Search results

Okay, the solution (after I've re-read Perl of Install.pm :) ) rm added-proxmox-ve_8.3-1-auto-from-partition.iso xorriso -indev proxmox-ve_8.3-1-auto-from-partition.iso -outdev added-proxmox-ve_8.3-1-auto-from-partition.iso \ -map /root/frr /proxmox/packages \ -boot_image any keep Where...

Hi Cheiss I've gone and read through and looked/considered that earlier today, BUT before I haven't installed the FRR (or anything else IS-IS) and have the IGP running to get a default gateway, I don't have "internet access" to fetch/apt anything. That is the reason I need the FRR package to...

Edit: xorisso solution I have this "use case" where the PVEs will have a meshed networking setup, ie. the transit/host/production network not connected to a switch/LAN/L2 with DHCP/PXE server somewhere broadcast domain reachable, but only PtPlinks in the cluster and up to the routers...

Seems to be included in 8.3's :)

The expected JSON format :) perhaps a versioned JSON format for when things change in PVE major versions etc. From a "sysadmin" perspective, a feature request: - perhaps GET requests with parameters (defined at "config" time, or in the DHCP/DNS/etc.) like...

Having goten bitten by this myself, the sequence of events that peefed me that I believe was the reason for this change: have "working" /etc/ssh/known_hosts --symlink-> /etc/pve/priv/ssh/known_hosts (can't recall the exact symlink destination) from a node you ssh to something/where else, and...

fun part, had an exact similar issue on PVE 8.2.x and this thread fixed it for me :D

Good day, ZANOG is busy setting up a community mirror for ZA (South Africa https://nog.net.za/mirrors/ ) Reason is the CDN sometimes resolves to NA, even when the fr.cdn.proxmox.com is the closer one, but still, EU traffic isn't always so nice and fast either and I'm also getting hit by the...

Apologies for English response, but this is the best fit for the problem from the other responses. Though the first answer (restore as Privileged container) does fix the problem, I have the problem also with the `random` device in UNprivileged containers. The ROOT CAUSE seems to be related to...

See: https://forum.proxmox.com/threads/linux-bridge-many-duplicates-in-mac-address-table-for-the-local-interface.117730/post-509703

I sometimes wonders even more about the Linux Kernel devs' NIHS views ;(

Going off topic: Those horses know to stay away from IT/computers and are running around freely.... do my 35+ years in this industry show?

Think it's still a problem in Devuan 5 too... need to get/make a Devuan5 template one day

Different horses for different courses Openstack is much beterer in the mass chicken world. In my world there are cows and alpacas and dogs and the cat or two. And OpenStack just doesn't handle that as nicely, while ProxMox I see will falter when you do need those loads of chickens in multiple...

Ease of identification of which bridge it uses, which client and/or which IP network - some of us does have OCD ;)

So, I got hit by this, from the perspective of *ADD*ing a PVE8.2 node to a 7.x cluster, with the intention to migrate the vMs/LXCs to that while decommissioning the old host it's replacing, rinse repeat for the rest of the new & old nodes. So, yes, the challenge is on to add a node to an old...

how do I "run" these in the image/VM right after creation? only fixed/static IP assignments, so networking can't rely on DHCP

I have needs to have multiple IPs/networks on the hypervisors, typically on different VLANs (like backups, corosyncing, ceph,etc. etc.) Having used OVS before, it is/was a charm to do similar, as you just add an OVSIntPort with the VLAN and IPs, and you are done. As the SDN is using Linux...

I have used it on just about all my PVE & PBS going to PVE 8.2 & PBS 3.2 as just about all of them had the Intel interfaces that have that virtual function causing the "change" that the System(&@#$(%&@#$%&(D implements I have loads of )&#@$(%)&@#$)%(& words for SystemD, as the correct way for...

Busy FAFOing with this in a totally meshed setup, and the use case is to have Ceph on that SDN created VxLan/EVPN vnets of it's own. Reason to NOT have it on the "public" network, but also to allow VMs (pointing fingers to K8s) that wants direct Ceph storage access on the access network, and in...