Running into an issue that honestly has me ready to pull out some of what little hair remains. ;)
Have an upgrade policy for a specific app which fails on MOST, but not all of our systems. I can't determine why.
Have a very simple package which simply places a large installer script into /private/var/tmp
Next, I'm running the install command via "Execute Command" under "Files and Processes".
Essentially, the package places a script in the location, which is then executed. The script itself is too large to deploy via Casper's normal script process (approx 7-8mb), thus this particular approach.
Now - in all of my initial testing, we had success. When deployed to production, on approximately 2/3 of the systems, I get the following in the logs when it attempts to execute the script:
Result of command:
/bin/sh: ./private/var/tmp/install.sh: No such file or directory
On the remainder, it executes perfectly.
I've tried both .dmg and .pkgs for the package, no change.
My main question - is "/private/var/tmp" a bad choice for temporary installers? Is there a "best practice" directory which would make more sense while I attempt to troubleshoot the issue?