LinuxQuestions.org - [SOLVED] Upgrade CentOS6.6 to 6.9 from new media on USB

- CentOS (https://www.linuxquestions.org/questions/centos-111/)

- - Upgrade CentOS6.6 to 6.9 from new media on USB (https://www.linuxquestions.org/questions/centos-111/upgrade-centos6-6-to-6-9-from-new-media-on-usb-4175615679/)

Upgrade CentOS6.6 to 6.9 from new media on USB

Hi all, hope everyone is OK. (Me not so much, mom is having serious heart problems :( :( )

I've tried this thing with upgrading a Linux installation many times but I could never figure out how or if I was doing something wrong so I always reinstalled it.
This time I NEED an upgrade from an old CentOS 6.6 system to the latest, 6.9.
I tried the usual YUM-thing but that made my system go kernel panic on me so I figured that I'll AGAIN try this feature to upgrade from media, but the exact same this happens.
This time I'm gonna ask what to do!

When I get to the screen asking me what type of installation I'd like to do, what should I answer there? I figured that "Replacing Existing Linux System(s)" would be the one, as I always tried, but NOPE. Answer is "Could not find enough free space for automatic partitioning, please use another method".

What installation type should I use to UPGRADE the system, not to REPLACE it, cos that I can do as I always used to do. I want to make the system to start working again with CentOS 6.9 after the failed 'yum update' from v6.6

There are applications installed that can't be reinstalled so.

What is wrong in my head here?

Please post fdisk -l (-> Automatic partioning).
Can you boot the old kernel?
Can you discribe the update process from Centos 6.6 to 6.9 in detail?

Start the old kernel, yes which is: 2.6.32-504.16.2-el6.x86_64

fdisk -l lists this:
Cannot open /dev/sdc
Cannot open /dev/sdb
Cannot open /dev/sda
Cannot open /dev/md126
Cannot open /dev/mapper/luks-2e......
Cannot open /dev/mapper/luks-cf......
Cannot open /dev/mapper/luks-8c......
Cannot open /dev/mapper/luks-c1......
Cannot open /dev/mapper/luks-38......
Cannot open /dev/mapper/luks-5d......

The upgrade was as this, first tried with yum as:
yum clean all
yum update

restarted and then I got kernel panic.

Tried to start it again now and I see this:
dracut Warning: No root device "block:/dev/disk/by-uuid/cb1b2......" found
dracut Warning: crypto LUKS UUID 381c...... not found
dracut Warning: Boot has failed. To debug this issue add "rdshell" to the kernel line.
...
...
Kernel panic - not syncing: Attempted to kill init!

so no root device, no wonder it starts to panic :P

I'll try the 'rdshell' thing and see what that gives me.
I'll be back...

I guess it has something to do with the encrypted partitions but I thought that it would handle that :(

Can you post grub.cfg? (-> dracut Warning: No root device "block:/dev/disk/by-uuid/cb1b2......" found)

And this error should not occur at all:

Cannot open /dev/sdc
Cannot open /dev/sdb
Cannot open /dev/sda

Code:

#boot=/dev/md127

default=0

timeout=5

splashimage=(hd0,0)/grub/splash.xpm.gz

hiddenmenu

title CentOS (2.6.32-696.13.2.el6.x86_64)

        root (hd0,0)

        kernel /vmlinuz-2.6.32-696.13.2.el6.x86_64 ro root=UUID=cb1b237a-c2ef-4106-9b45-4c923be6bceb rd_MD_UUID=82edd843:2cfa32f2:dd799763:4908a7f5  KEYBOARDTYPE=pc KEYTABLE=sv-latin1 SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_LUKS_UUID=luks-381c31cd-9788-4719-be7f-f8daa4316036 rd_MD_UUID=f882865d:d1f00a5b:22a0bc38:a440c5c6 rd_NO_LVM rd_NO_DM LANG=en_US.UTF-8

        initrd /initramfs-2.6.32-696.13.2.el6.x86_64.img

title CentOS (2.6.32-504.16.2.el6.x86_64)

        root (hd0,0)

        kernel /vmlinuz-2.6.32-504.16.2.el6.x86_64 ro root=UUID=cb1b237a-c2ef-4106-9b45-4c923be6bceb rd_MD_UUID=82edd843:2cfa32f2:dd799763:4908a7f5  KEYBOARDTYPE=pc KEYTABLE=sv-latin1 SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_LUKS_UUID=luks-381c31cd-9788-4719-be7f-f8daa4316036 rd_MD_UUID=f882865d:d1f00a5b:22a0bc38:a440c5c6 rd_NO_LVM rd_NO_DM LANG=en_US.UTF-8

        initrd /initramfs-2.6.32-504.16.2.el6.x86_64.img

title CentOS (2.6.32-504.12.2.el6.x86_64)

        root (hd0,0)

        kernel /vmlinuz-2.6.32-504.12.2.el6.x86_64 ro root=UUID=cb1b237a-c2ef-4106-9b45-4c923be6bceb rd_MD_UUID=82edd843:2cfa32f2:dd799763:4908a7f5  KEYBOARDTYPE=pc KEYTABLE=sv-latin1 SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_LUKS_UUID=luks-381c31cd-9788-4719-be7f-f8daa4316036 rd_MD_UUID=f882865d:d1f00a5b:22a0bc38:a440c5c6 rd_NO_LVM rd_NO_DM LANG=en_US.UTF-8

        initrd /initramfs-2.6.32-504.12.2.el6.x86_64.img

title CentOS 6 (2.6.32-504.el6.x86_64)

        root (hd0,0)

        kernel /vmlinuz-2.6.32-504.el6.x86_64 ro root=UUID=cb1b237a-c2ef-4106-9b45-4c923be6bceb rd_MD_UUID=82edd843:2cfa32f2:dd799763:4908a7f5  KEYBOARDTYPE=pc KEYTABLE=sv-latin1 SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_LUKS_UUID=luks-381c31cd-9788-4719-be7f-f8daa4316036 rd_MD_UUID=f882865d:d1f00a5b:22a0bc38:a440c5c6 rd_NO_LVM rd_NO_DM LANG=en_US.UTF-8

        initrd /initramfs-2.6.32-504.el6.x86_64.img

it also won't ask for the password for LUKS as it does when I start 2.6.32-504

Does bash (and commands) work correctly?

in the 2.6.32-504 version yes
probably not in 2.6.32-696 since I can't even boot into it :P

I assume, your server has been hacked.

You should reinstall with Centos version 6.9.

well, that can't be the case since it's internal only and it worked perfect with the 6.6 version.
I thought it would be smart to get the latest version so I tried to upgrade it and that made this error.

but the main question was how to use a DVD to try to upgrade the installation.
The selection "Replacing Existing Linux System(s)" seems right but it I only get: "Could not find enough free space for automatic partitioning, please use another method".
So how do I do this?

Then use gparted to investigate and clean the partition table.
https://gparted.sourceforge.io/download.php

If I do I'll wipe the installation like I would do if I made a clean installation and that is what I don't want to do.

Quote:

Originally Posted by Basher52 (Post 5770333)

If I do I'll wipe the installation like I would do if I made a clean installation and that is what I don't want to do.

Honestly, you really SHOULD do that...why upgrade from 6.6 to 6.9, when the latest is 7.4? If you're upgrading anyway, you really should get the latest-and-greatest, and at least put off (for a while), the inevitable EOL you're going to hit with the 6.x versions. Since you're already planning downtime and have things ready to go, a fresh install would get you the cleanest results. This will also give you the added benefit of testing your backup/recovery plans, because unless you can RESTORE your backups, they're useless.

And if this is a on a single-disk system, that's even easier. Spend the $59 or so and pop for a new drive, along with another $20 for a USB-to-SATA enclosure for the old drive. Do your install, and you just have to copy files from old drive to new. And if things go REALLY bad, you just swap drives and you're back up in a few minutes.

OK then, I'll do that, to do a clean install of 6.9 I mean.
Can't go to 7 though since I got some hardware that only work with kernel up to 2.6.32

Quote:

Originally Posted by Basher52 (Post 5770398)

OK then, I'll do that, to do a clean install of 6.9 I mean. Can't go to 7 though since I got some hardware that only work with kernel up to 6.32

Which hardware is that?? Because it seems unlikely, because hardware is typically better supported as it gets older, while the 'bleeding edge' hardware may have problems. What device(s) are you working with? And have you *TRIED* the latest CentOS 7.4 to see if it will function? You WILL be in a situation where even 6.9 won't be supported and/or won't work on whatever new hardware you have to buy. Things WILL DIE in the future, so clinging to an old OS/hardware is just going to bring you more bad times in the future. Worse, because you won't have ANY time to test things, but will be in disaster-recovery mode.

Take your time now, do it right, and work through your problems.