Ubuntu trashed

[kilaueabart]

A couple days ago I opened up my X1000 to see if it was dirty and overheating, or had bad connections, causing failure to reboot after a period of use. I found it remarkably clean inside and didn't do much but wiggle some of the boards and cord connections.

The next day I wanted to boot Ubuntu. I first tried the one with kernel 18, and it failed, flashing by a lot of text. Once before my kernel 18 option had failed to boot, but without all the chatter -- it had just left me with a permanent Amiga OS4 splash screen. On that previous occasion I then tried a kernel 14 version and that had worked fine. So I tried kernel 14 again, but this time it didn't work. Lots of text scrolling by but not as much as kernel 18 had produced.

I wonder what my next step should be. I must have the disk I originally installed Ubuntu with and maybe I should do so again, perhaps on my primary hard drive this time (I seem to have lots of extra space on it). That would leave me with the loss of everything I had added or changed in the to-be-abandoned version.

Or would it make sense to post the photos I took of the messages that failure to boot brought in hopes that someone would glance at them and see immediately that there was a reasonably simple fix? Following an "EXT3-fs (sdb7: error: couldn't mount because of unsupported optional features (24)" and a duplicate of that starting "EXT2-fs ..." there's lots of "ata2.00:" stuff, ending with "revalidation failed" in one case of boot failure. The other case followed that stuff with much EXT4-fs and "(sdb7)" errors repeated over and over.

(Meanwhile, my various OS4's are booting and rebooting just fine now.)

[xeno74]

Or would it make sense to post the photos I took of the messages that failure to boot brought in hopes that someone would glance at them and see immediately that there was a reasonably simple fix?

Yes please.

[kilaueabart]

What happened to the two jpegs I attached to a reply yesterday? I'll try again, but in a separate reply, since my whole reply seems to have disappeared.

[kilaueabart]

Okay, today it's telling me I'm not writing enough. I'm not sure how much I need to write, but I'll say that this is a photo of what the screen finally looked like when it quit trying to launch UbuntuLT with kernel 18. I didn't try to take a picture until the writing stopped although I could make a video and catch it pretty much from the beginning.

Maybe I have typed enough. Preview shows my text and my jpeg...

[kilaueabart]

Now I'll try to write enough stuff to get the other jpeg posted. This one stopped as I recall with just the one page, but I had to press the reset key on the X1000 box.

OK, no complaint about any shortage of words, so maybe sending them separately works? I tried two to a single post last time.

[xeno74]

Kilaueabart,

Thank you very much for these photos. OK I understand what is happened. This is a bug we also have in the RC1+2+3 of kernel 3.19. It was fixed in kernel 3.19 RC4. But I didn't know that the kernel 3.18 has the same problem. Thank you for posting that problem with kernel 3.18. Could you try the latest RC of kernel 3.19, please?

Download: vmlinux-3.19.0-rc7-KVM-PR-AMIGA_one_X1000.tar.bz2

Thanks in advance.

Cheers,

Christian

[kilaueabart]

I'm sure it said my response several hours ago had been posted!

OK, I try again. I was a bit pessimistic, because a kernel 18 problem shouldn't cause the kernel 14 Ubuntu to go bad at the same time, but I have installed kernel 19. Sure enough, it does not work; there must be something more basic wrong. But I attach another photo, this time to the end where it promises to reboot in 180 seconds.

kernel19.jpg

By the way, I'm not sure I know what I'm doing. How much of the stuff I have in cf0 really belongs there? And is anything missing?
Directory "Kingston:" on Thursday 05-Feb-15
nemo_3.18-3.patch 30053 ----rwed 26-Nov-14 05:42:38
kernel_3.12.0.config 75428 ----rw-d 04-Nov-13 04:35:54
vmlinux-3.12 18338734 ----rwed 04-Nov-13 04:32:44
vmlinux-3.14 18584865 ----rwed 30-Mar-14 19:52:36
.Trash-1000 Dir ---arwed 28-Oct-14 15:19:34
vmlinux-3.17.0_KVM-PR-FTRACE-AmigaOneX1000 Dir ---arwed 13-Oct-14 11:54:22
vmlinux-3.17 21571096 ----rwed 05-Oct-14 18:33:12
vmlinux-3.18.0-rc6-2-KVM-PR-OVERLAYFS-AMIGA_one_X1000 Dir ---arwed 26-Nov-14 07:15:56
OS4.1-Linux-icon.lha 15956 ----rwed 25-Mar-14 22:03:00
nemo_3.18-1.patch 28758 ----rwed 21-Oct-14 03:19:24
nemo_3.18.0-rc1_KVM-PR_FTRACE.config 79787 ----rwed 21-Oct-14 04:03:06
vmlinux-3.18

Added today only three:
vmlinux-3.19
nemo_3.19-rc7_KVM-PR.config
nemo_3.19-2.patch

[xeno74]

I'm sure it said my response several hours ago had been posted!

OK, I try again. I was a bit pessimistic, because a kernel 18 problem shouldn't cause the kernel 14 Ubuntu to go bad at the same time, but I have installed kernel 19. Sure enough, it does not work; there must be something more basic wrong. But I attach another photo, this time to the end where it promises to reboot in 180 seconds.

kernel19.jpg

By the way, I'm not sure I know what I'm doing. How much of the stuff I have in cf0 really belongs there? And is anything missing?
Directory "Kingston:" on Thursday 05-Feb-15
nemo_3.18-3.patch 30053 ----rwed 26-Nov-14 05:42:38
kernel_3.12.0.config 75428 ----rw-d 04-Nov-13 04:35:54
vmlinux-3.12 18338734 ----rwed 04-Nov-13 04:32:44
vmlinux-3.14 18584865 ----rwed 30-Mar-14 19:52:36
.Trash-1000 Dir ---arwed 28-Oct-14 15:19:34
vmlinux-3.17.0_KVM-PR-FTRACE-AmigaOneX1000 Dir ---arwed 13-Oct-14 11:54:22
vmlinux-3.17 21571096 ----rwed 05-Oct-14 18:33:12
vmlinux-3.18.0-rc6-2-KVM-PR-OVERLAYFS-AMIGA_one_X1000 Dir ---arwed 26-Nov-14 07:15:56
OS4.1-Linux-icon.lha 15956 ----rwed 25-Mar-14 22:03:00
nemo_3.18-1.patch 28758 ----rwed 21-Oct-14 03:19:24
nemo_3.18.0-rc1_KVM-PR_FTRACE.config 79787 ----rwed 21-Oct-14 04:03:06
vmlinux-3.18

Added today only three:
vmlinux-3.19
nemo_3.19-rc7_KVM-PR.config
nemo_3.19-2.patch

Hi Kilaueabart,

It could be, that you hard disk is the problem. On your CF card you only need the kernel files itself. For example vmlinux-3.19.

Rgds,

Christian

[mechanic]

Looking at the output you posted I would say the filesystem is trashed, it could also be the drive or a loose connection/bad cable.

You could check the drive with one of the USB Linux systems or use the install CD to see if the filesystem is still good, as long as you do not tell the install system to 'write to disk' it should not hurt anything. Be sure to use the Manual disk prep option, not the ones that will automatically partition and format the drive.

Also, if you must re-install everything I highly recommend you use ext3 for the filesystem as ext4 does not build cleanly. I suspect there is still some little endian code floating around in the ext4 source code. Ext2, ext3, and DOS all build clean.

There is also a way to break the install script and use fdisk to poke around, but I would have to refresh my memory to guide you through that procedure.

[kilaueabart]

On your CF card you only need the kernel files itself. For example vmlinux-3.19.

Good! ".conf" and ".patch" sound important, but maybe I should assume those are just data on how the kernel was fixed up?

A new curiousity. I can't even see the CF card from my backup 4.1u6 anymore. It's quite evident from FE.

I too have been suspecting a disk problem. I'm going to try to use mechanic's suggestions and see what turns up.