Repeated apport failures causing long boot delays

[Fritz Hudnut]

Full log in comment #4.

Been having problems with my installation of 20.10 U-MATE . . . reported here earlier to little fanfare . . . but the issues with very slow boot times, and then when finally arriving at GUI log in window, the log in "tray" is positioned upper left corner of the display and it's like a TTY, where it wants user name and then password. I think today it literally took 5 minutes to get logged into the GUI . . . .

As reported previously I have an install of Lu 20.10 in the same drive . . . so far is not having the same problems that U-MATE is having . . . .

ProblemType: Bug
DistroRelease: Ubuntu 20.10
Package: xorg 1:7.7+19ubuntu14
ProcVersionSignature: Ubuntu 5.4.0-42.46-generic 5.4.44
Uname: Linux 5.4.0-42-generic x86_64
NonfreeKernelModules: wl
ApportVersion: 2.20.11-0ubuntu45
Architecture: amd64
BootLog: Error: [Errno 13] Permission denied: '/var/log/boot.log'
CasperMD5CheckResult: skip
CompositorRunning: None
CurrentDesktop: MATE
Date: Wed Sep 2 07:28:33 2020
DistUpgraded: Fresh install
DistroCodename: groovy
DistroVariant: ubuntu
DkmsStatus: broadcom-sta, 6.30.223.271, 5.4.0-42-generic, x86_64: installed
DpkgLog:

ExtraDebuggingInterest: Yes, if not too technical
GraphicsCard:
 NVIDIA Corporation GK110 [GeForce GTX 780] [10de:1004] (rev a1) (prog-if 00 [VGA controller])
   Subsystem: eVga.com. Corp. GK110 [GeForce GTX 780] [3842:0781]
InstallationDate: Installed on 2020-01-20 (226 days ago)
InstallationMedia: Ubuntu-MATE 20.04 LTS "Focal Fossa" - Alpha amd64 (20200119)
MachineType: Apple Inc. MacPro5,1
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-5.4.0-42-generic root=UUID=e795b2e6-5bef-4a6e-9217-539a6debda90 ro quiet splash
SourcePackage: xorg
Symptom: display
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 07/30/2018
dmi.bios.vendor: Apple Inc.
dmi.bios.version: 138.0.0.0.0
dmi.board.asset.tag: 0
dmi.board.name: Mac-F221BEC8
dmi.board.vendor: Apple Inc.
dmi.chassis.type: 7
dmi.chassis.vendor: Apple Inc.
dmi.chassis.version: Mac-F221BEC8
dmi.modalias: dmi:bvnAppleInc.:bvr138.0.0.0.0:bd07/30/2018:svnAppleInc.:pnMacPro5,1:pvr0.0:rvnAppleInc.:rnMac-F221BEC8:rvr:cvnAppleInc.:ct7:cvrMac-F221BEC8:
dmi.product.family: MacPro
dmi.product.name: MacPro5,1
dmi.product.sku: System SKU#
dmi.product.version: 0.0
dmi.sys.vendor: Apple Inc.
version.compiz: compiz N/A
version.libdrm2: libdrm2 2.4.102-1ubuntu1
version.libgl1-mesa-dri: libgl1-mesa-dri 20.1.5-1
version.libgl1-mesa-glx: libgl1-mesa-glx N/A
version.xserver-xorg-core: xserver-xorg-core 2:1.20.8-2ubuntu4
version.xserver-xorg-input-evdev: xserver-xorg-input-evdev N/A
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:19.1.0-1ubuntu1
version.xserver-xorg-video-intel: xserver-xorg-video-intel 2:2.99.917+git20200714-1
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:1.0.16-1

[Daniel van Vugt (vanvugt)]

I can't find a simple answer for the slowness or any evidence of it in the log. However I can see this unsupported package installed, which will affect practically everything:

  libc6 2.31-2ubuntu1 [origin: unknown]

Please revert that package to a supported version:

  http://launchpadlibrarian.net/495370423/libc6_2.31-0ubuntu11_amd64.deb

Please also reproduce the very long boot again and then run:

  journalctl -b0 > journal.txt

and attach the resulting text file here.

[Fritz Hudnut (este-el-paz)]

@Daniel:

Thanks for the reply . . . it'll be a few hours before I can get back to U-MATE to test out your requests . . . .

But, after I filed this bug I posted the issue on the U-MATE developmental sub-forum and I got one reply with a link to a recent post on LM forum expressing a similar issue.

https://forums.linuxmint.com/viewtopic.php?f=46&t=329115&hilit=slow+boot

[Fritz Hudnut (este-el-paz)]

@Daniel:

Well, I couldn't exactly reproduce the extremely slow boot to a distorted GUI log in window . . . it wasn't exactly "zippy" but maybe it took two minutes to get from restart to log in window, which was in the proper position.

I ran the "journalctl" command and then the plan was to remove the "libc6 2.31-2ubuntu1" package . . . I tried to use GDebi to open you link to the regressed package, but GDebi said, "A newer package is already installed" . . . so then I tried to "apt rm libc6 2.31-2ubuntu1" . . . but it said, "Invalid operation rm" . . . .

Then I searched synaptic with "libc6 2.31-2ubuntu1" and it found nothing??? I tried to get to root in the terminal and I guess I haven't been added as sudoer by U-MATE?? Kind of a pain . . . . So, for now I'm "stuck" on how to get that package regressed . . . do I have to use TTY? Does this package even exist? If so, how or why didn't synaptic "see" it as "installed" and/or give me choices to regress back to??

Looks like the journalctl file is showing the endless streams of "Failed to start process" . . . again?? Had this issue a couple days back and seemed to get it cleaned up in the TTY, now back? TTY2 seems to be "open for business" . . . .

[Daniel van Vugt (vanvugt)]

To install the correct package version, run this in a terminal:

  wget http://launchpadlibrarian.net/495370423/libc6_2.31-0ubuntu11_amd64.deb
  sudo dpkg -i libc6_2.31-0ubuntu11_amd64.deb

and then reboot.

Separately, I think the slowness might be caused by bugs in the automatic crash reporter. To stop that happening please delete any files in /var/crash/

[Fritz Hudnut (este-el-paz)]

@Daniel:

Even though the previous time I booted in U-MATE seemed to go faster, this time again was very slow and again the log inwindow was squeezed into upper left corner.

Thanks for providing the wget data . . . I ran that and it came up with some errors, so I'm pasting that here before I reboot.

[CODE]$ sudo dpkg -i libc6_2.31-0ubuntu11_amd64.deb
[sudo] password for nonspace1:
dpkg: warning: downgrading libc6:amd64 from 2.31-2ubuntu1 to 2.31-0ubuntu11
(Reading database ... 345902 files and directories currently installed.)
Preparing to unpack libc6_2.31-0ubuntu11_amd64.deb ...
De-configuring libc6:i386 (2.31-2ubuntu1) ...
Unpacking libc6:amd64 (2.31-0ubuntu11) over (2.31-2ubuntu1) ...
dpkg: error processing package libc6:amd64 (--install):
 package libc6:amd64 2.31-0ubuntu11 cannot be configured because libc6:i386 is at a different version (2.31-2ubuntu1)
dpkg: error processing package libc6:i386 (--install):
 package libc6:i386 2.31-2ubuntu1 cannot be configured because libc6:amd64 is at a different version (2.31-0ubuntu11)
Errors were encountered while processing:
 libc6:amd64
 libc6:i386 [/CODE]

I also checked the /var/crash/ folder via the GUI and there were maybe 6 or 7 crash files in there, three of them were "locked" and properties said, "You are not the owner" . . . I right clicked on the other files and an error came up saying, "File can't be moved to trash" and then it had "cancel" "delete" or "something else???? (some other choice) . . . so I just closed the window and left the files as they were??? Is this something that has to be done in the console as root, or, clicking "delete" would "delete" the file "permanently" . . . bypassing "trash" and just going straight to linux "heck"????

[Fritz Hudnut (este-el-paz)]

On reboot again it took a looonnnngggg time to get to the tiny log in window . . . and the international red dot with line through it applet in toolbar upper right said, "unmet dependencies" . . . launched synaptic, seems like it wants to add the "2.31-Oubuntu" package back in?? and remove three others?? It says "4 broken packages" . . . .

Anyway, I left things as they are there . . . .

Maybe I'll try to "delete" the items that aren't locked in /var/crash/ . . . .

Basically the slow boot issue is still continuing . . . to continue.

[Fritz Hudnut (este-el-paz)]

OK . . . I deleted "immediately" the three or four crash logs that weren't locked, and on reboot it went faster and login window was in proper place . . . . But an "error" sound sounded when the log in window opened on the screen . . . and the "broken packages" symbol is still there.

I'll have to try a few more times, because the problem seems to come and go, etc.

Seems like cutting down on the crash logs maybe helped???

[Daniel van Vugt (vanvugt)]

To resolve the remaining slowness please open a Terminal window and run:

  sudo rm /var/crash/*

and then reboot.

Separately, to try and resolve your broken package problem try:

  sudo apt update
  sudo apt full-upgrade
  Reboot.

If you still can't solve it after that then I suggest reinstalling the whole OS. But next time use Ubuntu 20.04 which is supported, stable, and won't have these kinds of problems.

  https://releases.ubuntu.com/20.04/ or
  http://cdimage.ubuntu.com/ubuntu-mate/releases/20.04/release/

[Fritz Hudnut (este-el-paz)]

@Daniel:

Thanks again for those commands . . . I didn't know if removing the whole crash directory was the way to remove those files . . . .

So I ran that command, then I did the "apt update" . . . but then I ran "apt -f install" . . . and that showed a bunch of things to add and subtract . . . so I ran that through and then I did your "apt full-upgrade" . . . and then "sudo reboot" . . . and reboot took it's time.

But on restart it was faster to the log in window . . . still got the "clang-a-lang" sound when it loaded, but logging in went "OK" . . . not the zippiest, but acceptable . . . and no universal error red dot symbol!!! So that is more or less OK . . . I probably won't need to re-install . . . I have enough linux distros that I don't need stone axe reliability, but I do prefer systems that "work" when clicking on stuff . . . not taking 60 seconds to reboot from console, etc.

For now I think we have "solved" the complaint in this bug report . . . however, I have another bug report filed on U-MATE && Lubuntu 20.10 and that bug happened again while running through the "apt full-upgrade" commands suggested to fix the slow boot issues of this bug . . . and that is that ubuntu seems to be wiping the EFI grub data . . . leaving only itself on the grub listing . . . . It's a pain to have to use the SG2 disk to boot over into OpenSUSE to fix a problem that ubuntu seems to be causing . . . devs don't seem to be able to address that issue, which has been going on for months and effects a fair number of people . . . . If you have any experience in the grub area and can offer some suggestions that would be helpful . . . . I can't believe it happened yet again today in spite of my looking through the packages and not seeing "grub-xxxx" as it has been in the recent events with this problem . . . . Ubuntu does not seem to be playing well with others . . . for might be the last 9 -12 months??

https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/1894242

[Brian Murray (brian-murray)]

This is likely a duplicate of bug 1891657.

[Daniel van Vugt (vanvugt)]

OK, duplicate until proven otherwise.