Re: Updated Debian Ports installation images 2021-09-23

Hi Ulrich!

On 10/11/21 20:44, Ulrich Teichert wrote:
> That's why I used two different CD-ROM drives *and* burned two CD-ROMs on different
> types of blank CDs. I think that rules out hardware issues, as the error is exactly
> the same, even happens on the same sector, when I try to mount the CD-ROM from
> a shell:
> 
> With Liftec media:
> Oct 11 20:15:08 main-menu[268]: INFO: Menu item 'di-utils-shell' selected
> Oct 11 20:15:32 kernel: [  260.904163] sr 0:0:4:0: [sr0] tag#33 unaligned transfer
> Oct 11 20:15:32 kernel: [  260.905139] blk_update_request: I/O error, dev sr0, sector 64 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
> Oct 11 20:15:32 kernel: [  260.905139] isofs_fill_super: bread failed, dev=sr0, iso_blknum=16, block=32
> 
> With Medion media:
> Oct 11 20:34:10 main-menu[269]: INFO: Menu item 'di-utils-shell' selected
> Oct 11 20:34:30 kernel: [  258.741078] sr 0:0:4:0: [sr0] tag#380 unaligned transfer
> Oct 11 20:34:30 kernel: [  258.742055] blk_update_request: I/O error, dev sr0, sector 64 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
> Oct 11 20:34:30 kernel: [  258.742055] isofs_fill_super: bread failed, dev=sr0, iso_blknum=16, block=32

OK, I don't want to rule out a software problem. However, I would advise to use
Taiyo Yuden media if you still can those. Those are usually those with the highest
quality and reliability.

> I am getting a lot of other messages for unaligned transfers on /dev/sr0, too:
> 
> Oct 11 20:11:49 kernel: [   28.658188] sr 0:0:4:0: [sr0] scsi-1 drive
> Oct 11 20:11:49 kernel: [   28.659165] cdrom: Uniform CD-ROM driver Revision: 3.--More-- (32% of 53020                           
> Oct 11 20:11:49 kernel: [   28.706039] sr 0:0:4:0: Attached scsi CD-ROM sr0
> ...
> Oct 11 20:11:49 kernel: [   30.755843] sr 0:0:4:0: [sr0] tag#20 unaligned transfer
> Oct 11 20:11:49 kernel: [   30.756820] blk_update_request: I/O error, dev sr0, sector 583127 op 0x0:(READ) flags 0x0 phys_seg 9 prio class 0
> Oct 11 20:11:49 kernel: [   30.759749] Buffer I/O error on dev sr0, logical block 583127, async page read
> Oct 11 20:11:49 kernel: [   30.760726] Buffer I/O error on dev sr0, logical block 583128, async page read
> ... etc..
> 
> Getting the full log would be a bit tedious, as I can't proceed to the network
> configuration...

I had another idea after I replied your last mail and I think one possible culprit could
be xorrisofs which was used for building the ISO image instead of genisoimage which Debian
used in the past.

I can actually create an installation image with genisoimage instead and you can test it.

Also, let's pull Thomas Schmitt from libburnia upstream who might know more.

>> FWIW, the change to re-enable EISA on alpha in the Debian kernel config has not been
>> merged yet due to bike-shedding over the proper changelog entry ... [1].
> 
> Oh well, eventually, it will get merged ;-?

Yes, it will be merged. No worries. Hopefully later this week. I didn't have the time for
a rebase yet.

Adrian