Posted on 10-02-2013 01:12 PM
Hello,
Not sure where to begin with this one. I have a mobile filter package that we are pushing out to all of our student Macs. I've looked at the chart and logs and see a hundred or so "Failed" reports, out of about 1600. That didn't seem too bad, so I started looking at the reports. The first dozen I opened showed the package as successfully downloading and installing. The failure came at some point during the Recon process. Either at the beginning, at the "Retrieving inventory preferences from https://jss-01.ccsd93.com:8443/..." or at the end with some form of "Error running recon: Connection failure: "The host jss-01.ccsd93.com is not accessible."
My first question is: Is it possible to not show Recon failures as "Failures"? Everything else processed successfully.
Second: Any idea why our JSS is sometimes inaccessible? We did update the server to 9.11. We are running our JSS on a Linux server that has been kept updated. Are there updated system recommendations for 9 on Linux that I did not notice?
Thanks, Tim
Posted on 10-17-2013 09:21 AM
Ditto on this one. I've had multiple problems that I've whittled down and am primarily looking at this specific failure as well as a few CRUD errors that we're chasing through different channels.
I am on OS X 10.8.5, JSS 9.12
The policies are working. We can resolve our host names just fine everywhere else, except when recon throws a fit.
Posted on 10-18-2013 08:32 AM
OK... kind of embarrassing. In working with JAMF I've found that I misconfigured the "Grant All" command when creating the new jamfsoftware database during last weeks rebuild. Instead of typing:
grant all on jamfsoftware.* to jamfsoftware@localhost identified by 'password';
I apparently added the '' around the database name which left the password BLANK! The JSS sort of got along with this because, well, not having a password grated quite a lot of access. I'm hoping that this will help resolve the issue.
Posted on 10-25-2013 07:45 AM
Unfortunately I am still seeing these errors form time to time on normal Inventory processes. I'm beginning my research now and will fill in more info later.
** UPDATE ** Looks like possible ARP Cache issues on our core switch. Once we resolve those I'll see if I still have failing connections to the jss
Posted on 04-21-2014 06:28 PM
Did you guys have an update to this?
I get email alerts from Casper and often get Recon errors the same as original post. If I ssh to the machine and run sudo jamf recon it will work. Doing host/nslookup shows that is nows the Casper server.
Not sure where to go from here. Thanks
Posted on 04-23-2014 08:42 AM
I've also begun to see these alerts coming in randomly. They don't seem consistent and tend to be different machines every time.
Posted on 04-23-2014 09:58 AM
Ditto, very random, very inconsistent.
Posted on 04-23-2014 10:19 AM
I've encountered this problem too. We weren't having this problem prior to updating to version 9.3. I have computers that are hard wired to the network and are always turned on have this issue not just my laptop users. I don't see any common thread between those computers experiencing this issue.
Posted on 04-23-2014 11:17 AM
Me too. It's random.
Posted on 04-23-2014 11:30 AM
I saw this at HDS- turned out to be our riverbeds not treating SSL traffic on a port other than 443 correctly. If you've got riverbeds, try setting your JSS url to bypass them.