cisco xrv9k-full keeps rebooting

[Sidney]

hi

my cisco ios xrv9k images keeps rebooting.
I tried 7.4.2, 7.6.2 and 7.8.2 both vrr and non-vrr. From using the cisco built .qcow2 file that you get from the .tar (Option 1:) file, to building my own .qcow2 (Option 2:)
(https://www.eve-ng.net/index.php/documentation/howtos/howto-add-cisco-xrv-9000/)

permissions was fixed, rebooted
and all 8 routers have different errors before the start rebooting.


Server info:
=====
Hypervisor: VMware ESXi, 7.0.3, 20036589
Model: UCSB-EX-M4-1
Processor Type: Intel(R) Xeon(R) CPU E7-4890 v2 @ 2.80GHz
Logical Processors: 120
NICs: 6
Virtual Machines: 7
State: Connected
Uptime: 36 days
=====



Also no new logs:
====
root@dumbledore:/opt/unetlab/data/Logs# ll
total 792
drwxr-sr-x 2 www-data www-data 4096 Apr 5 11:49 ./
drwxr-sr-x 4 www-data www-data 4096 Jul 8 2022 ../
-rw-r--r-- 1 www-data www-data 0 Apr 5 11:48 api.txt
-rw-r--r-- 1 www-data www-data 23432 Apr 5 17:45 php_errors.txt
-rw-r--r-- 1 www-data www-data 625661 Apr 5 18:26 ssl-access.log
-rw-r--r-- 1 www-data www-data 1061 Apr 5 17:07 ssl-error.log
-rw-r--r-- 1 www-data www-data 134266 Apr 5 17:45 unl_wrapper.txt
root@dumbledore:/opt/unetlab/data/Logs# date
Wed Apr 5 18:26:57 UTC 2023
root@dumbledore:/opt/unetlab/data/Logs#
====

[IceCat]

First, disable UKSM in the lab (GUI -> Lab Status) - it's common requirement to run xrv9k on all eve-ng versions and all qemu versions.

For eve-ng 5.x try non-default QEMU versions: 2.12.0, etc.

If eve-ng 5.x still works not stable, try latest eve-ng 2.x with default qemu 2.4.0. This combinations should give maximum stability.

UPDATE:

Cisco xrv9k on AMD 5950X CPU works stable on eve-ng 6.x (with newer Linux core) and default QEMU version for AMD platform (tpl-5.2.0). I successfully ran 10 xrv9k nodes, assigning 2 vCPU and 8192 MB RAM to each. But: UKSM (KSM) still needs to be disabled!