recent kernel

[faeychild]

I've stumbled into my own confusion here
I was trying to clean up old kernels via MCC and listed the installed
kernels



kernel-desktop-5.15.23-1.mga8-1-1.mga8
kernel-desktop-5.15.25-1.mga8-1-1.mga8
kernel-desktop-5.15.28-1.mga8-1-1.mga8
kernel-desktop-5.15.32-1.mga8-1-1.mga8
kernel-desktop-5.15.4-1.mga8-1-1.mga8
kernel-desktop-5.15.6-2.mga8-1-1.mga8

So I thought , remove the kernels 5.15.23 and 5.15.25 and leave the last
two 28 and 32

But a closer look shows 5.15.4 and 5.15.6
are they more recent?

uname -a
Linux unimatrix.cryptid 5.15.32-desktop-1.mga8 #1 SMP Mon Mar 28
08:31:19 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux

cat /proc/version
Linux version 5.15.32-desktop-1.mga8 (***@rabbit.mageia.org) (gcc
(Mageia 10.3.0-2.mga8) 10.3.0, GNU ld (GNU Binutils) 2.36.1) #1 SMP Mon
Mar 28 08:31:19 UTC 2022

I would seem to be running 5.15.32
and I believe that I'm confused by the kernel numbering system and the
sort order of MCC, but I'm not sure




regards

[faeychild]

On 11/4/22 15:13, faeychild wrote:

OK I found the change log. They date from late last year.

I can't imagine why they are still installed and never removed. It did
rattle my chain though

regards

[David W. Hodgins]

Post by faeychild
OK I found the change log. They date from late last year.
I can't imagine why they are still installed and never removed. It did
rattle my chain though

Old kernels and their associated packages such asvirtualbox-kernel are not
automatically removed in order to allow fallback if booting a newer kernel fails.

I use the attached script to remove old kernels.

Regards, Dave Hodgins

[Doug Laidlaw]

Post by faeychild
OK I found the change log. They date from late last year.
I can't imagine why they are still installed and never removed.  It did
rattle my chain though
regards

My experience with Kernel 5-15-4. has been similar. I am guessing that
it is installed by the OS installer, and for some reason, it doesn't
appear as "old."

[William Unruh]

Post by Doug Laidlaw

Post by faeychild
OK I found the change log. They date from late last year.
I can't imagine why they are still installed and never removed.  It did
rattle my chain though

They are not removed so that if your new kernel does not work, you are
not left up shit creek without a boat or paddle.
(Actually there is a creek in College Station Tx called Sewer Creek--
runs through the middle of the golf course.)

If you want to remove your old kernels, you can do so. Urpme is your
friend.

Post by Doug Laidlaw

Post by faeychild
regards

My experience with Kernel 5-15-4. has been similar. I am guessing that
it is installed by the OS installer, and for some reason, it doesn't
appear as "old."

Mageia does not remove "old" kernels. However it does put them onto the
orphans list so, if you are silly enough to remove orphans it will be
removed ( It is silly because if one of those "Package" rpms gets
orphaned its removal can destroy your whole system. And they do get
orphaned at times. )

[David W. Hodgins]

Post by faeychild
I've stumbled into my own confusion here
I was trying to clean up old kernels via MCC and listed the installed
kernels
kernel-desktop-5.15.23-1.mga8-1-1.mga8
kernel-desktop-5.15.25-1.mga8-1-1.mga8
kernel-desktop-5.15.28-1.mga8-1-1.mga8
kernel-desktop-5.15.32-1.mga8-1-1.mga8
kernel-desktop-5.15.4-1.mga8-1-1.mga8
kernel-desktop-5.15.6-2.mga8-1-1.mga8
So I thought , remove the kernels 5.15.23 and 5.15.25 and leave the last
two 28 and 32
But a closer look shows 5.15.4 and 5.15.6
are they more recent?
uname -a
Linux unimatrix.cryptid 5.15.32-desktop-1.mga8 #1 SMP Mon Mar 28
08:31:19 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux
cat /proc/version
(Mageia 10.3.0-2.mga8) 10.3.0, GNU ld (GNU Binutils) 2.36.1) #1 SMP Mon
Mar 28 08:31:19 UTC 2022
I would seem to be running 5.15.32
and I believe that I'm confused by the kernel numbering system and the
sort order of MCC, but I'm not sure

Use 'rpm -qa|grep kernel|sort -V", or "rpm -qa --last|grep kernel" which
will show the date installed, with most recent first.

Or check a mirror.
http://mirror.math.princeton.edu/pub/mageia/distrib/8/x86_64/media/core/updates/
shows ...
kernel-desktop-5.15.4-1.mga8-1-1.mga8.x86_64.rpm 2021-11-21 13:29 68M
kernel-desktop-5.15.32-1.mga8-1-1.mga8.x86_64.rpm 2022-03-28 09:35 68M

So .32 is newer than .4.

Regards, Dave Hodgins

[Jim]

I've stumbled into my own confusion here I was trying to clean up old
kernels via MCC and listed the installed kernels
kernel-desktop-5.15.23-1.mga8-1-1.mga8
kernel-desktop-5.15.25-1.mga8-1-1.mga8
kernel-desktop-5.15.28-1.mga8-1-1.mga8
kernel-desktop-5.15.32-1.mga8-1-1.mga8
kernel-desktop-5.15.4-1.mga8-1-1.mga8
kernel-desktop-5.15.6-2.mga8-1-1.mga8
So I thought , remove the kernels 5.15.23 and 5.15.25 and leave the last
two 28 and 32
But a closer look shows 5.15.4 and 5.15.6 are they more recent?

You have sorted the kernel names, but neglected the difference between
numerical sort and ascii sort.

Sorted in ascii fashion, 5.15.3... precedes 5.15.4...
Sorted in numerical fashion, 5.15.4... precedes 5.15.32...

Cheers!

jim b.

[faeychild]

Post by Jim
You have sorted the kernel names, but neglected the difference between
numerical sort and ascii sort.
Sorted in ascii fashion, 5.15.3... precedes 5.15.4...
Sorted in numerical fashion, 5.15.4... precedes 5.15.32...

Yes I've done all those things. What I find amusing is that in all the
previous kernel cleanups I managed to completely ignore the two early
kernels. They stand out clearly, obviously and yet I missed seeing them
on at least three occasions.

And despite plodding carefully through the list each time, this is a
stunning lack of observation unless I have some temporary dyslexia I an
unaware of.

I have saved David's script
He has beaten the Bit Twister to it :-)

regards

[Bit Twister]

Post by faeychild
I have saved David's script
He has beaten the Bit Twister to it :-)

I had seen David's script in previous posts somewhere.
My script did/does not query rpm like David's does.

My ck_4_kernel_rpms script checks for more than one kernel in /boot with
ls -1h /boot/initrd-*mga* | grep -v $(uname -r) | wc -l
and converts ls -1 initrd-*mga* | grep -v $(uname -r) release into
urpme kernel rpm name(s) and uses xmessage to pop up the list to be removed.

My install_updates scripts runs ck_4_kernel_rpms to tell me to remove the
extra kernels which I do after I have booted the new kernel and verify
I do not seem to have problems.

[William Unruh]

Post by faeychild

Post by Jim
You have sorted the kernel names, but neglected the difference between
numerical sort and ascii sort.
Sorted in ascii fashion, 5.15.3... precedes 5.15.4...
Sorted in numerical fashion, 5.15.4... precedes 5.15.32...

Yes I've done all those things. What I find amusing is that in all the
previous kernel cleanups I managed to completely ignore the two early
kernels. They stand out clearly, obviously and yet I missed seeing them
on at least three occasions.
And despite plodding carefully through the list each time, this is a
stunning lack of observation unless I have some temporary dyslexia I an
unaware of.
I have saved David's script
He has beaten the Bit Twister to it :-)

Note that you can also look at the date of installation of the file
(well the modification time) with
ls -t (newest first)
or ls -tr (oldest first)
instead of alphabetical.cd /boot

Post by faeychild
regards

[TJ]

Post by faeychild
I've stumbled into my own confusion here
I was trying to clean up old kernels via MCC and listed the installed
kernels
kernel-desktop-5.15.23-1.mga8-1-1.mga8
kernel-desktop-5.15.25-1.mga8-1-1.mga8
kernel-desktop-5.15.28-1.mga8-1-1.mga8
kernel-desktop-5.15.32-1.mga8-1-1.mga8
kernel-desktop-5.15.4-1.mga8-1-1.mga8
kernel-desktop-5.15.6-2.mga8-1-1.mga8
So I thought , remove the kernels 5.15.23 and 5.15.25 and leave the last
two 28 and 32
But a closer look shows 5.15.4 and 5.15.6
are they more recent?
uname -a
Linux unimatrix.cryptid 5.15.32-desktop-1.mga8 #1 SMP Mon Mar 28
08:31:19 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux
cat /proc/version
(Mageia 10.3.0-2.mga8) 10.3.0, GNU ld (GNU Binutils) 2.36.1) #1 SMP Mon
Mar 28 08:31:19 UTC 2022
I would seem to be running 5.15.32
and I believe that I'm confused by the kernel numbering system and the
sort order of MCC, but I'm not sure

This first time I remember running into this situation was back with
Mandriva. As I recall, there was about 2-3 seconds of confusion before I
realized the reason for the way the kernels were listed, but it had been
a long day and I was tired.

Actually, I had remembered installing and later updating the
later-listed kernels, so I knew they were older without looking them up.

TJ

[faeychild]

Post by TJ

Post by faeychild
I would seem to be running 5.15.32
and I believe that I'm confused by the kernel numbering system and the
sort order of MCC, but I'm not sure

This first time I remember running into this situation was back with
Mandriva. As I recall, there was about 2-3 seconds of confusion before I
realized the reason for the way the kernels were listed, but it had been
a long day and I was tired.

YES! And the mind half on something else thinking I'll just do this and
get back to... and then confrontation with the anomaly
and everything is sidetracked.

[William Unruh]

Post by faeychild

Post by TJ

Post by faeychild
I would seem to be running 5.15.32
and I believe that I'm confused by the kernel numbering system and the
sort order of MCC, but I'm not sure

This first time I remember running into this situation was back with
Mandriva. As I recall, there was about 2-3 seconds of confusion before I
realized the reason for the way the kernels were listed, but it had been
a long day and I was tired.

YES! And the mind half on something else thinking I'll just do this and
get back to... and then confrontation with the anomaly
and everything is sidetracked.

ls -t or ls -tr is your friend.