Disk errors on VBOX HARDISK - Ubuntu 18.04.2 LTS - on Mojave 10.14.3

Discussions related to using VirtualBox on Mac OS X hosts.
Post Reply
fustbariclation
Posts: 14
Joined: 21. Feb 2013, 10:09

Disk errors on VBOX HARDISK - Ubuntu 18.04.2 LTS - on Mojave 10.14.3

Post by fustbariclation »

I'm getting errors on the Virtualbox hard drive:

Code: Select all

ata1.0 - /dev/mapper/ubuntu--vg-root

$ df -H
/dev/mapper/ubuntu--vg-root  245G  102G  131G  44% /
I'm not getting any errors on the MacOS disk where the virtualbox image resides.

Such as:

Code: Select all

kernel: [  201.293055] ata1.00: device reported invalid CHS sector 0
kernel: [  411.734386] ata1.00: configured for UDMA/133
kernel: [ 3997.800467] ata1.00: exception Emask 0x0 SAct 0x1e00 SErr 0x0 action 0x6 frozen
kernel: [ 3997.800519] ata1.00: failed command: WRITE FPDMA QUEUED
kernel: [ 3997.800540] ata1.00: cmd 61/98:48:88:9b:54/00:00:0e:00:00/40 tag 9 ncq dma 77824 out
kernel: [ 3997.800579] ata1.00: status: { DRDY }
kernel: [ 3997.800593] ata1.00: failed command: READ FPDMA QUEUED
kernel: [ 3997.800612] ata1.00: cmd 60/18:50:28:bd:d3/00:00:00:00:00/40 tag 10 ncq dma 12288 in
kernel: [ 3997.800654] ata1.00: status: { DRDY }
kernel: [ 3997.800664] ata1.00: failed command: READ FPDMA QUEUED
kernel: [ 3997.800683] ata1.00: cmd 60/20:58:b8:93:41/00:00:00:00:00/40 tag 11 ncq dma 16384 in
kernel: [ 3997.800724] ata1.00: status: { DRDY }
kernel: [ 3997.800734] ata1.00: failed command: READ FPDMA QUEUED
kernel: [ 3997.800754] ata1.00: cmd 60/c0:60:20:1b:c6/00:00:07:00:00/40 tag 12 ncq dma 98304 in
kernel: [ 3997.800795] ata1.00: status: { DRDY }
kernel: [ 4002.999673] ata1.00: configured for UDMA/133
kernel: [ 4002.999684] ata1.00: device reported invalid CHS sector 0
kernel: [ 4002.999689] ata1.00: device reported invalid CHS sector 0
kernel: [ 4002.999690] ata1.00: device reported invalid CHS sector 0
kernel: [ 4002.999691] ata1.00: device reported invalid CHS sector 0
kernel: [ 5796.931332] ata1.00: exception Emask 0x0 SAct 0x100000 SErr 0x0 action 0x6 frozen
kernel: [ 5796.931386] ata1.00: failed command: WRITE FPDMA QUEUED
kernel: [ 5796.931414] ata1.00: cmd 61/40:a0:a8:48:55/00:00:0e:00:00/40 tag 20 ncq dma 32768 out
kernel: [ 5796.931451] ata1.00: status: { DRDY }
kernel: [ 5802.196581] ata1.00: configured for UDMA/133
kernel: [ 5802.196589] ata1.00: device reported invalid CHS sector 0
I've tried fsck on boot - with:

Code: Select all

# touch /forcefsck
It reports and fixes various missing inodes, but gives no serious errors.

I have also tried checking that there's no software corruption:

Code: Select all

# apt-get update --fix-missing
# debsums -cs
# apt-get -f install
I get no errors from the above.

This is causing very slow boots. When I try to identify what is causing the problem, I get:

Code: Select all

# systemd-analyze blame
    7min 30.310s rc-local.service
     7min 4.939s mysql.service
    2min 43.725s vboxadd.service
    2min 39.277s loadcpufreq.service
    2min 20.262s tor@default.service
     2min 7.504s virtualbox.service
    1min 48.285s cachefilesd.service
    1min 37.875s dev-mapper-ubuntu\x2d\x2dvg\x2droot.device
    1min 26.501s apt-daily.service
    1min 22.152s vncserver.service
    1min 20.911s grub-common.service
    1min 18.190s nfs-server.service
    1min 16.583s apport.service
    1min 15.635s mixmaster.service
     1min 5.954s gdomap.service
     1min 1.492s hddtemp.service
      1min 785ms systemd-udev-settle.service
       1min 93ms ModemManager.service
         55.609s init_urandom.service
         55.476s pppd-dns.service
         54.513s speech-dispatcher.service
         53.274s rsyslog.service
         52.715s stunnel4.service
         46.240s sysstat.service
         45.487s apparmor.service
         44.978s cgroupfs-mount.service
         41.368s systemd-udev-trigger.service
         40.405s cpufrequtils.service
         39.431s systemd-user-sessions.service
         38.549s networkd-dispatcher.service
         33.118s wpa_supplicant.service
         31.148s alsa-restore.service
         30.187s postfix@-.service
         29.006s nmbd.service
         28.497s nfs-mountd.service
         27.084s vncserver@1.service
         26.235s networking.service
         26.096s sysfsutils.service
         25.845s keyboard-setup.service
         24.661s vncserver@:1.service
         24.331s kerneloops.service
         23.220s systemd-logind.service
         20.106s systemd-tmpfiles-clean.service
         19.622s tor.service
         18.554s thermald.service
         15.504s systemd-journald.service
         12.812s smbd.service
         12.611s systemd-journal-flush.service
         11.729s dev-hugepages.mount
         11.703s proc-fs-nfsd.mount
         11.452s kmod-static-nodes.service
         11.349s packagekit.service
         11.285s binfmt-support.service
         11.131s systemd-sysctl.service
         10.869s avahi-daemon.service
         10.779s run-rpc_pipefs.mount
         10.348s systemd-remount-fs.service
Any suggestions gratefully received!

I've attached the latest vbox log file - but I can't find any errors related to disk problems in it.
Attachments
VBox.log
(76.79 KiB) Downloaded 23 times
Last edited by socratis on 27. Mar 2019, 19:43, edited 3 times in total.
Reason: Replace use of "list" with code brackets.
mpack
Site Moderator
Posts: 39156
Joined: 4. Sep 2008, 17:09
Primary OS: MS Windows 10
VBox Version: PUEL
Guest OSses: Mostly XP

Re: Disk errors on VBOX HARDISK - Ubuntu 18.04.2 LTS - on Mojave 10.14.3

Post by mpack »

VirtualBox drives are simulated using host media. If you get a physical media error inside a VM then the host drive is faulty. There is certainly evidence of host I/O problems in the log:
02:33:17.476434 AHCI#0: Port 0 reset
02:33:18.095754 VD#0: Write request was active for 31 seconds
02:33:18.095785 VD#0: Cancelling all active requests
18:22:30.074787 AHCI#0: Port 0 reset
18:22:32.946674 VD#0: Write request was active for 34 seconds
18:22:33.587127 VD#0: Cancelling all active requests
18:22:33.587398 AHCI#0: Port 0 reset
18:22:33.598493 VD#0: Cancelling all active requests
Long delays on write usually mean lots of retries, e.g. because of a bad sector. There would have been retry noises comes from a rotational drive. A bad sector will not be found using a simple host filesystem check - it requires a thorough scan, which may need to run overnight. The log is truncated at that point - evidently it was grabbed while the VM was still running.

Btw, this host cannot spare 4 cores for a VM as that is 100% of available cores - it leaves nothing for the host.

Also, you have duplicate shared folders.


I would try reducing guest cores to two. Perhaps the sector failures are just a mirage that a more sensible sharing of CPU will eliminate.
fustbariclation
Posts: 14
Joined: 21. Feb 2013, 10:09

Re: Disk errors on VBOX HARDISK - Ubuntu 18.04.2 LTS - on Mojave 10.14.3

Post by fustbariclation »

Thank you for the recommendations - I'll try them.
Post Reply