Re: 6.1.0-15/6.1.66-1 broken too?

2023-12-11 Thread Kevin Price
Please see https://bugs.debian.org/1057967 for follow-up.

Am 11.12.23 um 06:21 schrieb Stephan Verbücheln:
[...]
> My hardware is a 2014 Macbook Pro (Intel CPU and graphics).

Stephan and all, would you please post your information there? TIA
-- 
Kevin Price



Re: 6.1.0-15/6.1.66-1 broken too?

2023-12-10 Thread Stephan Verbücheln
Hello everybody

I can confirm the same problems. At first I thought the network problem
was due to proprietary Broadcom driver because network connectivity was
the most obvious problem.

However, most problems persisted after removing the driver. I do not
have any other proprietary or custom kernel modules.

My hardware is a 2014 Macbook Pro (Intel CPU and graphics).

Regards
Stephan


signature.asc
Description: This is a digitally signed message part


Re: 6.1.0-15/6.1.66-1 broken too?

2023-12-10 Thread Kevin Price
I confirm that 6.1.66-1 (6.1.0-15) severely breaks my
amd64/bookworm/gnome physical machine, which runs fine with 6.1.52-1 and
6.1.55-1.

Am 10.12.23 um 20:24 schrieb Andrew M.A. Cater:
> On Sun, Dec 10, 2023 at 08:02:03PM +0100, Kevin Price wrote:
>> Am 09.12.23 um 19:09 schrieb Dan Ritter:
>>> The new kernel release is reported to contain an ext4 data
>>> corruption bug. It's prudent not to upgrade, or if you have
>>> started to upgrade, not to reboot, until a new kernel release
>>> is prepared.
>>
>> Thanks for your announcement. I'm running out of time to properly report
>> a bug against 6.1.66-1.
>>
>> #1057843 states that it's fixed with 6.1.0-15/6.1.66-1. This I highly
>> doubt. I handpicked that version from
>> https://deb.debian.org/debian/pool/main/l/ , and installed the 6.1.66-1
>> packages[1]. That was totally messed up and made my amd64 system highly
>> unresponsive and erratic to the point it would hang shutting down. So I
>> booted back into 6.1.0-13, which still works fine, and purged 6.1.0-14
>> and 6.1.0-15, and IOT have two working options, I reinstalled 6.0.1-12,
>> which had been autoremoved with the update.
>>
>> [1] Packages I used:
>>
>> [src:linux]
>> linux-compiler-gcc-12-x86_6.1.55-1_amd64.deb
>> linux-headers-6.1.0-13-amd64_6.1.55-1_amd64.deb
>> linux-headers-6.1.0-13-common_6.1.55-1_all.deb
>> linux-kbuild-6.1_6.1.55-1_amd64.deb
>> linux-libc-dev_6.1.55-1_amd64.deb
>>
>> [src:linux-signed-amd64]
>> linux-headers-amd64_6.1.55-1_amd64.deb (optional meta-pkg)
>> linux-image-6.1.0-13-amd64_6.1.55-1_amd64.deb
>> linux-image-amd64_6.1.55-1_amd64.deb (optional meta-pkg)
>>
>> Other versions I tried:
>> 6.1.0-12/6.1.52-1 works
>> 6.1.0-13/6.1.55-1 works
>> 6.1.0-14/6.1.64-1 broken, according to #1057843
>> 6.1.0-15/6.1.66-1 broken as experienced
>>
>> For now, I've purged the broken ones, and put the working ones on hold
>> with dpkg.
>>
>> Any ideas?
>> -- 
>> Kevin Price
>>
> 
> If you hand-picked packages: I would suggest using apt to upgrade the whole
> system .

Done that, now that 6.1.0-15/6.1.66-1 is available through apt.

> Kevin: Do you have any logs to show brokenness?

Which ones?

> For what it's worth,
> there were other updates and fixes besides just #1057843 in the new
> release.

Some of that must be the culprit, breaking my system.

What happens, at a first glance:

1. Bootup to the gdm greeter works, but there I get an unusal keyboard
layout. And everything takes longer.

2. There seems to be no network connectivity. (no WiFi icon)

3. Launching Firefox apparently does nothing.

4. Launching gnome-teminal works, but many commands just stall, such as
"sudo dmesg" or "ip a".

5. Shutting down takes ages, with systemd waiting for a bunch of
services to terminate, most of which seem to be network-related. After
much more that 10 min I used hard power-off.

I'm uncertain whether such an unstable system is even capable of going
through the reportbug script. Any ideas, which logfiles might be useful
to debug this issue, and whom to address this to, IOT to prevent similar
experiences from many more users?

Also what kind of testing could I usefully perform? Any kernel
parameters maybe?
-- 
Kevin Price



Re: 6.1.0-15/6.1.66-1 broken too?

2023-12-10 Thread Andrew M.A. Cater
On Sun, Dec 10, 2023 at 08:02:03PM +0100, Kevin Price wrote:
> Am 09.12.23 um 19:09 schrieb Dan Ritter:
> > The new kernel release is reported to contain an ext4 data
> > corruption bug. It's prudent not to upgrade, or if you have
> > started to upgrade, not to reboot, until a new kernel release
> > is prepared.
> 
> Thanks for your announcement. I'm running out of time to properly report
> a bug against 6.1.66-1.
> 
> #1057843 states that it's fixed with 6.1.0-15/6.1.66-1. This I highly
> doubt. I handpicked that version from
> https://deb.debian.org/debian/pool/main/l/ , and installed the 6.1.66-1
> packages[1]. That was totally messed up and made my amd64 system highly
> unresponsive and erratic to the point it would hang shutting down. So I
> booted back into 6.1.0-13, which still works fine, and purged 6.1.0-14
> and 6.1.0-15, and IOT have two working options, I reinstalled 6.0.1-12,
> which had been autoremoved with the update.
> 
> [1] Packages I used:
> 
> [src:linux]
> linux-compiler-gcc-12-x86_6.1.55-1_amd64.deb
> linux-headers-6.1.0-13-amd64_6.1.55-1_amd64.deb
> linux-headers-6.1.0-13-common_6.1.55-1_all.deb
> linux-kbuild-6.1_6.1.55-1_amd64.deb
> linux-libc-dev_6.1.55-1_amd64.deb
> 
> [src:linux-signed-amd64]
> linux-headers-amd64_6.1.55-1_amd64.deb (optional meta-pkg)
> linux-image-6.1.0-13-amd64_6.1.55-1_amd64.deb
> linux-image-amd64_6.1.55-1_amd64.deb (optional meta-pkg)
> 
> Other versions I tried:
> 6.1.0-12/6.1.52-1 works
> 6.1.0-13/6.1.55-1 works
> 6.1.0-14/6.1.64-1 broken, according to #1057843
> 6.1.0-15/6.1.66-1 broken as experienced
> 
> For now, I've purged the broken ones, and put the working ones on hold
> with dpkg.
> 
> Any ideas?
> -- 
> Kevin Price
>

If you hand-picked packages: I would suggest using apt to upgrade the whole
system . 

For anybody who followed my initially erroneous advice to remove
linux-image-amd64 - installing that package should pull in 6.1.0-15
correctly as a dependency.

Kevin: Do you have any logs to show brokenness? For what it's worth,
there were other updates and fixes besides just #1057843 in the new
release.

With every good wish, as ever,

Andy
[amaca...@debian.org]



6.1.0-15/6.1.66-1 broken too?

2023-12-10 Thread Kevin Price
Am 09.12.23 um 19:09 schrieb Dan Ritter:
> The new kernel release is reported to contain an ext4 data
> corruption bug. It's prudent not to upgrade, or if you have
> started to upgrade, not to reboot, until a new kernel release
> is prepared.

Thanks for your announcement. I'm running out of time to properly report
a bug against 6.1.66-1.

#1057843 states that it's fixed with 6.1.0-15/6.1.66-1. This I highly
doubt. I handpicked that version from
https://deb.debian.org/debian/pool/main/l/ , and installed the 6.1.66-1
packages[1]. That was totally messed up and made my amd64 system highly
unresponsive and erratic to the point it would hang shutting down. So I
booted back into 6.1.0-13, which still works fine, and purged 6.1.0-14
and 6.1.0-15, and IOT have two working options, I reinstalled 6.0.1-12,
which had been autoremoved with the update.

[1] Packages I used:

[src:linux]
linux-compiler-gcc-12-x86_6.1.55-1_amd64.deb
linux-headers-6.1.0-13-amd64_6.1.55-1_amd64.deb
linux-headers-6.1.0-13-common_6.1.55-1_all.deb
linux-kbuild-6.1_6.1.55-1_amd64.deb
linux-libc-dev_6.1.55-1_amd64.deb

[src:linux-signed-amd64]
linux-headers-amd64_6.1.55-1_amd64.deb (optional meta-pkg)
linux-image-6.1.0-13-amd64_6.1.55-1_amd64.deb
linux-image-amd64_6.1.55-1_amd64.deb (optional meta-pkg)

Other versions I tried:
6.1.0-12/6.1.52-1 works
6.1.0-13/6.1.55-1 works
6.1.0-14/6.1.64-1 broken, according to #1057843
6.1.0-15/6.1.66-1 broken as experienced

For now, I've purged the broken ones, and put the working ones on hold
with dpkg.

Any ideas?
-- 
Kevin Price