Beta 9 iso seems broken:
Creating a USB installer for 9 and I keep getting this message:
View attachment 88768
tried via many methods that I've done before. Same issue. Same process with 8.4 ISO is fine.
alright will check the checksumNeither our successful internal testing on a wide range of HWs nor the feedback from the community support your statement.
That would rather point to the downloaded ISO. Did you ensure your downloaded PVE 9 ISO file is valid?
You could, e.g., compare the sha256sum with one listed in: https://enterprise.proxmox.com/iso/SHA256SUMS
My upgrade path ended up working so didn't need to do a fresh install so gave up on the usb. I was using a 32GB patriot USB on a Z890 Creator WiFI with latest BIIOS. I had no issues running the 8.4 installer on this machine with the same process, and a mini but will try some more when i build the new backup host. I always use the DD method, which I have for years. I tried etcher on windows and popos and had the same issues.And if that's not wrong, then please open a new thread, mention my nick here with @ and provide more details about which method you used to prepare the medium, which USB pen drive model you use–which should allow to interfer if it's attached vias USB as SCSI (UAS) or older methods–and the hardware you try to use it, e.g. EFI or legacy BIOS, which CPU and motherboard model.
Webhook notification targets using POST method that worked in 8.4.5 seem to have issues in 9.0.0~18 beta. With working config details from 8.4.5 transposed, it returns the following error when Test is clicked: "Could not test target: bad uri: POST is missing scheme (500)"
There don't seen to be any invisible characters mucking it up... anyone else experienced this?
View attachment 88776
Yes, this happens to me as well on PVE 9.0.0~19 and PBS 4.0.8 for my Discord webhooks. I have tried creating a brand new one with no avail.
Webhook notification targets using POST method that worked in 8.4.5 seem to have issues in 9.0.0~18 beta. With working config details from 8.4.5 transposed, it returns the following error when Test is clicked: "Could not test target: bad uri: POST is missing scheme (500)"
There don't seen to be any invisible characters mucking it up... anyone else experienced this?
I hope that snapshot trees will come in a future version.yes, this is an intentional limitation for now. while technically we could extend the feature to support arbitrary snapshot trees, the code (and in particular the failure scenarios) would get a lot more complex, so we limited it to snapshot chains and rollback to the most recent snapshot for the time being.
Currently each Snapshot lv use the same space than base image. In the future, we ll look to add thin provisionning with Dynamic lv resizing.Hi, I also think that the >> vendor agnostic << snapshot function for LVM shared block storage is a highly anticipated feature for many Proxmox VE users ... especially coming/migrating from a Vsphere/ESXi and enterprise (SAN / FC) environment.
I haven't tested it so far (will do soon) - the current Beta documentation regarding this feature is a little bit sparse .... so is there maybe some more explanation available how this works under the hood ? e.g. as soon as a snapshot is taken - is the initial/original LV/VM disk marked read-only (= backing disk) and then for every additional snapshot a new LV (delta/overlay LV) is created (automatically) out of the corresponding VG - with which size ? and must the LV size be actively monitored not running out of space ? etc .... thank you
yep, it's working now in beta 9.0.0~22. thanks!Thanks for the reports, this should be fixed by
https://lore.proxmox.com/pve-devel/20250804080604.118202-1-l.wagner@proxmox.com/T/#u
Thanks for confirming!yep, it's working now in beta 9.0.0~22. thanks!
Works for me as well on beta 9.0.0~22, thank you for being so quick!Thanks for the reports, this should be fixed by
https://lore.proxmox.com/pve-devel/20250804080604.118202-1-l.wagner@proxmox.com/T/#u
if you want to continue using it as a test system, you can simply upgrade using apt:Guys, now 9.0.3 has been released, do you have any advice about the upgrade ؟
For production systems you shouldn't have used the beta to begin withStarting fresh is recommended there.
OSPF is unique since FRR uses two daemons for IPv4 / 6 (OSPFv2 / 3) and we initially decided to only support IPv4 because of this. When we implement BGP, we certainly want to support dual-stack setups.Regarding Fabrics, are there any plans to add IPv6 with OSPF or BGP or is the plan to stick to IPv6 with only OpenFabric?
We use essential cookies to make this site work, and optional cookies to enhance your experience.