Recent content by alexskysilk

this doesn't work like you think. In effect, this creates a loop which will either have one interface blocked by STP, or the whole switch would be shut down. In other words, this is a poor substitute for an active/passive bond. That is a more appropriate method. If LACP is available to you...

Most likely yes. It tooks me a second to read the above discussion when I realized they're no longer speaking in terms of your question. The answer to your question depends on other factors you didnt mention, namely: 1. are you trying to improve virtual machine performance? if not, bigger...

I wasnt even aware... oh well. it wasnt a particularly useful thing to do.

really? why is that? I havent tried this on pve9 but I dont see the network stack changing that dramatically...

One thing I can say with certainty- this is caused by something that is present on your system, which was not when the OS was initially installed. To verify, install the OS (PVE) from scratch and set up your zpool, and see if it "disappears."

your disks have been wiped. this doesnt happen on its own.

having iscsi links separated is IDEAL, and works in most instances. iscsi over LACP can, AT BEST, match the performance of seperated links, but in practice this can be quite challenging. Storage vendors rarely have any control over the end user network stack. As mentioned above, separate...

try not to filter unnecessarily; what you want to see if there are PARTITIONS on those drives. lsblk output will tell. if you have partitions, its just a matter of a race condition which could be solved. if there arent- you're doing stuff beyond what you are describing here. dont bother...

Sure. the problem isnt with the networking per se; its that you're shoving 3 vlans into two nics, 2 of which really want their own interface. this would be very simply solved if you just added another interface (or 2) for non iscsi traffic. That should probably be the first layer in your...

Understatement of the year :lol: this is really prone to MTU problems and uneven IO.

It will. I've used this kind of config in the lab before. like I said, not ideal for a production environment but can be done. edit- I'll explain the logic. nic1 has mac1. nic 2 has mac2 in the above config, either mac1 or 2 is present in V0; mac1 is present in v100, and mac2 is present in...

have less vms move. the cluster manager will not do all operations concurrently.

You can. it works. Its a bad idea, but it can be done. (as an aside, its a bad idea on vmware too; iscsi interfaces dont like to share the link.) the interfaces would look like this: iface nic0 inet manual mtu 9000 iface nic1 inet manual mtu 9000 auto bond0 iface bond 0 inet...

well, you have quite the loadout. Before we begin, make sure you have multipath-boot-tools installed. it is critical for the proper ordering of devices during boot. next, I assume the disks in question are the 600GB drives. assuming you can create a new zpool with striped mirrors: zpool...

We're looking at it from opposite perspectives. the word "prefer" tells the story; to facilitate high load, you will need to do this. The key remains that backup systems are essentially passive- all the work have to occur from the sender end; If you have no load to produce, why is "optimizing"...