Been fighting an issue since trying to bring up a new instance of eve-ng at home. The issue occurs in the Community edition 5.0.1-13 in the following installations:
Code: Select all
1. Dell R620 server, E5-2660 8core 256GB ram 2x300gb SAS raid0
a. Bare metal install
b. ESXI 8 free edition licensed Install
1. given 8 cores and 128GB
c. Proxmox 7.2 (latest? as of 2 weeks ago)
1. given 12 cores and 128GB
2. SuperMicro 1u server SYS5018D-FN8T Xeon-D 1518, 32GB ram 4x300GB SATA/NVME
a. ESXI 7.03 free edition licensed install
1. given 4 cores and 16GB ram
SSH stays connected throughout the issues, and I see panics in the syslog - however the system stays active. On the remote VNC side though, the video crashes - and sometimes, the subsystem restarts and prompts me with a login screen (put in password). I can do so, but on other crashes, the display subsystem just goes "gray" in VNC and doesn't respond.
I've also seen graphical glitches as I move the mouse around (almost seeming like the video memory is getting corrupted somehow?)
What works correctly: if I use either a bare metal Eve-NG pro (5.0.1-20) or I use my older setup - vmware workstation player 17 free on an AMD system (community edition 2.0.3-112 with an older 4.20.17 linux kernel). I do the exact same process as above and do NOT experience the crashes with the exact same ISO files (md5 verified etc.)
It appears that there are fixes (?) to QEMU in 5.0.1-20 that need brought forward to the community version. Does anyone have any thoughts on this - and potentially how I can patch this issue so I can move forward on my home lab environment without this issue?