New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Yeah. You go to their website and hit F5 a few dozen times a day.
Yeah. You go to their website and hit F5 a few dozen times a day.
Or hang around in LET. The best deals are always here.
Just random thought - may be this issues are caused by too new kernel on too old hardware?
Running something like centos 6 just to check may be worth a try....
Because mine is running since ~november 2016 without any issues apart from this HDDs being old and slow. With centos 7.
You do realize the rather ignorant ideology of this, right? Linux is not Windows. It'll still run happily on a Pentium Pro, given you have enough RAM. The BIOS is a little weird, and this is from a time when AMD really fucking sucked (IOMMU is very, very buggy), but all you need to do is know how to setup the BIOS and hardware, accordingly.
Running an ancient distribution just becase "it works" is NOT going to do the internet, nor your administrative abilities any good.
@WSS like where the fuck do you find any kind of option to set the AGP aperture in the bios? Theres shit all in my northbridge settings. Trying various kernel settings dont seem to help shit either.
NorthBridge->IOMMU->64MB. "Porblem solved".
setting iommu=soft in your kernel load should make it more stable, but you're going to lose VMX, which basically means no KVM hosting. amd_iommu=off will certainly disable VMX, but I haven't tested that. I've just got a neutered system with dying disks, soooo..
@WSS Im not quite that retarded (I hope):
Edit: Already tried iommu=soft. didnt help. Havent tried turning iommu off completely, cause id rather stay on debian 8 then.
Any news?
They are replacing my disk as of now.
I have Debian 9 stable- it's just my softraid that isn't. Under Memory config, AGP should be under that. First option.
I rebooted after I had realized they took back the KVM, so now I'm waiting in line for a KVM to figure out why it didn't come back up.
I do, however i still think that trying will not hurt, at least to determine if it is hardware or some sort of compatibility issue.
And yes, in theory linux should run on any old hardware. In practice, however, issues still occur. Removed support/drivers for older hardware is not uncommon, and theoretically some old hardware-specific things like some cpu/chipset/bios bug workarounds could have been removed from kernel too.
Whats the issue with your softraid?
I assume you installed your own ISO created the mds etc?
Disk issues too?
Had checked in there already:
They are still working on it. Will update once fixed.
I have asked for KVM time extension twice due to these issues...
Yup. Hard drive with 6 years spin time and known corrupt firmware that makes the controller shit itself.
We're talking about Linux. The only stuff ever removed from the kernel are things that shouldn't be.
I guess... No one got a fully working server yet?
I did! Old drives but with no issues in smart.
After getting another dead KVM, I got one that works, but it doesn't send anything to the host, so I can stare at the pre-boot BIOS screen, but that's it.
This is like fucking OneProvider all over again.
I'm glad I didn't click the order button.
Yeah reboot doesnt work here as well. Just run a pxe install but cancel it in the next second. It will reboot, then press F11 to boot from virtual drive.
I was having so many issues with the stupid KVM that I let it do a PXE install, then DD'd my ISO over /dev/sda and installed that way. Yet another "Hope I never have a problem" setup with this beast.
I've had my second drive replacement completed now, will take a look at what's been done later. Also going to ask what the mysterious 'issue' with the drives is....
High mileage and bad firmware with Seagates in mine.
Y'see, that's what scares me - I've still got at least two of these aged 750GB drives and from the power on hours they were probably all in then same machine so I fear for their sanity.
Heh.
Lots of IOMMU errors. There's no such setting in the BIOS.
As per their reply:
Also checking online, this particular BERT error also appears to be something related to Kernel 4.8+
**Please note that we do not offer Debian 9 via our automated system and, as such, it has not been tested for compatibility on any of our server models. That OS appears to be provided with kernel 4.9 by default. If you would like, we can perform an installation of an older OS version, such as Debian 8 or Ubuntu 14, to test and confirm if this is indeed a compatibility error. Doing so would, of course, remove all existing data from the server.
Nah, I really want to get this running Debian 9 fine. It wasn't working with CentOS 7 either and I already got Proxmox 5 installed.
Probably going to ask for a refund on this one. I was sitting at it for roughly 6 hours yesterday and Im still not any closer to having a fix for the mysterious drive problems under any linux kernel >4.
The same problem happens on 4.4 as well btw @MikePT.
I installed Debian 8. All works fine. Install proxmox 4.4 on top (and its 4.4 kernel) and all goes to shit. Booting the older 3.10 kernel makes it go smooth again.
Alrighty then. I am currently running kernel 4.9 and debian 9. The only issue I have is that bootup seems to take like 8-12 minutes.
The memory value is set to 128mb on my bios, but I haven't touched any of the settings. I dont have kvm currently set up, but should I ask for it to write down all the settings?
I installed Debian 8 with kvm and then upgraded it to Debian 9. It didnt boot at first, but I dont recall doing anything else than generating new initramfs for my raids.
Found a fix for this:
Sources: https://communities.vmware.com/thread/34340 / http://whiteboard.ping.se/Linux/IOMMU
Add to your kernel line: iommu=force,memaper=3
No errors now. All gone.
For those running Debian 9 + Proxmox 5, here you go lazy assholes:
linux /boot/vmlinuz-4.10.17-1-pve root=UUID=52bb0696-7caa-4fc3-8a40-a55c45d321f1 ro quiet iommu=force,memaper=3
echo 'Loading initial ramdisk ...'
initrd /boot/initrd.img-4.10.17-1-pve