top the dates spanning into the future on a scroll (this should be okay because the scroll only need to go so far as the max end date). Open to suggestions on this!
- Reps need to assign specific machines to projects. It's a bit more complex than I described but to keep the example simple let's assume the reps do assign manually. Correct the dashboard is meant to be the way to see whats available. That is a good point, I need to factor in a timing difference. Going back to your first point about a single template I need to think about what works best here. Same thing about your last point regarding multiple files or a single file per rep.
What do you think about creating some type of booking form, similar to surveys, and having the results sent to me. Then I can timestamp entries and allocate accordingly.But this takes any editing control out of the hands of the reps. Hmm need to think about this.
Then you will get this output if it was there was a conflict. I added a row to the original data set that showed project stream reserved for the same day on a different machine to mock up this collision.
We've got a Daily_Detect_All schedule which runs every day at 2am and performs a Detect on all machines. We have about 350 machines and this detect schedule seems to run throughout a large portion of the day despite kicking off in the early morning. One particular user keeps complaining that their applications function extremely slowly while the Inventory process is running (they open up Task Manager and see it) on their machine. None of our other users have mentioned this. Is this something that anyone else has noticed or heard of? I guess I could break it up and have it only hit smaller groups of machines and spread it out over 3 or 4 days so that it hits machines before anyone comes in to work, but are there any issues with only performing an inventory every few days?
Not good:
"We've got a Daily_Detect_All schedule which runs every day at 2am and performs a Detect on all machines.".
Use Labels based on Software and O.S. version installed in your device.( please do not detect all patches, you making the patching process waste time by checking patches for Apps you don't even have, use Smart Labels).
Also there's no need to detect patches all every single day, even if it's outside business hours, plus Are those Desktop Devices? Because If you have Laptops, How do you know your detection did not happen at 2AM, and it was not deferred to the next Online State. (that means when your users login in the morning).
Anyways, see if an AV is blocking the patching process.
Review the KAGENT log, it will tell what is doing when the device is Slow (ask your users when that happens and note the time), if it's patching, AV might be blocking your patching EXEs.
Or you are detecting thousand of patches and the process is still running at 8AM... plus AV auditing the process....
See:
-systems-management-appliance/kb/111785/which-directories-and-files-do-i-need-to-whitelist-for-the-sma-agent-
I have read here that this is not the recommended thing, but still, how do we go about deleting machines completely from FNMS completely from Compliance and Inventory databases. I have seen some scripts here, but unable to find those posts.
If a computer is removed from the Inventory database (specifically, the Computer_MT table) AND it exists in no other 3rd-party source, then the next full reconciliation will remove the associated machine from the Compliance database.
For deleting a computer from the Compliance database, I suppose it is ok to delete it from All Inventory, and that will be all? Or would some other steps be needed? We are all assuming that the computer is not from a third party source.
The overnight reconciliation will remove the device from the ImportedComputer table in the Compliance database, and subsequently the ComplianceComputer table (which will then make it disappear from the UI).
If you want to delete computers based on how long it is since they have returned inventory, check out the following post about a new capability for doing this: Auto-cleanup of Obsolete FlexNet Inventory.
I just noticed that my certificate inventory for the login keychain seems to be out of date. I'm curious if something changed (potentially with Catalina) where Jamf can no longer inventory certificates in the login keychain.
From what I observed, when things are working as expected, the user level certificate gets deployed to the machine when a user logs in. When taking the machine out of the configuration profile scope and then when user logs in, the user level certificate gets removed from the machine. These are all reflected in the Inventory > Certificates section. When the certificate is added, it has that username associated with it. So for a machine with multiple user profiles, I will see multiple entries in the certificates list with the different usernames. Obviously this is the managed user certificate deployment scenario.
However, here come the problems. if the user manually removes the managed user certificate from login keychain:
1. Jamf inventory does not get updated and will still show the certificate as deployed. I can only tell the certificate is missing on the client machine either in keychain or the User Profile in the System Preferences > Profiles will show (Certificate) Error: Not found in keychain
2. The only way that I know to force the Jamf inventory to update the Certificates info reading is by running sudo jamf mdm -userLevelMdm command. But this will require the user to manually approve the MDM Profile again.
3. In order to put the user certificate back into user's login keychain, I had to add the machine into the Exclusions of the Configuration Profile scope, log off and log back in the user (the deployment completed record will disappear from the Configuration Profile log). Then remove the machine from the Exclusions, log off and log back in the user. Finally the certificate is back there. It is pity that the Configuration Profile doesn't detect the change and reverse it by automatically applying again at the user's next login. I think even the system level configuration profile doesn't do that either.