Die Virtualisierungslösung Proxmox VE wurde vor ein paar Wochen in der Version 5.4 veröffentlicht. Laut Entwickler konzentriert sich das Update auf eine einfache Bedienbarkeit und Integration in die Unternehmesstruktur. Neu ist auch die komplette Konfiguration von Ceph-Storage über das Webinterface von Proxmox VE. Proxmox VE 5.4 basiert auf Debian Stretch 9.8 mit Kernel 4.15.18.
Proxmox VE 5.4 Features
- Debian Stretch 9.8 and Linux Kernel 4.15
- New installation wizard for Ceph in the UI,
- New HA policies freeze/fail-over/default for greater flexibility,
- Suspend to disk/Hibernation support for Qemu guests,
- Support for Universal 2nd Factor (U2F) authentication,
- Improved ISO installation wizard,
- New options for Qemu guest creation wizard
Quelle: https://www.proxmox.com/de/training/video-tutorials/item/what-s-new-in-proxmox-ve-5-4
Proxmox VE 5.4 Release Notes
- Based on Debian Stretch 9.8
- Kernel 4.15.18
- QEMU 2.12.1
- LXC 3.1.0
- ZFS 0.7.13
- Ceph 12.2.11 (Luminous LTS, stable), packaged by Proxmox
- Ceph can now be installed via the GUI, thus closing the gap of needing the CLI for this initial step, before configuring a complete (hyperconverged) Ceph setup via PVE’s GUI.
- QDevice support via `pvecm`
- primarily for small 2-node clusters adding a qdevice can help mitigate the downside of not being able to reboot one node without losing quorum (and thus the ability to make any changes in the cluster) ** Can also help in clusters with a larger even number of nodes by providing a tie-break vote.
- Integration into pvecm and PVE stack vastly simplifies adding a qdevice (it was possible manually before as well)
- Wake On Lan support via `pvenode`
- It is now possible to associate a MAC-Address with a PVE-node.
- This can be used to send a WOL (wake on lan) packet to that node from any other node in the PVE-Cluster, and thus power it on on-demand
- Provides a comfortable way of saving power by keeping only part of the nodes permanently online, and starting others on demand (e.g. increased load, or for running guests during upgrades to the PVE-nodes.)
- GUI Container wizard creates unprivileged containers by default
- HA improvements and added flexibility
- It is now possible to set a datacenter wide HA policy which can
change the way guests are treated upon a Node shutdown or reboot. The
choices are:
- freeze: always freeze servivces, independent of the shutdown type (reboot, poweroff)
- failover: never freeze services, this means that a service will get recovered to another node if possible and if the current node does not comes back up in the grace period of 1 minute.
- default: this is the current behavior, freeze on reboot but do not freeze on poweroff
- It is now possible to set a datacenter wide HA policy which can
change the way guests are treated upon a Node shutdown or reboot. The
choices are:
- Suspend to disk/hibernate support for Qemu/KVM guests
- qemu guests can be ‚hibernated‘ (have their RAM contents and internal state saved to permanent storage) and resumed on the next start.
- This enables users to preserve the running state of the qemu-guests across most upgrades to and reboots of the PVE-node.
- Additionally it can speed up the startup of guests running complex workloads/ workloads which take lots of resources to setup initially, but which need not run permanently.
- Support for guest (both Qemu and LXC) hookscripts:
- Hook-scripts are small executables which can be configured for each guest, and are called at certain steps of the guest’s lifetime (‚pre-start‘, ‚post-start‘, ‚pre-stop‘, ‚post-stop‘).
- This gives Administrators great flexibility in the way they can prepare the environment for the guest (e.g. adding necessary network resources (routes, vlans), firewall-rules, unlocking encrypted files/devices,…) and cleaning them up when the guest is shutdown or stopped.
- Improved Qemu Guest creation wizard:
- Certain often requested options (e.g. Machine-type (q35, pc-i440fx), Firmware (Seabios, UEFI), SCSI controller) can now be selected directly in the VM Creation Wizard, and dependent options get set to sensible values directly.
- Clusterwide Subscription Check and Display
- The Datacenter Summary page now gives Administrators a global overview of the subscription status of all nodes in a PVE cluster, thus helping to verify the support-status of the complete cluster on one quick glance
- Improved UI in the installer
- It is now possible to go back to a previous screen in the pve-installer, and adapt choices made there, without the need to restart the complete installation process
- Before the actual installation is performed an information-page is displayed containing all relevant information
- U2F Authentication for the WebUI
- PVE’s WebUI now can optionally use two-factor authentication for users.
- This is generally considered good security practice, and in certain domains and environments it is required.
- The new U2F authentication and the TOTP second factor authentication can be configured by each user for themselves (before all 2FA solutions needed to be configured by a user with User.Modify permission).
- Improved reference documentation
- Countless bug fixes and package updates (for all details see bugtracker and GIT)
Interessiert in verschiedenste IT Themen, schreibe ich in diesem Blog über Software, Hardware, Smart Home, Games und vieles mehr. Ich berichte z.B. über die Installation und Konfiguration von Software als auch von Problemen mit dieser. News sind ebenso spannend, sodass ich auch über Updates, Releases und Neuigkeiten aus der IT berichte. Letztendlich nutze ich Taste-of-IT als eigene Dokumentation und Anlaufstelle bei wiederkehrenden Themen. Ich hoffe ich kann dich ebenso informieren und bei Problemen eine schnelle Lösung anbieten. Wer meinen Aufwand unterstützen möchte, kann gerne eine Tasse oder Pod Kaffe per PayPal spenden – vielen Dank.