Reading the docs on storage, it says:
"A number of storages, and the Qemu image format qcow2, support thin provisioning. "
It doesn't actually tell you which storages support thin provisioning, but I can I assume that dir storage does not (unless the images stored on it are qcow2)?
Not sure why your guests are using tap devices - are they connecting to the outside via a VPN or something? Either way it seems like an upstream issue to me if they've stopped being able to get ARP requests.
If you have changed nothing before this happened, I'd think that implies an upstream problem. It's possible that your ISP has introduced a change in the way they operate their network. This in fact happened to me recently with similar consequences.
Hard to say, but can you ping the gateway from the VM? The tcp dump implies you can. If so, I'd contact your ISP to check if they haven't changed something upstream from you.
This appears to be a similar issue to mine (although I'm using an older version of PVE).
When you say "yougetsignal.com/tools/open-ports says that all the ports are open" do you mean open on the guest(s) or the host itself? I note that with me, ports are coming up (correctly) as "filtered" when...
I'm trying to set some firewall rules at the guest level but they aren't being applied (see system information attached). This is what I have right now:
/etc/pve/firewall/cluster.fw:
[OPTIONS]
enable: 1
policy_in: DROP
[IPSET management] # Identifies the LAN
10.10.10.0/24
[RULES]
GROUP admin...
After some more investigation, I still have no clue as to why this might be happening. So I'm just going to drop WAN traffic from inside the guest in question.
Using PVE 4.4-22/2728f613 with the following config for a Linux VM:
net0: virtio=C2:75:29:81:BE:D4,bridge=vmbr0,firewall=1
net1: virtio=36:DE:0C:03:12:1B,bridge=vmbr1
net0 is the WAN interface, which I need to disconnect. However, when I do (using the PVE GUI) the WAN on another VMs goes down...
I'm running PVE 4.4-22 and want to enable the firewall on one of my VMs so that it drops all but some ports on its WAN network interface.
Using the GUI, I have enabled the firewall on the datacenter, at node 'host' and on the VM. The input policy on the VM is DROP. I have also enabled the...
Thanks, but the boot process used to work fine. Only after a PVE update a couple of months ago did it start refusing to boot (it actually goes into a loop, rebooting after getting past the POST routine I think).
So it would be good to know why the device map is present at all. Is there a reason...
I've got a weird problem with a PVE host machine that refuses to reboot without me having to explicitly choose the boot device from the BIOS.
Looking for clues on this I see /boot/grub/device.map contains the following. Does anyone know what this is trying to do?
(hd0)...
If at first USB doesn't boot, try another make/model of USB key. Some sticks just don't boot things, even if they boot other things. It's just black magic really.
If more than one stick is still refusing to boot, then try every possible permutation of making the bootable media: dd, unetbootin...
As hinted at in the man page, it's probably best to use an NFS server for the dump itself (massively faster than dumping straight to tape I would think), then copy the dump files to the tape drive at your leisure.
If you have no alternative but to dump straight to tape, then do you know the...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.