Guru Meditation

[Jamie_Lambertstock]

Hello

I'm having a few problems trying to get VirtualBox to work on my Fedora 24 host. I have two servers, both running Fedora, which one which runs faultlessly, while the other keeps giving this Guru error.

Since one server is working, I'm guessing it isn't the install, although maybe something hardware related as they are both different models of Dell's poweredge series, which have different components.

I'll attach the full error log, although here is the point it reports a fault:

Code: Select all

00:00:22.635794 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
00:00:22.635806 !!
00:00:22.635808 !!                 Guru Meditation 1155 (VINF_EM_TRIPLE_FAULT)
00:00:22.635895 !!
00:00:22.636049 !! Skipping ring-0 registers and stack, rcErr=VINF_EM_TRIPLE_FAULT
00:00:22.636088 !!
00:00:22.636091 !! {mappings, <NULL>}
00:00:22.636095 !!

I'm very new to VirtualBox, so sorry if I'm being plain stupid.

EDIT #1: Since file size is to large, here is a pastebin of the log file:

www . pastebin . com /JJ1isA61

EDIT #2 - Guest was CentOS 7 although tried a few others, including Windows Server, not luck.

[mpack]

Extracts from a log are no good to us. Please post the complete log.

With the VM fully shut down, right click it in the GUI. Select "Show Log" and save "VBox.log" (ONLY) to a zip file. Attach the zip here. Compressed formats other than zip are ok too.

[Jamie_Lambertstock]

Thanks for the reply, attached is the compressed log as requested.

I posted a pastebin link above of the log, as it was to large to upload, completely forgot about compressing it - sorry!

[mpack]

I see that you still have the install ISO attached to the VM, and the guest is booting from it. Does that mean that you are still trying to install this guest OS? Or running it as a live CD (that would be odd for a server OS).

If yes, then do an MD5 checksum test on the ISO: triple faults are (IME) always the result of the guest doing something really nasty, such as executing illegal opcodes. A corrupted ISO could easily have that result.

[Jamie_Lambertstock]

Indeed it was , this was with a fresh install of both the guest (CentOS 7) and the host (Fedora 24). The installer failed just after the Test Media or Install selection.

I've also tried copying over a working guest servers, from our working VirtualBox host, which boot with the same error.

Checked the install media, no problems there sadly.

The plan was to make two Windows 2008 R2 guests on this server, which also give the error while trying to boot their install media.

[mpack]

Checked the install media, no problems there sadly.

Did you check the actual media being used by the VM, not copies held elsewhere? ISTM you have a suspect transport mechanism for getting ISOs onto that host.

[Jamie_Lambertstock]

Did you check the actual media being used by the VM, not copies held elsewhere? ISTM you have a suspect transport mechanism for getting ISOs onto that host.

I've just had a play, having no luck still.

I redownloaded the CentOS iso from their sites they suggest from here. (E.g. Here & Here)

Also transferred the ISO I used to install Cent on the other server, which worked fine at the time, no luck with that.

I still feel it could be hardware based, not 100% sure about the two Xeons in it. Would there be a way in which I could test to see if that are happy?

[mpack]

The CPU choice can't corrupt opcodes or cause any similar errors in the VM, so that doesn't sound likely to me.

The guest OS's you mention are so varied - and so well established as reliable guests in VirtualBox - that I'm struggling to see any common factor except the media.