Search results

Hi, I'm not sure about what I'm seeing there.. if you use LACP, make sure it's enabled on the switch though.. cat /proc/net/bonding/bond1 will tell you what pve sees. ip link / ip a wlil tell you more about your NICs, but configuration seems OK..

I don't like Discord, personally, but find it funny because, if people wanted to chat.. There's an unofficial IRC channel ##proxmox @ Libera Chat (485 people there, something like less than 10 actually chatting..) and I just realized it also lists a discord server in the topic haha.. well.. do...

welll, that's why we have a forum heh!

well.. if it's untagged, just set it up as a bond (say bond0), and set an IP on it, that will be untagged by default (all those things are using straight Linux networking stack). If you wanted a VLAN, then once the bond is set up, you can add a VLAN to it. But here, just use bondX directly, it's...

If you want migration to use both "vmotion" interfaces, you still need to set up a bond, and set an IP on that bond on each node. And of course, set that IP/network in the migration network in datacenter options. A bridge is only required if you want to attach a VM to that bond, and in that...

sure, but that has a CPU cost that can be a problem given a certain VM density, and because that traffic probably goes to VMs, it's used twice. There are good reasons to use an actual device for a firewall or a router, but simplicity is sure not one of them!

Also, if you have a bit of CPU to spare, and 1 Gbps max, I'd say that virtio would be enough, and provide more ease of use than passthrough.. however, it won't change that you have to fix your openwrt config..

Throw in a dmesg -Tw while you're at it.. also you could use qemu on the host to debug what's happening as well..

ah, oops, indeed.. and I'm French haha. Well, I saw something like 4 setups with that weird configuration, and 3 were OceanStor.. One was  PureStorage, but it was clearly misconfigured..

What fba suggested is mentionned here: https://forum.proxmox.com/threads/iscsi-huawei-oceanstore-dorado-3000-lvm-and-multipath.159971/ and.. maybe there is a configuration issue on your SAN heh.

I think Huawei (is it a Huawei storage server? :P) announces separate portals for each controller, with a different iqn for the target.. You might want to switch something like the path_grouping_policy or some other options in multipath.conf. I don't have the dorado at hand, but will ask the...

Also, what is the hardware configuration of the VM? I think you need to migrate the template to the node you want to clone locally to.. that is true also for VMs I think.

You probably saw that there is a question mark there.. If you think it's trying to mount it twice (it probably isn't), try disabling it in the UI, then explicitly unmount it (umount /mnt/pve/backups) or reboot if that's an option. Then well, it's telling you what to do.. use dmesg and check what...

This looks more like an OpenWRT issue than Proxmox.. Start with the second nic as disconnected, so as to set it up nicely in OpenWRT.. Probably dealing with multiple gateways, if you have DHCP on both sides. And make sure you're not creating a loop as well..

There have been serious issues reported with some versions of the kernel, and Ryzen CPUs, I don't remember the specifics, but check that your BIOS is up to date, and maybe try another kernel inside proxmox. You don't specify the versions there, but maybe try more recent kernels there...

Hi, Maybe share your storage.cfg. and check there that your local storage is set to allow Disk Images ?

This is more of a Samba question than Proxmox VE, I'd say. I never feel using containers for NAS software, but I guess samba should work that way. As for uid mapping, well, why would those on the pve host matter? Do you need the Samba shares to be mounted on the host? For vzdump backups maybe...

Use the recovery mode from grub or bypass systemd to see what is happening.. that little information.. is not enough. Could be a hardware issue. Or a failing network configuration (though it would tell..), or a filesystem issue..

It is not about keeping 20% free, it means the host won't start inflating the balloons until it has that much memory used. So no impact on the VMs regarding ballooning until you have less than 20% RAM available on the host (then, priorities kick in)

Use tcpdump/wireshark to check what's happening, -e will go down to ethernet.. I'm not sure what 10UP (untagged, pvid) and 1T (tagged) really means, though I imagine all untagged trafic is tagged 10, and vlan 1 is allowed. So either you need to have a separate NIC set for VLAN1 or try to set in...