Please test that persistent live drives of Focal Fossa work in your computer(s)

[guiverc]

I can continue tomorrow; i'm too tired now & calling it quits for today...

[sudodus]

Your results are promising: Most of your tested computers are not affected, and in those affected Michael Hudson-Doyle's hack squashes the bug.

Have a good night's sleep

[C.S.Cameron]

I am using Focal from 20020325, Rufus 3.9, UNetbootin 677 Windows and mkusb-plug 2.5.5.

Mkusb-plug works fine for me with Persistent live drive except for the amount of time it takes to boot and "...end of device" scrolling.

Rufus worked ok until I dropped casper-initrd into casper folder and renamed it initrd.
Then I got Busybox
I named initrd back to casper-initrd and changed grub.cfg to suit.
It booted and worked, however it looked like it was booting syslinux.
I then edited txt.cfg to use casper-initrd and again got Busybox.

Unetbootin booted persistent with stock initrd, when casper-initrd was added and syslinux.cfg edited, Busybox resulted.

Computer s used EliteBook and Travelmate booting BIOS.

[sudodus]

@C.S.Cameron,

Thanks for sharing these results

I have a question and a few comments:

- Did you check which label was used in the different cases? We are focusing here to test if persistence works with casper-rw, and I think you used a version of mkusb-plug that creates the label writable.

- I'm afraid that you are using a deprecated version of Michael Hudson-Doyle's hack, 'casper-initrd'. The current version (downloaded from the same link as the original one) works with Rufus when used by Pete Batard. You can try to download the 'hack', casper-initrd.gz again. Or maybe you can download my compressed image file. It works for this test.

- On the other hand, if your HP Elitebook and Acer Travelmate work with the standard 'initrd' and 'casper-rw', they are not affected by the bug.

- Maybe you can test in some other computer.

---

So far (2020-04-09) we have found no HP that is affected (several models are tested), but the following computer brands (in most cases specific details can be found via links from this thread),

. three Dells (two by guiverc and one by me)
. two Intel NUCs (by Pete Batard)
. one ASUS (by Pete Batard)
. one Toshiba (by me)
. one Lenovo (by me)

and Michael Hudson-Doyle's hack squashes the bug in all these computers.

---

Edit: more questions:

- Which model of Acer Travelmate is it?
- Do I understand correctly, that your Travelmate is not affected by this bug? That persistence works also with the standard 'initrd'?

[C.S.Cameron]

Hi Sudodus:
I clicked the "This Hack" link (https://people.canonical.com/~mwh/casper-initrd.gz), in your Apr 9 PM and casper-initrd started downloading automatically.
I will check a mkusb-plug drive in my wife's computer, a gigabyte, (I am pretty sure a mkusb-plug drive works okay on it), if not I will look for the latest hack in the bug report and try it. (do you have the latest link handy)?
The Travelmate is a 8572T-6806, (10 years old). I have been testing 20.04 persistent drives on it in many forms, both persistent and Full install.
Cork

[sudodus]

Hi Cork,

I noticed that the current Lubuntu Focal daily iso file contains the 'hack' of Michael Hudson-Doyle: The package casper version 1.445 is bundled with the current Lubuntu iso file dated 2020-04-09. I tested it (made a persistent live USB-connected SSD) in one of my computers that were affected by this bug, the Toshiba. Persistence works with the label 'casper-rw'

So it is easy to

- simply use mkusb-plug with the current Lubuntu Focal daily iso file
- change the label of the partition for persistence from 'writable' to 'casper-rw'
- and then test if persistence works.

I guess this is what you mean in post #15.

---

Yesterday and the day before yesterday we used the image file from this link and described in post #4 above. And if you want to see the difference with and without the 'hack' of Michael Hudson-Doyle it is still a good way to test. (It is easier than to make an extracted system and add the 'hack' of Michael Hudson-Doyle into that system manually.)

[guiverc]

I finished my list

box (copied from list; most detail from lshw)	debug (/cow size)	start (/cow size)
hp dc7700 (c2d-e6320, 5gb, nvidia quadro nvs 290)	4.3/4.7gb	4.3/4.7gb
hp dc7900 (c2d-e8400, 4gb, intel 4 series integrated i915)	4.3/4.7gb	1.8/1.8gb
dell [optiplex] 745 (c2d-6600, 6gb, amd/ati radeon rv516/x1300/x1550)	4.2/4.7gb	4.2/4.7gb
dell [optiplex] 755 (c2d-e6850, 5gb, amd/ati radeon rv516/x1300/x1550)	4.4/4.7gb	2.4/2.4gb
dell [optiplex] 755 (c2d-e8300, 8gb, amd/ati radeon rv610/radeon hd2400 pro/xt)	4.4/4.7gb	4.4/4.7gb
dell [optiplex] 780 (c2q-q9400, 8gb?, amd/ati cedar radeon hd 5000/6000/7350/8350)	4.4/4.7gb	4.4/4.7gb
dell [optiplex] 960 (c2q-q9400, 8gb, amd/ati cedar radeon hd 5000/6000/7350/8350)	4.4/4.7gb	4.2/4.7gb
hp 8200 elite sff (i5-2400, 8gb, nvidia quadro 600)	4.3/4.7gb	4.3/4.7gb
dell vostro 430 (i7-870, 12gb, ??)	-	-
dell [optiplex] 990 (i7-2600, 16gb, nvidia geforce gt 6600 gt)	4.3/4.7gb	7.8/7.8gb
lenovo thinkpad sl510 (c2d-t6570, 2gb ram, i915)	4.4/4.7gb	4.4/4.7gb
lenovo thinkpad x201 (i5-m520, 4gb, i915)	4.4/4.7gb	4.4/4.7gb
motion computing j3400 (c2d-u9400, 4gb, intel mobile 4 series)	4.2/4.7gb	4.2/4.7gb
sony vaio ultrabook (i5-9400u, 4gb, intel haswell-ULT)	4.2/4.7gb	4.2/4.7gb

the dell 780 may have had a different amount of ram to that listed (I forgot to check; I'm borrowing ram from it in testing vostro_430 that's currently complaining about its ram).

list at https://docs.google.com/spreadsheets...DMM/edit#gid=0

[sudodus]

Thanks @guiverc,

- You found 3 computers that are affected by this bug.

- For the first time an HP computer is affected: hp dc7900 (c2d-e8400, 4gb, intel 4 series integrated i915).

- The 'hack' by Michael Hudson-Doylefixes it in these 3 computers.

- And all your computers work with the 'hack'.

Edit: Maybe you would like to copy your results into the corresponding bug report.

[C.S.Cameron]

Got my hands on my wife's Gigabyte GB-BXi7-5775
The UNetbootin syslinux drive worked okay with standard initrd.
It did not boot in BIOS mode with casper-initrd, (I was using version on Index of /~mwh).
It also did not boot in UEFI mode.
A 20.04 drive made with mkusb-plug worked fine on the Gigabyte, but took >15min to boot.
I will try with your current Lubuntu Focal daily iso file above, after midnight when rates are low.
(Decided to ride this Virus thing out in Sri Lanka).

Discovered that pressing Enter stops the scrolling.

[C.S.Cameron]

Tested "current Lubuntu Focal daily iso file" in three computers, works with both writable and casper-rw partitions in all three.