Kernel warning when cloning a website

[timscha]

Hello!

I have installed a fresh Plesk Obsidian on a Debian 9.
I try to clone a website with the Wordpress toolbox, but random times I got a kernel warning like

Message from syslogd@plesk at Sep 26 11:19:56 ...
 kernel:[ 3075.069955] NMI watchdog: BUG: soft lockup - CPU#2 stuck for 41s! [sw-engine:10607]

Message from syslogd@plesk at Sep 26 11:27:42 ...
 kernel:[ 3540.934139] NMI watchdog: BUG: soft lockup - CPU#3 stuck for 40s! [sw-engine-fpm:10849]

Message from syslogd@plesk at Sep 26 11:27:42 ...
 kernel:[ 3540.934346] NMI watchdog: BUG: soft lockup - CPU#1 stuck for 37s! [mysqld:1555]

Message from syslogd@plesk at Sep 26 11:29:13 ...
 kernel:[ 3629.704344] NMI watchdog: BUG: soft lockup - CPU#1 stuck for 45s! [cron:10888]

Message from syslogd@plesk at Sep 26 11:29:13 ...
 kernel:[ 3629.704366] NMI watchdog: BUG: soft lockup - CPU#2 stuck for 44s! [mysqld:10902]

Also in general, it tooks years to clone a website.
I already tried to add a subdomain manually and than start the clone operation with the same result.
It's a normal Wordpress installation with a 90 mb database.

Mabye someone has an idea?

Tim

 

[IgorG]

Please tell me, is your server installed on some kind of virtualization platforms like VMware or something else?

 

[timscha]

Yes, using Hetzner Cloud (Truly thrifty cloud hosting - Hetzner Online GmbH, modul CX41), they are using KVM. Previous Plesk version was working fine there.

Edit: After killing the server I got this error in Plesk:

• Task is not responding: id=37, pid=10125, type=ext-wp-toolkit-backgroundtasksimplementation\cloning\cloningtask

 

[IgorG]

According to this KB article 1009996 one possible reason could be high levels of over-commitment (especially memory over-commitment) or other virtualization overheads. If you think that over-commitment is the reason, you can use the following:

# echo [time] > /proc/sys/kernel/watchdog_thresh

where time cannot be more than 60.

Source: CentOS 7 - kernel: BUG: soft lockup - CPU#3 stuck for 23s! [rcuos/1:19] - CentOS

 

[timscha]

Thanks for the informations. My hoster moved me to another system and now it is working fine again.

 

[seqoi]

Thanks for the informations. My hoster moved me to another system and now it is working fine again.

Hello. Since i also use CX41 at Hetzner for some of my projects i would like to know to what system they (Hetzner) switched you and from what exactly? From KVM to ?

I am sure other will also benefit from your response.

IgorG i am using Onyx at CX41 (KVM) on CentOS 7 and it's working perfectly. Why is Obsidian demanding something else? Or this is part of the unsolved upgrade issues?

 

[timscha]

Hello. Since i also use CX41 at Hetzner for some of my projects i would like to know to what system they (Hetzner) switched you and from what exactly? From KVM to ?

I am sure other will also benefit from your response.

IgorG i am using Onyx at CX41 (KVM) on CentOS 7 and it's working perfectly. Why is Obsidian demanding something else? Or this is part of the unsolved upgrade issues?

they moved my virtual Maschine from one to another host, I don‘t have more details.

 

[Ales]

IgorG i am using Onyx at CX41 (KVM) on CentOS 7 and it's working perfectly. Why is Obsidian demanding something else? Or this is part of the unsolved upgrade issues?

Obsidian requirements aren't that different, the issue in this case was caused by an overloaded server. The provider simply put too many customers on it.

 

[seqoi]

Obsidian requirements aren't that different, the issue in this case was caused by an overloaded server. The provider simply put too many customers on it.

Thank you.

 

[KITYFA]

• Task is not responding: id=37, pid=10125, type=ext-wp-toolkit-backgroundtasksimplementation\cloning\cloningtask