Bug 2022191

Additional info:
Also, after having tried out the metrics rhel-system-role on RHEL 9 system, insights-client started getting continuously spun up and consumes considerable CPU time.

Reproducer:
```
1. Install RHEL 9 BETA
2. Install system roles and ansible:
dnf install ansible-core ansible-pcp rhel-system-roles 

3. Create playbook:
cat << 'EOF' >playbook.yml
---
- hosts: the-system
  vars:
    metrics_graph_service: yes
  roles:
    - rhel-system-roles.metrics
EOF

4. Run playbook:
ansible-playbook ./playbook.yml
```

Ignore the additional info, it's unrelated.

After having registered RHEL 9 Beta to Insights, I experience performance issues on my host. CPU usage and network traffic rise up drastically.
I disabled all Insights related services and rebooted the VM, unfortunately the same occurs again, until I unregister the system from Insights.
Strange side effect is that even though the system isn't registered anymore, it re-appears after a restart on console.redhat.com again. Ideas ?

(In reply to Christian Labisch from comment #7)
> After having registered RHEL 9 Beta to Insights, I experience performance
> issues on my host. CPU usage and network traffic rise up drastically.
> I disabled all Insights related services and rebooted the VM, unfortunately
> the same occurs again, until I unregister the system from Insights.
> Strange side effect is that even though the system isn't registered anymore,
> it re-appears after a restart on console.redhat.com again. Ideas ?

You can mask the insights-client-results.service to stop this issue from happening anymore until a fix is released.

This issue is caused by this line in /usr/lib/systemd/system/insights-client-results.path

[Unit]
Description=Monitor @pkgsysconfdir@/.lastupload for modifications
Documentation=man:insights-client(8)
PartOf=insights-client.timer

[Path]
PathExists=@pkgsysconfdir@/.lastupload      # <<<
PathModified=@pkgsysconfdir@/.lastupload

[Install]
WantedBy=insights-client.timer

https://github.com/RedHatInsights/insights-client/blob/master/data/systemd/insights-client-results.path.in#L16

Because the file /etc/insights-client/.lastupload exists, the service is being triggered continuously.

Thanks @Link ! I have checked it and removed the .lastupload and the .last-upload.results files on January 29th.
Unfortunately that did not change the behavior ... the system re-appeared today after I had booted it yesterday.
I have removed it manually again today from console.redhat.com - below you you can see the facts of the system :

$ cat /usr/lib/systemd/system/insights-client-results.path
[Unit]
Description=Monitor /etc/insights-client/.lastupload for modifications
Documentation=man:insights-client(8)
PartOf=insights-client.timer

[Path]
PathExists=/etc/insights-client/.lastupload
PathModified=/etc/insights-client/.lastupload

[Install]
WantedBy=insights-client.timer

$ ls -la /etc/insights-client
total 1556
drwxr-xr-x.   2 root root    4096 29. Jan 16:04 .
drwxr-xr-x. 105 root root    8192 31. Jan 11:43 ..
-rw-------.   1 root root  145114 29. Jan 16:02 .cache.json
-rw-------.   1 root root     811 29. Jan 16:02 .cache.json.asc
-r--r--r--.   1 root root    4586  2. Mär 2021  cert-api.access.redhat.com.pem
-rw-r--r--.   1 root root     185  2. Mär 2021  .exp.sed
-rw-r--r--.   1 root root  143458 18. Okt 16:31 .fallback.json
-rw-r--r--.   1 root root     811 18. Okt 16:31 .fallback.json.asc
-rw-r--r--.   1 root root    1490 24. Aug 16:00 insights-client.conf
-rw-r--r--.   1 root root     151 20. Okt 16:34 insights-client.motd
-rw-r--r--.   1 root root      52 29. Jan 16:00 .insights-core.etag
-rw-r--r--.   1 root root      52 29. Jan 16:00 .insights-core-gpg-sig.etag
-rw-r--r--.   1 root root      36  2. Nov 19:17 machine-id
-r--r--r--.   1 root root    1147  2. Mär 2021  redhattools.pub.gpg
-rw-r--r--.   1 root root 1224762 18. Okt 16:31 rpm.egg
-rw-r--r--.   1 root root     811 18. Okt 16:31 rpm.egg.asc
-rw-r--r--.   1 root root      26 29. Jan 16:02 .unregistered

You need to remove line 16 from /usr/lib/systemd/system/insights-client-results.path and run `systemctl daemon-reload` in order to fix it. Removing /etc/insights-client/.lastupload will not change the behavior of the systemd unit file.

Another, better workaround would be to create an override unit file that redefines insights-client-results.path until we push a bug fix. Run `systemctl edit --full`, edit the text buffer by deleting line 16, and then save and quit.

Thank you @Link ! I'm gonna check it out ... more important though will be that the major issue (high CPU usage and network traffic) will get resolved when one registers a RHEL 9 system to Insights as soon as possible.

*** Bug 2027263 has been marked as a duplicate of this bug. ***

FYI @Link : I removed line 16 from /usr/lib/systemd/system/insights-client-results.path two days ago as you suggested - but still the system re-appears.
I have removed the system manually from console.redhat.com again yesterday and rebooted - well, today I found it being back on console.redhat.com again.

Removing line 16 from /usr/lib/systemd/system/insights-client-results.path will not remove the system from Insights Inventory. It only stops the insights-client-results.service unit from continuously restarting.

Thank you @Link ! Seems I misunderstood ... I have re-registered to Insights, and you are right - continuously restarting stopped. Thanks again Link ! :)

*** Bug 2062153 has been marked as a duplicate of this bug. ***

I just upgraded to RHEL 9.0 Beta 5 and it seems that the package insights-client 3.1.7-5.el9 has resolved the problem.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (new packages: insights-client), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2022:2645