Search results

I typically get a similar 10% boost too. In any event, I’d be curious if someone intimate with this sort of thing could weigh in as at least on the surface having the same data in ARC and a VM’s page cache, especially at scale, would be a waste.

I found an academic paper on this double cache issue. Their argument was KSM partially solves this as if block XYZ is cached in ARC and cached in the VM’s RAM/page cache, it should be swept up as a duplicate by KSM. I think your approach is right and what I’ve always done, but this occurred to...

It occurs to me that I may be wasting RAM due the fact that there are two caches at play. There is the ARC at the Proxmox layer. Then, my VMs have a kernel buffer/page cache caching reads for whatever file system they use … ext4, XFS, etc. Does anyone disable the page cache within their VMs...

Upgrading to kernel 5.15 fixed my Asus PN50 on a Ryzen 4800U should anyone be Googling and find this thread.

Just as the title says, I rebooted today to apply a new kernel and my box gets stuck forever during boot at the “Import ZFS pools by cache file” step. I run ZFS for my boot disk too. I have seen a few articles on how to fix the issue - assuming I can get into the OS. Is there a way to “safe...

It seems the Proxmox servers are down. This seems to be a "not you" issue as I can't get to download.proxmox.com for weekend patching. :-/

Yes, this is the host. I didn’t know if USB ran the hubs in parallel or they were physically separate. In copious digging, USB hubs are not linked - which may seem obvious to everyone but me. I think this is a hardware/physical issue and not OS at this point. Appreciate the reply - but I’m off...

No matter what I do can't get my drives working at USB3 speeds. Everything looks right, so I am hoping some other eyes can give me some hints. # lsusb Bus 004 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub Bus 004 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root...

That’s what I was after. No fsync() means async in my book. :) I suspect a SLOG won’t help at all then. I’ll try disabling the ZFS property via /sys and see if it changes anything, but yeah, I think this disk is just not up to the task and it’s not a double write ZIL issue. Thanks!

Yes indeed, I am talking about the server side. You lost me though. leads me to believe writes are async by default? Are you saying that writes are synchronous or no as then you say which makes me think writes ARE synchronous? And how would I enable/disable synchronous writes? Thanks!

I am troubleshooting a disk that has realtively horrible performace and I am curious... does Proxmox Backup write using sync or async? As if it's sync, I probably need a separate ZIL or to forget about ZFS. If async, well... I keep digging.

I figured it out. I booted into an older kernel, deleted /etc/modprobe.d/zfs.conf, and then ran 'apt reinstall pve-kernelXYZ' to regenerate the initramfs. I'm back in business and slightly smarter.

I am not a smart person. I tweaked ZFS via /etc/modprobe.d/zfs.conf (specifically options zfs l2arc_noprefetch=0), ran 'update-initramfs -u' and I must have screwed something up in zfs.conf. The root file system is ZFS and now the box won't boot. Is there a way to fix this and ignore my fat...