Search results

I should mention the WAL is on an NVME

I have a 3 node cluster and I replaced 3x8TB drives and the recovery/rebalance is going to take 20+ days and maxes out at 10MiB which seems exceptionally slow. I've changed all relevant parameters, though setting them at a global level doesn't have any effect. It does change when set at the...

Nevermind, found a really good article that explains this and how to avoid it https://www.percona.com/blog/2019/08/02/out-of-memory-killer-or-savior/.

I have some LXC containers that are purposefully going to have high memory and CPU consumption and I DON'T want the container killed when this happens. How do I prevent those containers from being shot in the head but still prevent other containers from doing so?

Nice, I didn't know about that. Thank you. Having the extra isolation that VMs bring is a benefit as long as there aren't processor vulnerabilities to worry about.

I'm curious if there are any future plans to support MicroVMs in Proxmox? https://firecracker-microvm.github.io/ looks very promising for creating very low resource usage VMs (not containers) and is the technology backing AWS Fargate 1.4 without the Docker Engine as outlined in...

Looks like I'll be answering my own question. The problem is I was running my subnet managers on the Voltaire switches. https://docs.mellanox.com/pages/viewpage.action?pageId=12005023#SingleRootIOVirtualization(SR-IOV)-VirtualFunctionInfiniBandPorts has the subtle note "The SM on QDR switches...

I have a Mellanox ConnectX-3 card configured with VFs connected to Voltaire 4036's that are running my subnet manager. I have ib_ipoib loaded on the Proxmox host and can successfully assign an IP to the VF and ping my other IB hosts. When I pass the VF to a [un]priviliged LXC container with or...