Search results

Hi Michael, depends on the SSDs. ZFS on Linux is very flexible, but not really fast... with SSD-only raids it's ok, but with HDDs it's depends on your IO-workload. An SSD or NVMe for journaling and cache can be very helpfull. But use the right SSD (enterprise - look, which ssd is usable for...

Hi, das ist aber nur bedingt ein Backup... Aber Du könntest znapzend von Tobias Oetiker verwenden https://github.com/oetiker/znapzend Damit kannst Du ein Plan erstellen, welche snapshots auf dem Ziel existieren sollen (nicht auf der Quelle). Damit benötigst Du nur (massenhaft) Platz auf dem...

Hi Spirit, I use OVS for network and it was an double vlan tag. For safety I will try the bnx-option. Udo

Hi Spirit, since that I'm aware not to use vlan tagging on an tagged vm-nic - so the issue don't occour (but if fixed it's will be much better!). Udo

Hi, if this isn't an cluster, the pvecm output is normal. What is the output of pve5to6 Udo

Hi, if you are sure, that the device are not used, you can follow this guide i wrote for me: dmsetup info /dev/sata/vm-210-disk-1 Name: sata-vm--210--disk--1 State: ACTIVE Read Ahead: 256 Tables present: LIVE Open count: 6 Event number: 0 Major...

Hi Chris, thanks, work like a charm. Udo

Hi, I want simply add an new created VM with pvesh to an pool (like "pvesh add pools/Dev -members 123"). OK, pvesh don't know add - but my try with set are not successfull. How is the right syntax? Udo

Hi, die "bloße Aussage" von LnxBil kommt nach meiner Meinung daher, weil "messungen" mit dd nicht für alle Storage-Type vergleichbar sind. So wird bei zfs z.b. komprimiert - und mit dd aus /dev/zero auf ein zfs-Filesystem zu schreiben, bringt zwar tolle Werte, aber die haben nichts mit der...

Hi, ich würde mal vermuten, dass es entweder an der usb-3 performance liegt, oder an der Raidperformance selber. Und mit 5400 U/min im Raid-1 ist so ziemlich worst case... Udo

Hi, @guletz: I will try volblock-sizes later. With the new Kernel, the test takes 40m27.5s and the load looks much better. pveversion -v proxmox-ve: 6.0-2 (running kernel: 5.0.21-1-pve) pve-manager: 6.0-6 (running version: 6.0-6/c71f879f) pve-kernel-5.0: 6.0-7 pve-kernel-helper: 6.0-7...

Hi Stefan, thanks for the feedback. I will do an test again, but need some time. Udo

Hi, if I understand you right, your read the compredd data from the same zfs-pool (inside an VM), where you write the output? If you look for the throughput: (read 14001635328 bytes, duration 75 sec - read 7000817664 bytes, duration 31 sec)/44/(1024*1024) you got 151MB/s. (tihs is readed...

Hi, I've installed pve-6 successfull on a 6 * 6TB-4kn-hdd raidz2 with the help of an single disk (with 512b sectorsize) and manual reconfigure (take some time). If you wan't I can post the HowTo... Udo

Hi Davide, if you look here: https://www.seagate.com/enterprise-storage/exos-drives/exos-e-drives/exos-7e8/ you see some models with 4k sector size and some with emulated 512bytes. I assume you have an 4k-model too. Because raid-0: are your backup not important? If yone disk die - all backups...

Hi Davide, I have the same issue with different 6TB-disks. Looks that the pve-installer can't handle native 4k sectors correct! Has your disks 4k sectors (4kn) or emulated 512b (512e)? Udo BTW. zfs raid 0 is an bad idea

Hi, sorry - you must measure caching! With an 5-OSD HDD cluster on 1GB-Network you will never ever got 80-100MB thorughput in an VM / single thread. Ok, with replica 3 and 5 OSDs you can got 100/(5/3) = 60MB inside an VM with 100MB/s per OSD - but I don't think that you reach such values! Try...

Hi, documentation: https://docs.ceph.com/docs/master/install/manual-deployment/#adding-osds I would NOT use ceph-mons outside the pve-cluster, only osd-nodes (which is fine, if you have enough resources). Esp. ceph-mons should have the same (and the newest version) - osds are not so critical...

Hi, if you look here: https://docs.ceph.com/docs/jewel/start/hardware-recommendations/ you see, that you need min 3GB free ram for one OSD. And newer versions need more ram - see also here: https://unix.stackexchange.com/questions/448801/ceph-luminous-osd-memory-usage?rq=1 second - linux use...

Hi, you can simply install an OS with the same ceph-version and use the same ceph.conf like the pve-cluster... BUT I would not recommend to run with 1GB any ceph services!! You will got much trouble (don't know your testing, but I'm quite sure, that you will run in trouble). Udo