I and other users are still using CentOs 7 images. we have to update RoclyLinux.
we foind CO 7,x > Co7.9 > RL 8.x .
but RL8 to RL9 is failed . depend on kits or repository. Web site update , some times repository lost.
we has a DVD images , can boot and install RL9.2
Old way to update from DVD image , overide OLD CentOS images.
hope to keep /etc/groups /etc/pass … system setting.
re-install SQL or Other products are acceptable.
or Old Cent OS user setting migration way or scrips are helpfull for Old CentOS still using users.
regards!
Disclaimer, I personally install fresh rather than even consider in-place conversion to different distro (and RHEL 7, RHEL 8, and RHEL 9 are three distinct distros).
Second, AFAIK there is no in-place conversion that Rocky would officially support, (except the XX8 → RL8 sidegrade).
That said, there is ELevate project in: AlmaLinux OS - Forever-Free Enterprise-Grade Operating System that might be able to convert some systems.
The CentOS Linux 7.9 was released November 2020. A properly maintained system would have been “Co7.9” since then and also have all updates released for 7.9 after that. Same applies to Rocky; one does not just install and forget, one must diligently deploy updates when they are released.
The services that run in RL9 may or may not be same as in CL7. Even if service is same, the version is not and appropriate settings have changed. Copy-paste is not a clean solution.
There are now configuration management systems – Rocky has package ansible-core that provides Ansible. One can create groups and accounts with it to existing systems. (Even passwords …) Admittedly, taking such system into use does require some effort, but will save work in the long run.
I have never quite understood the concept of “image”.
Many CentOS 7 users have experienced deletion or suspension of repositories, KIT, and boot DVDs on mirror servers and even the source of CentOS due to EOL, and have taken the opportunity to take advantage of the fact that new security patches are no longer available. , we are in the process of migrating to RockyLinux.
The path from CentOS7 to 8 is already closed.
If I were to upgrade from CentOS7 to 8, I would choose to migrate to ROcky or alma.
“My goal” is CentOS7 to Rocky9. only Update OS.
looked at several sites and performed the steps to get Rocky8 with leapp. I stumbled on ROcky8 to 9.
I don’t mind redoing system settings on Linux or reinstalling Psql, MySQL, or other products.
All you need to do is move the PASSWORD and Group of the user using the server appropriately.
Moving user accounts is most important with SSH, as it does not rely on passwords and integrates with other systems.
The important point is moving ~.ssh/authorized_keys and shadowed accounts and groups.
By copying some files from CentOS to New Rocky linux 9
Is there any script or method that can easily solve this problem?
The below is an Ansible playbook that should ensure that target system has in groups
john:*:2001:
jane:*:2002:
and in users
john:x:2001:2001:John Doe:/home/john:/usr/sbin/bash
jane:x:2002:2002:Jane Doe:/home/jane:/usr/sbin/bash
The playbook, if in file users.yml would be deployed with
ansible-playbook users.yml
Note though that Ansible will use ssh to connect to destination(s), you do need to define inventory that lists the destinations (but could be just localhost), and if the connection is not as root, then account must be able to sudo (so it can run the ‘useradd’).
This play will not set any password nor create home directory (i.e. copy files from /etc/skel). The assumption is that you have mounted an existing filesystem to /home and it already has directories for 2001 and 2002 (and .ssh/authorized_keys within).
- hosts: all
become: true
vars:
local_users:
- name: john
uid: 2001
gecos: John Doe
- name: jane
uid: 2002
gecos: Jane Doe
tasks:
- name: Create private groups
ansible.builtin.group:
name: "{{ item.group | default(item.name) }}"
gid: "{{ item.gid | default(item.uid) | int }}"
loop: "{{ local_users }}"
loop_control:
label: "{{ item.group | default(item.name) }}"
- name: Create users
ansible.builtin.user:
create_home: false
name: "{{ item.name }}"
uid: "{{ item.uid | int }}"
group: "{{ item.group | default(item.name) }}"
comment: "{{ item.gecos | default(item.name) }}"
home: "{{ item.base | default('/home') }}/{{ item.name }}"
shell: "{{ item.shell | default('/usr/bin/bash') }}"
loop: "{{ local_users }}"
loop_control:
label: "{{ item.name }}"
Repositories for CentOS 7 do still exist and they do receive security patches. (The latest I see is new kernel on 2023-10-20.) The EOL of CentOS 7 is June 2024.
Thank you for your cooperation.
For passwords and groups,
Prepare a work folder,
Copy the group and passwd of Rocly working with CentOS in a distinguishable form.
cat group_co >> /etc/group
cat group_RL >> /etc/gruop
cat passwd_CO >> /etc/passwd
cat passwd_RL >> /etc/passwd
Remove duplicates using grock.
You will be asked if you want to delete the duplicates.
With pwck, you will be warned about deleting duplicates and if the Directory does not exist, so create it.
Refer to the created directory in CentOS and match the owner and group.
Restart the server and check network connectivity.
These check commands seem to have FiLO priority.
If you append in the wrong order, the network or service will not start up.
After this, we will move on to introducing Psql and HTTPD Rsync.
/minami
passwd and group shadow files also , may be works append .
iit’s your own risks. not enough test was done.
I can , RL9 not made users can logon CentOS’s user account/pass
regards!.
This seems dangerous to me if you are duplicating the system users and then deleting duplicates. Why? Because the UID’s and GID’s for the system users may be different from the two different systems as they are not all hard coded. How do you know which duplicate to delete?
which duplicate to delete?
This oepration doing RL system.
these file are text file , as you know .
grchk and pwck works delete or keep and integrity. whitch delete is your own risk.
before pwck , tou can seel Working system using UID and GID .
for examole , these 2 files root information writed in
RL root
CO root
RL root
data in passwd. files
pwck delete 1st find RL root and CO rool record keeps 2nd RL root (RL original) ,
This just symple teick. system Users UID and GID also shadow record keeps Originals.
is your own risk and keep Original file xxx_RO , after reboot , failed networks and other operation,
Console single logon , and replace Original files ,back to before merge
Enjoy !
Upgrades, regardless of the method, is unsupported. You are much better off setting up configuration management, backing up data, installing a new system, and restoring your backups where possible.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.