Hi to everyone. I'm new to the forum and building a server at home. ive very little experience so hoped some on here could advice. Ive been told elswhere that my best bet would be to buy a tower server. As i am only dipping my toe in the water, i dont want to spend vast amounts at this stage. What I am looking for is a server to run proxmox and create vm's with pci/gpu passthrough. Ive seen a Dell t110 ii, but ive seen there are problems finding compatible gpu's. What would anyone advice for my needs? I dont want a rack server as they are too noisy for where it needs to go so hopefully a reasonably quite tower would be good. Any suggestions??
Posted by: Jimez - 07-22-2021, 05:17 PM - Forum: General
- No Replies
Hello, can someone please help me identify this hardware with no brand or identifying markings? The back just has a couple of fans. Please and thank you!!
on my home server I have replaced the motherboard from a Gigabyte B450 to an Asrock X470d4u AM4 Server board. But after replacing the board the LSI MegaRAID SAS 9261-8i is not working on the new board. I can not access the array mangement and the stoage is also not reconised by the system?! Any help or idea...
I have a question, we have a Hyper-V host (Windows Server 2016) that hosts several virtual machines. The host is well utilized (CPU) at peak times, I'm just wondering one thing, the CPU load of the host never exceeds 75% CPU utilization. Now I wonder if this value is somehow limited by a rule?
I know that the host always needs computing power itself and that you shouldn't overbook it, that just aroused my curiosity and I would like to know that. I'm not talking about the vCPU limitation of each virtual machine.
I'm testing a scenario and reaching my limits somewhere. I am currently playing with a VM with two virtual hard disks of 35GB each. Here I installed Windows Server 2019 Standard. I have set up a mirror for hard disk C via the disk management. The system automatically converted both disks into dynamic disks. If I now remove the main hard disk during operation (boot partition), the system first runs as usual and then reboots. Here I didn't manage to get the system up and running again with the Windows Server installation ISO. What do you have to do to run Software Raid 1 to get the system up and running again after a crash? Is it possible to create a recovery boot image etc.?
I know that raid with a raid card is the more reasonable choice, but I would like to do that with the software raid and hope someone can help me here.
I have a Hyper-V server at home with about 21 VMs including a web server. My problem is that I don't have a fixed IP at home. My question is, can you get a fixed IP address through a VPN tunnel at home via my Windows server in the data center and forward the web server traffic or other services arriving in the data center to the VMs in the home network? If so, could someone explain to me in more detail how this could work?
My setup:
Fritzbox 7490 via fiber optic into the Internet 1GB Down 500Mbit Upload
Hyper-V Server directly connected to Fritzbox 7490 via Lan and gets its IPs for the VMs from there.
Fritzbox 7390 is still lying around here, if necessary I can use it to build a tunnel in RZ.
it is sometimes desperate and apparently there is no one in the world who has even one meaningful piece of advice.
For some weeks now we have been fighting on our managed servers at Strato with a high CPU load and load. But not from now on, it's gradually building up over days and weeks.
Strato's support refers to faulty scripts, but can't tell me how to find them either. Data protection reasons. Now I'm hoping to find the culprit here.
On the webspace only a Wordpress installation and a few plugins are running, but even switching them off and switching to the standard WP template doesn't help.
I have access to the Access Log, Error Log, and the SSH console with some commands like Top. Also, the Kill All command doesn't help there, CPU immediately goes up again.
How exactly do you proceed to find the error? I am grateful for any advice!
Compared to RAID-1, RAID-5 has the advantage that more usable net storage space is created from the physical gross storage space of the disks.
Some time ago someone told me, however, that it is much easier to recover data from a defective RAID-1 due to the complete redundancy (i.e. in much the same way as from a hard disk without RAID) than from a RAID-5, where the data is distributed over three hard disks. I am thinking in particular of cases where the whole RAID system is broken. E.g. in case of a water damage it happens regularly that the RAID controller first writes randomly confused stuff on the disks, until it (finally) chews completely.
Is it true that it is easier to recover data from a DEFECT RAID-1 array than from a DEFECT RAID-5 array?
