Two issues with upgrade from 20.04 to 20.10

I have been using 20.04 on a RPi4/4GB since its release and am very pleased with the performance.
Today, I upgraded to 20.10 and all works well, except for a couple of issues:

  1. After every boot, a "System program problem detected" window opens. It has no "details" tab, so there is no hint of what is the problem. I have checked the /var/log/syslog file, but there doesn't appear to be anything obvious. How can I found out what is the problem?

  2. Chromium takes about 90 seconds to open to my home page (for 20.04 was 2-3 seconds). I have tried removing and reinstalling, but that does not resolve the issue. Once my home page opens, chromium responds normally.
    The following entries from /var/log/syslog relate to the problem. As you can see from the timestamps, the whole process takes 1 minute 28 seconds and includes many "DENIED" messages.

Oct 24 12:09:19 jon-desktop systemd[1]: tmp-snap.rootfs_ZhBbN6.mount: Succeeded.
Oct 24 12:09:19 jon-desktop systemd[1469]: tmp-snap.rootfs_ZhBbN6.mount: Succeeded.
Oct 24 12:09:19 jon-desktop kernel: [ 1098.245474] audit: type=1400 audit(1603530559.267:57): apparmor="DENIED" operation="capable" profile="/usr/lib/snapd/snap-confine" pid=3054 comm="snap-confine" capability=4 capname="fsetid"
Oct 24 12:09:51 jon-desktop dbus-daemon[1492]: apparmor="DENIED" operation="dbus_method_call" bus="session" path="/org/freedesktop/DBus" interface="org.freedesktop.DBus" member="ListActivatableNames" mask="send" name="org.freedesktop.DBus" pid=3054 label="snap.chromium.chromium" peer_label="unconfined"
Oct 24 12:10:03 jon-desktop dbus-daemon[1492]: [session uid=1000 pid=1492] Activating service name='io.snapcraft.Settings' requested by ':1.86' (uid=1000 pid=3316 comm="dbus-send --print-reply=literal --session --dest=i" label="snap.chromium.chromium (enforce)")
Oct 24 12:10:03 jon-desktop dbus-daemon[1492]: [session uid=1000 pid=1492] Successfully activated service 'io.snapcraft.Settings'
Oct 24 12:10:03 jon-desktop io.snapcraft.Settings[3319]: userd.go:98: Starting snap userd
Oct 24 12:10:44 jon-desktop kernel: [ 1183.566954] [UFW BLOCK] IN=eth0 OUT= MAC=01:00:5e:00:00:01:10:5a:f7:0e:dc:8d:08:00 SRC=10.100.102.1 DST=224.0.0.1 LEN=32 TOS=0x00 PREC=0xC0 TTL=1 ID=54730 PROTO=2
Oct 24 12:10:47 jon-desktop kernel: [ 1186.436509] audit: type=1400 audit(1603530647.458:58): apparmor="DENIED" operation="open" profile="snap.chromium.chromium" name="/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.4/1-1.4.2/busnum" pid=3054 comm="ThreadPoolForeg" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0
Oct 24 12:10:47 jon-desktop kernel: [ 1186.436562] audit: type=1400 audit(1603530647.458:59): apparmor="DENIED" operation="open" profile="snap.chromium.chromium" name="/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.4/1-1.4.2/devnum" pid=3054 comm="ThreadPoolForeg" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0
Oct 24 12:10:47 jon-desktop kernel: [ 1186.440317] audit: type=1400 audit(1603530647.462:60): apparmor="DENIED" operation="open" profile="snap.chromium.chromium" name="/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb2/2-1/busnum" pid=3054 comm="ThreadPoolForeg" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0
Oct 24 12:10:47 jon-desktop kernel: [ 1186.440440] audit: type=1400 audit(1603530647.462:61): apparmor="DENIED" operation="open" profile="snap.chromium.chromium" name="/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb2/2-1/devnum" pid=3054 comm="ThreadPoolForeg" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0

Any suggestions would be greatly appreciated.

1 Like

Fresh installs are the best everytime.

1 Like

Fresh installs are the best everytime.

You are no doubt correct. But I use this system as my main work computer, and a fresh install carries a significant setup overhead. It's true that the majority of the setup information can be used directly from /home/XXX/.config, but I would still have to install programs and carry out various setups which are in the root.

1 Like

I too experienced several problems after upgrade and was forced to do a clean install. The clean install works great without any of the problems with an upgrade.

I also have this problem but I did not investigate because I was running 20.10 from alpha stage and have done lot of fiddling. Planning to do clean install to see if it is issue with my setup or with 20.10 release.

I guess that is (new) normal :grimacing: Chromium is a snap package now instead of .deb and as of today snaps do take a while to launch compared to traditional packages (although in my case it is about 20s not 90s).

I guess that is (new) normal :grimacing: Chromium is a snap package now instead of .deb and as of today snaps do take a while to launch compared to traditional packages (although in my case it is about 20s not 90s).

No, that's not the problem. It was also a snap package for 20.04, but then chromium startup took 2-3 seconds.
By the way, this long delay only happens the first time after startup. If I close chromium and then reopen, it reopens immediately.

You are right. I checked chromium, it is slow only first time but starts in couple of seconds thereafter. I also do not have same apparmor DENIED msgs.

In 20.10 AppArmor has been upgraded to version 3.

Please create Launchpad bugs in regards to any issues you may experience.

Please check to see if you have crash files in /var/crash

1 Like

Please check to see if you have crash files in /var/crash

I just rebooted to create a new file in /var/crash

The first few lines are:

ProblemType: Crash
Architecture: arm64
CrashCounter: 1
CurrentDesktop: MATE
Date: Sun Oct 25 19:19:15 2020
DistroRelease: Ubuntu 20.10
ExecutablePath: /usr/libexec/ayatana-indicator-datetime/ayatana-indicator-datetime-service
ExecutableTimestamp: 1597738096
ProcCmdline: /usr/libexec/ayatana-indicator-datetime/ayatana-indicator-datetime-service

Does that help? How do you suggest I proceed in order to resolve this? I don't see any issues with displaying the correct date and time.

I have just read about the change to "Ayatana Indicators" in the 20.10 release notes.

Could there be a conflict with my configuration settings from 20.04?

1 Like

I'm getting this same 'crash' problem with Ayatana too. Is there a fix to sort this out at all?
Plus the clock is missing (although I can add a clock back using the Clock panel item)...

(Bluetooth icon was missing but managed to add this back reticking the AppIndicator item in Bluetooth plugin setup)

TIA!

I've tried a clean install, without setting up my whole system, and that seems to work OK.

Anyway, as I use this as my main computer, I've decided to stay with 20.04 LTS, and not waste time getting the system to work rather than just using it.

1 Like

If you look at the other recent crash thread JonG, I think I figured this problem out... it was from having the indicator-datetime package still installed as well as the new ayatana-indicator-datetime. I guess it is a bug that the old one is not removed during the upgrade process.

From the Ayatana Indicators github:
"When using the -datetime Ayatana Indicator, make sure that the -datetime Ubuntu Indicator (package name: indicator-datetime) is not installed."

1 Like

The best solution I found was just do a clean install. I believe that Eisenhiem is correct about the indicator-datetime package. Good work Eisenhiem.

1 Like

Same problem for me on a clean install of 20.10 (sha=06e26aa197eb8e7fc....). In fact, several crashes and the report dialog either does not come up or does not respond. Here is my 1st 3 hour history in /var/crash:

-rw-------  1 root   whoopsie   38809        Nov  2 14:58 samba-common.0.crash
-rw-r-----  1 ubuntu whoopsie 2482161 Nov  2 15:01 _usr_lib_aarch64-linux-gnu_brisk-menu_brisk-menu.1000.crash
-rw-r-----  1 ubuntu whoopsie   71061    Nov  2 17:40 _usr_bin_update-manager.1000.crash
-rw-r-----  1 ubuntu whoopsie   89755    Nov  2 17:42 _usr_bin_software-properties-gtk.1000.crash

Possibly related to that, all software installs and update stopped working with:
dpkg: unrecoverable fatal error, aborting:
files list file for package 'avahi-autoipd' is missing final newline

So things are apparently in a dead state. Not able to apt install anything or apt upgrade.

For "non-savant user" adoption of your work, please Linux/ubuntu/mate developers consider working on fundamental ways to mask/tolerate errors that are not relevant, keep real ones from corrupting the system, and/or ways to revert major functions to working state if corruption does occur.

Please make sure to create Bug Reports on these upgrade failures.