The message:
apport-autoreport.service loaded failed failed Process error reports when automatic reporting is enabled
is probably unrelated to the package lightdm.
The message can be confirmed by running:
$ sudo systemctl --failed
using "sudo systemctl --failed" to confirm apport-autoreport.service failure message
Recent upgrades/updates on Ubuntu seem to trigger this apport-autoreport.service error. The apport automatic crash report uploading, called whoopsie, becomes disconnected tfrom the local system's /var/crash store of reports, and dies, reporting nothing to Canonical.
Here is a solution:
$ # Do we have the problem?
$ sudo systemctl is-enabled whoopsie.path
disabled
$
$ # Yes, let's enable whoopsie.path
$ sudo systemctl enable whoopsie.path
Created symlink /etc/systemd/system/multi-user.target.wants/whoopsie.path → /lib/systemd/system/whoopsie.path.
$
$ sudo systemctl is-enabled whoopsie.path
enabled
$
$ # Now start the service and check if failure is cleared.
$ sudo systemctl start apport-autoreport
$ sudo systemctl --failed
UNIT LOAD ACTIVE SUB DESCRIPTION
0 loaded units listed.
$ # Problem: solved!!! Reboot and "systemctl --failed" to double-check
Details:
The whoopsie.path is a systemd stanza that tells whoopsie package, which uploads things from /var/crash to Canonical for reporting, that /var/crash is the directory that should be checked (and monitored) for crash reports, which itself is directed via apport-autoreport.path .
Note: The apport mechanism had previously been a security problem.
To see the apport crash reports your system has submitted to errors.ubuntu.com, do the following:
$ # Grab system's unique ID and form URL to look at logs
$ echo https://errors.ubuntu.com/user/$(sudo cat /var/lib/whoopsie/whoopsie-id)
https://errors.ubuntu.com/user/(128_character_hex_id_here)
$ # and open the URL generated