kde partition manager is waiting it appears for /dev/fd0 (no floppy drive)

[Chris Guiver]

Lubuntu 20.04.1 daily LIVE QA-test on
dell [optiplex] 745 (c2d-6600, 6gb, amd/ati radeon rv516/x1300/x1550)

Opening random programs, one program selected was KDE Partition Manager, which failed to be usable.

** Actual results

Opened KDE Partition Manager
A Scanning box/window opened and is still reporting "Scanning device '/dev/fd0'" ~five minutes later, and is not advancing. A circling logo occurs when pointer is over the KDE Partition Manager window.

** Expected Results

The box has no floppy drive, so I don't expect it to stop at scanning /dev/fd0, and KDE Partition Manager should be usable.

This issue has been noticed before, but only reported in comments (I believe), or if reported it didn't include `dmesg` output & I changed it to incomplete.. (when/if I find it, I'll mark it duplicate of this)

ProblemType: Bug
DistroRelease: Ubuntu 20.10
Package: partitionmanager 4.1.0-1
ProcVersionSignature: Ubuntu 5.4.0-42.46-generic 5.4.44
Uname: Linux 5.4.0-42-generic x86_64
ApportVersion: 2.20.11-0ubuntu43
Architecture: amd64
CasperMD5CheckResult: pass
CasperVersion: 1.452
CurrentDesktop: LXQt
Date: Tue Aug 4 05:12:22 2020
LiveMediaBuild: Lubuntu 20.10 "Groovy Gorilla" - Alpha amd64 (20200801)
SourcePackage: partitionmanager
UpgradeStatus: No upgrade log present (probably fresh install)

[Chris Guiver (guiverc)]

What was missing on prior reporting of this issue I believe (dmesg)

Of note - last time when I rebooted (in hopes of getting `dmesg`) it didn't occur again..
This is also a different box; last box was d755-5 I believe (different but similar model dell)

[Chris Guiver (guiverc)]

On d745 or box this issue occurred on, i used `killall` to kill partitionmanager. It closed, then I ran (hoping to get more detail, or repeat the issue)

lubuntu@lubuntu:~$ partitionmanager
Could not load plugin for core backend "" : "The shared library was not found."
Loaded backend plugin: "pmsfdiskbackendplugin"
Icon theme "elementary" not found.

but unexpected no kde partition manager opened (a flash did occur, brief and about correct size for partitionmanager windwo, but it was just a quick flash). Attempting to re-start it again from terminal had the same result.

I didn't see anything helpful added to `dmesg`, however `ournalctl` had the following which maybe related

Aug 04 05:30:23 lubuntu dbus-daemon[1059]: [system] Rejected send message, 1 matched rules; type="method_call", sender=":1.67" (uid=999 pid=3322 comm="partitionmanager " label="unconfined") interface="org.freedesktop.DBus.Introspectable" member="Introspect" error name="(unset)" requested_reply="0" destination="org.kde.kpmcore.helperinterface" (uid=0 pid=1864 comm="/usr/lib/kauth/libexec/kpmcore_externalcommand " label="unconfined")
Aug 04 05:31:59 lubuntu dbus-daemon[1059]: [system] Rejected send message, 1 matched rules; type="method_call", sender=":1.68" (uid=999 pid=3330 comm="partitionmanager " label="unconfined") interface="org.freedesktop.DBus.Introspectable" member="Introspect" error name="(unset)" requested_reply="0" destination="org.kde.kpmcore.helperinterface" (uid=0 pid=1864 comm="/usr/lib/kauth/libexec/kpmcore_externalcommand " label="unconfined")
Aug 04 05:33:39 lubuntu kernel: perf: interrupt took too long (3145 > 3141), lowering kernel.perf_event_max_sample_rate to 63500

[Chris Guiver (guiverc)]

Rebooted box & tested again on
dell [optiplex] 745 (c2d-6600, 6gb, amd/ati radeon rv516/x1300/x1550)

(Before booting I checked BIOS, floppy is set to 'internal' for box; box has no floppy drive though. Maybe a change to 'disabled' would cause this hang to not occur)

only KDE partition manager was loaded, it appeared the same result (it paused) so I went to other room to report findings in iso.qa.ubu, when I got back kde partition manager was busy waiting.

Found in `dmesg` is
[ 127.478906] blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 127.478912] floppy: error 10 while reading block 0
[ 165.636346] blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 165.636352] floppy: error 10 while reading block 0
[ 203.949342] blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 203.949348] floppy: error 10 while reading block 0
[ 242.091826] blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 242.091832] floppy: error 10 while reading block 0
[ 280.238087] blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 280.238093] floppy: error 10 while reading block 0

What I think of as useful (line or two extra) from `journalctl`
Aug 04 05:40:02 lubuntu systemd[1293]: Starting Virtual filesystem metadata service...
Aug 04 05:40:02 lubuntu dbus-daemon[1379]: [session uid=999 pid=1379] Successfully activated service 'org.gtk.vfs.Metadata'
Aug 04 05:40:02 lubuntu systemd[1293]: Started Virtual filesystem metadata service.
Aug 04 05:40:02 lubuntu dbus-daemon[1059]: [system] Successfully activated service 'org.freedesktop.UPower'
Aug 04 05:40:02 lubuntu systemd[1]: Started Daemon for power management.
Aug 04 05:40:02 lubuntu ofonod[1078]: enum_devices_reply: no battery detected
Aug 04 05:40:02 lubuntu dbus-daemon[1059]: [system] Activating via systemd: service name='org.freedesktop.UDisks2' unit='udisks2.service' requested by ':1.41' (uid=999 pid=1652 comm="/usr/bin/lxqt-panel " label="unconfined")
Aug 04 05:40:06 lubuntu kernel: blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
Aug 04 05:40:06 lubuntu kernel: floppy: error 10 while reading block 0
Aug 04 05:40:20 lubuntu NetworkManager[1060]: <warn> [1596519620.6199] dhcp6 (enp3s0): request timed out
Aug 04 05:40:20 lubuntu NetworkManager[1060]: <info> [1596519620.6200] dhcp6 (enp3s0): state changed unknown -> timeout
Aug 04 05:40:27 lubuntu dbus-daemon[1059]: [system] Failed to activate service 'org.freedesktop.UDisks2': timed out (service_start_timeout=25000ms)
Aug 04 05:40:30 lubuntu systemd-udevd[999]: fd0: Worker [1015] processing SEQNUM=3315 is taking a long time
Aug 04 05:40:44 lubuntu kernel: blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
Aug 04 05:40:44 lubuntu kernel: floppy: error 10 while reading block 0
Aug 04 05:40:56 lubuntu systemd[1]: udisks2.service: start operation timed out. Terminating.
Aug 04 05:40:57 lubuntu dbus-daemon[1059]: [system] Activating se...

[Chris Guiver (guiverc)]

This box (d745) just reported battery needs replacing, maybe BIOS settings have invalid values which could be impacting or causing bug to occur, maybe also why it occurred first time, but didn't on reboot.

Regardless I went into BIOS & changed setting to 'disable' (floppy drive)

On rebooting into Lubuntu daily 'live', I started KDE partition manager, and box (scanning) did open but was too quick for me to read anything, so issue has no gone completely, or at least didn't re-occur this time.

[Ubuntu QA Website (ubuntuqa)]

This bug has been reported on the Ubuntu ISO testing tracker.

A list of all reports related to this bug can be found here:
http://iso.qa.ubuntu.com/qatracker/reports/bugs/1890236

[Andrius Štikonas (stikonas)]

This should have been upstream bug report, it doesn't look distro specific. Anyway, to get some useful info, it would be better to run partitionmanager with some debugging:

$ KPMCORE_DEBUG=1 partitionmanager

Then it will print to terminal which commands it is running. This might tell you where it is getting stuck.

[Chris Guiver (guiverc)]

@Andrius/stikonas

If you'd like this filed anywhere else, please just ask. (I usually verify reports on opensuse/fedora/debian before I file upstream; and that wasn't done, maybe hard to reproduce due lack of floppy support in many machines BIOS that is possibly essential to re-create issue)

Today's re-test on box documented here had a delay only on starting (less than a minute) before it responded and was usable. This test is using groovy/20.10 (not focal/20.04.1/2 daily)

** system in operation
Lubuntu groovy daily (live) 2020-10-18 on
- dell [optiplex] 745 (c2d-6600, 6gb, amd/ati radeon rv516/x1300/x1550)

apt-cache policy partitionmanager shows
partitionmanager:
  Installed: 4.1.0-1
        500 http://archive.ubuntu.com/ubuntu groovy/universe amd64

Full KPMCORE_DEBUG=1 output can be seen at
https://paste.ubuntu.com/p/QRnXcSBNsS/

It did not STOP & fail to continue today.. it just sat there for some time (less than a minute I believe but noticeable) with the last messages on terminal during 'pause' being
--
"Scanning devices..."
"Command: lsblk --nodeps --paths --sort name --json --output type,name"
"Command: blockdev --getsize64 /dev/fd0"
--
The message box that opened during this time read
"Scanning device '/dev/fd0'"
--

Scanning `dmesg` I see
--
[ 853.260553] blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 853.260560] floppy: error 10 while reading block 0
--

Closing and re-starting (after delay) partitionmanager caused the following to appear in `dmesg
--
[ 2476.713193] perf: interrupt took too long (2502 > 2500), lowering kernel.perf_event_max_sample_rate to 79750
[ 2529.179311] blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 2529.179318] floppy: error 10 while reading block 0
--
(ie. same result except for 'perf:' line that maybe unrelated)

On fourth run of `KPMCORE_DEBUG=1 partitionmanager`
the following is a paste from `journalctl`
--
Oct 19 05:32:09 lubuntu dbus-daemon[1002]: [system] Activating service name='org.kde.kpmcore.externalcommand' requested by '>
Oct 19 05:32:09 lubuntu dbus-daemon[1002]: [system] Successfully activated service 'org.kde.kpmcore.externalcommand'
Oct 19 05:32:14 lubuntu sudo[2688]: pam_unix(sudo:session): session closed for user root
Oct 19 05:32:47 lubuntu kernel: blk_update_request: I/O error, dev fd0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio cla>
Oct 19 05:32:47 lubuntu kernel: floppy: error 10 while reading block 0
--

[Chris Guiver (guiverc)]

Same ISO/thumb-drive as prior comment (#8), so
Lubuntu groovy daily (live) 2020-10-18 on
- dell [optiplex] 755 (c2d-e6850, 5gb, amd/ati radeon rv516/x1300/x1550)

NO NOTICEABLE DELAY was expericenced with floppy inserted, or not-inserted during execution on this box in running `KPMCORE_DEBUG=1 partitionmanager`

This box HAS A FLOPPY DRIVE, box was booted with floppy inserted.

If helpful, messages can be see via
--
lubuntu@lubuntu:~$ pastebinit kpmcore_floppy_inserted.txt
https://paste.ubuntu.com/p/gW99hVSys4/
lubuntu@lubuntu:~$ pastebinit kpmcore_no-floppy_inserted.txt
https://paste.ubuntu.com/p/Nz2sHG9ygy/
--

I did NOT experience hangs on either box in comment #8 or this #9 using Ubuntu groovy (just a delay on d745 or box that doesn't have a floppy drive)

The initial report mentions FOCAL (20.04.1)?; though I notice it was filed on a groovy box?? That was just hours/days before 20.04.1's release, so I'd have been testing focal far more at the time than groovy.

[Andrius Štikonas (stikonas)]

Hmm, so it looks like blockdev --getsize64 /dev/fd0 is taking a long time.

Can you confirm this without partitionmanager by running

sudo blockdev --getsize64 /dev/fd0

But from dmesg it looks like blockdev just waits for kernel to reply.