Solved How to Turn off automatic update of the operating system

[solidsnake]

I saw the following message in the daily logs. Do I need to turn off any service to prevent this from repeating?

Jul  9 22:50:50 testserver1 systemd[1]: Starting dnf makecache...
Jul  9 22:52:47 testserver1 systemd[1]: Starting Cleanup of Temporary Directories...
Jul  9 22:52:47 testserver1 systemd[1]: systemd-tmpfiles-clean.service: Succeeded.
Jul  9 22:52:47 testserver1 systemd[1]: Started Cleanup of Temporary Directories.
Jul  9 22:52:50 testserver1 dnf[7454]: Oracle Linux 8 BaseOS Latest (x86_64)           0.0  B/s |   0  B     02:00
Jul  9 22:52:50 testserver1 dnf[7454]: Errors during downloading metadata for repository 'ol8_baseos_latest':
Jul  9 22:52:50 testserver1 dnf[7454]:  - Curl error (28): Timeout was reached for https://yum.oracle.com/repo/OracleLinux/OL8/baseos/latest/x86_64/repodata/repomd.xml [Connection timed out after 120001 milliseconds]
Jul  9 22:52:50 testserver1 dnf[7454]: Error: Failed to download metadata for repo 'ol8_baseos_latest': Cannot download repomd.xml: Curl error (28): Timeout was reached for https://yum.oracle.com/repo/OracleLinux/OL8/baseos/latest/x86_64/repodata/repomd.xml [Connection timed out after 120001 milliseconds]

Are the following messages related to this process?

Jul  9 22:58:03 testserver1 systemd[1]: Starting Check PMIE instances are running...
Jul  9 22:58:03 testserver1 systemd[1]: Started Check PMIE instances are running.
Jul  9 22:58:03 testserver1 systemd[1]: pmie_check.service: Succeeded.
Jul  9 22:59:03 testserver1 systemd[1]: Starting Check and migrate non-primary pmie farm instances...
Jul  9 22:59:03 testserver1 systemd[1]: Started Check and migrate non-primary pmie farm instances.
Jul  9 22:59:03 testserver1 systemd[1]: pmie_farm_check.service: Succeeded.
Jul  9 23:00:03 testserver1 systemd[1]: Starting system activity accounting tool...
Jul  9 23:00:03 testserver1 systemd[1]: sysstat-collect.service: Succeeded.
Jul  9 23:00:03 testserver1 systemd[1]: Started system activity accounting tool.
Jul  9 23:10:03 testserver1 systemd[1]: Starting system activity accounting tool...
Jul  9 23:10:03 testserver1 systemd[1]: sysstat-collect.service: Succeeded.
Jul  9 23:10:03 testserver1 systemd[1]: Started system activity accounting tool.

 

[dos2unix]

First, it's a good idea to keep your systems updated.
You should be running a dnf update at least every month.

It looks like that may not be possible in your case, as it seems you can't reach some internet sites.
Typically dnf tries to do a cache update every day. This ensures that the latest versions are lsited
and available. You could disable this, but it doesn't hurt anything. The error you are getting
is simply saying I can't get to the internet. Which isn't an error, it's a true statement.

pmie_check is part of the performance co-pilot suite. Are you using PCP?
It's not directly related to the dnf errors, but it does rotate logs every so often.

If you really, really want to do this. Add this line to your /etc/dnf/dnf.conf file.

exclude=*

 

[solidsnake]

First, it's a good idea to keep your systems updated.
You should be running a dnf update at least every month.

It looks like that may not be possible in your case, as it seems you can't reach some internet sites.
Typically dnf tries to do a cache update every day. This ensures that the latest versions are lsited
and available. You could disable this, but it doesn't hurt anything. The error you are getting
is simply saying I can't get to the internet. Which isn't an error, it's a true statement.

pmie_check is part of the performance co-pilot suite. Are you using PCP?
It's not directly related to the dnf errors, but it does rotate logs every so often.

If you really, really want to do this. Add this line to your /etc/dnf/dnf.conf file.

exclude=*

Thanks

yes I want to do the updates whenever I want.

But I couldn't figure out how to cancel

 

[Condobloke]

yes I want to do the updates whenever I want.

This appears to be a growing trend.....and the incidence of OS failures grows with it

Not being critical of you, @solidsnake, but the trend I speak of comes from ex windows users, who have developed this habit over the years to avoid 'windows crazy updates', or whatever they call them now.

It really is unnecessary with Linux.

 

[bob466]

In Linux Mint Cinnamon...Automatic Updates are disabled by default and I don't enable it because of this...
https://easylinuxtipsproject.blogspot.com/p/mintupdate.html#ID5

I do have Automatic Maintenance Enabled as shown here...

Hope this helps.

 

[Aristarchus]

I don't know Oracle Linux, but it looks like the issue with the server you are trying to connect. You can try another server if option exist or ask about this on Oracle Linux forums.

I am running Linux for 26yrs, originally updates weren't automatic and all was working. I don't have automatic updates set now either: Slackware do not provide it, OpenIndiana (last when I was using it) does not provide it, neither do BSDs.
Automatic updates are not required, they are just a convenience (in normal, sane word).

I have seen enough posts about broken systems after automatic update.

What is needed is user taking full responsibility for the tools.

Just be careful and understand what you do.

 

[Condobloke]

Just to be REALLY clear

Software & Security Automatic Updates are NOT in place as the default

It is a CHOICE.......if you wish to turn them on, then you have the ability to do so.

There may be some Linux OS's that do have this 'feature' by default....if so I am not aware of them

 

[dos2unix]

The OS updates are not automatic, nothing gets installed unless you tell it to. ( there is an auto-update feature, but it's disabled by default ). However the cache update is automatic. Nothing gets installed, it just updates the package list with any new version or new packages. These don't get installed, they just show up in "What would be updated if you updated".
By putting the exclude line in the dnf.conf, it won't even update the cache.

 

[solidsnake]

The OS updates are not automatic, nothing gets installed unless you tell it to. ( there is an auto-update feature, but it's disabled by default ). However the cache update is automatic. Nothing gets installed, it just updates the package list with any new version or new packages. These don't get installed, they just show up in "What would be updated if you updated".
By putting the exclude line in the dnf.conf, it won't even update the cache.

I have critical applications on my server that are interconnected. when a package is updated, I need to investigate whether it will affect the application or not, or I need to consult the company.

So I wanted to do the updates myself whenever I want.

I didn't want /var/log/messages to be constantly

Let me consult the support page of the distribution I use

 

[GatorsFan]

Have a look here since you are using Oracle 8 - https://support.oracle.com/knowledge/Oracle Cloud/2522975_1.html
you will need an account to view it

 

[f33dm3bits]

Are the following messages related to this process?

Oracle Linux/RHEL Clones and RHEL don't have automatic updates enabled.

Jul 9 22:50:50 testserver1 systemd[1]: Starting dnf makecache... Jul 9 22:52:47 testserver1 systemd[1]: Starting Cleanup of Temporary Directories... Jul 9 22:52:47 testserver1 systemd[1]: systemd-tmpfiles-clean.service: Succeeded. Jul 9 22:52:47 testserver1 systemd[1]: Started Cleanup of Temporary Directories. Jul 9 22:52:50 testserver1 dnf[7454]: Oracle Linux 8 BaseOS Latest (x86_64) 0.0 B/s | 0 B 02:00 Jul 9 22:52:50 testserver1 dnf[7454]: Errors during downloading metadata for repository 'ol8_baseos_latest': Jul 9 22:52:50 testserver1 dnf[7454]: - Curl error (28): Timeout was reached for https://yum.oracle.com/repo/OracleLinux/OL8/baseos/latest/x86_64/repodata/repomd.xml [Connection timed out after 120001 milliseconds] Jul 9 22:52:50 testserver1 dnf[7454]: Error: Failed to download metadata for repo 'ol8_baseos_latest': Cannot download repomd.xml: Curl error (28): Timeout was reached for https://yum.oracle.com/repo/OracleLinux/OL8/baseos/latest/x86_64/repodata/repomd.xml [Connection timed out after 120001 milliseconds]

What you are seeing here is just the yum cache being refreshed, you still have to run "yum update" to actually install the updates.