Latest activity
-
Uuzumo replied to the thread yawgpp (Yet Another Windows Guest Performance Post).Past comparisons regarding HVICs (One of the functions grouped under the name VBS) are as follows. VBS itself does not reduce processing speed. Only certain components (HVIC) reduce processing speed. *Performance degradation is equivalent to...
-
EllerholdAG replied to the thread IPERF3 not exceeding the bandwidth beyond 10G even if proxmox hosts are connected on 25G.You need 3.16+ to use multiple CPU cores. Before 3.16 it still uses only one core, so youd have to start multiple instances of iperf at the same time and add them together manually. -
EllerholdAG replied to the thread IPERF3 not exceeding the bandwidth beyond 10G even if proxmox hosts are connected on 25G.Which iperf verson did you use? Is it the single-core-only variant?
-
Eermanishchawla posted the thread IPERF3 not exceeding the bandwidth beyond 10G even if proxmox hosts are connected on 25G in Proxmox VE: Installation and configuration.Sender Side iperf3 -c 10.XX.YY.8 -u -b 10G -i 1 -t 10000 -P2 Connecting to host 10.25.240.8, port 5201 [ 5] local 10.25.240.7 port 49254 connected to 10.25.240.8 port 5201 [ 7] local 10.25.240.7 port 48981 connected to 10.25.240.8 port 5201 [...
-
Ttchaikov replied to the thread VMs freeze after node failure/reboot in 5-node Ceph cluster.MGR's actual role: MgrClient in the OSD (src/osd/OSD.cc) only does: send_pgstats(), update_daemon_health(), set_perf_metric_query_cb() — all async, none on the I/O path PeeringState.cc has zero references to MGR PrimaryLogPG::do_request() never...
-
Ttchaikov replied to the thread VMs freeze after node failure/reboot in 5-node Ceph cluster.Hi @torbho, before recommending solutions, I'd need more information: Reproducibility: Does this happen every time a node goes down/reboots, or was this a one-time event? If reproducible, this strongly suggests a structural issue (network...
-
JJohannes S reacted to UdoB's post in the thread Whole disk backup every time on stopped VM with
Like.
Maybe. I know for Datasets there are other attributes (meta-data) to evaluate. This thread is about a VM/ZVOL. Yeah, sounds clever! Thanks for adding this. But nevertheless the source ZVOL needs to be read, completely. And that was the main... -
JJohannes S reacted to IsThisThingOn's post in the thread Whole disk backup every time on stopped VM with Like.
Like others already said, by default it reads everything. The money quote line is this one: 107: 2026-02-28 21:00:07 INFO: scsi0: dirty-bitmap status: created new Since the VM is not running and/or you have the stop backup mode, it can't make...
-
JJohannes S reacted to Impact's post in the thread Whole disk backup every time on stopped VM with Like.
Keep the VM running so the dirty bitmap stays intact. Why?
-
JJohannes S replied to the thread Backup job creates non-conventional folder path; QNAP cannot read.Is using USB disc as removable datastore directly on the PBS an option? The idea would be to exclude the PBS datastores from the backup on the qnap and use a sync-job on the PBS to sync your pbs backups to the USB discs.
-
JJohannes S replied to the thread DHCP für PBS im Pullbetrieb.Und für den Edgecase, dass einen regulärer Datentransfer zu lange dauern würde, kann man ja außerdem auch eine externe Platte als removable Datastore verwenden. Die kriegt man auch im Taxi mit ;) Ein Featurerequest für DHCP-Unterstützung ist...
-
JWozu für diesen selten Fall das tun? Wenn das wirklich notwendig wird, passt man einmal die IP in der /etc/network/interfaces an. Das dauert 5 Sekunden. Da ist die Wahrscheinlichkeit für Probleme mit DHCP in vielen Jahren Betrieb größer.
-
JBei jedem Server habe ich ein BMC wo ich auf die Konsole komme. Das mache ich Weltweit Remote regelmäßig. Das BMC bleibt auf DHCP weil unkritisch und Server statisch. Das funktioniert seit Jahren wunderbar. Wer Consumerhardware nutzt, kann auch...
-
SSuBun7 replied to the thread [SOLVED] Users Connecting to Minecraft Server in VM Cause ProxMox Server to Lose All Internet.I'm having trouble figuring out how to find my ethernet device name - disregard, found
-
Ppulipulichen reacted to gfngfn256's post in the thread P2V/V2V errors and storage "failed to stat" errors preventing migration to PVE cluster hosts with Like.
Before committing, I would first run: qm importovf <new_vmid> <path_to_ovf_file> <target_storage> --dryrun to check that the OVF manifest is correctly populated. This is advisable as different systems produce different manifests.
-
DDerekG replied to the thread Connection Error with USB-Device.The USB ports in these older system are notoriously unreliable, and using that for PBS which is a heavy on the I/O appliance is really asking for trouble. In addition to that using a spinning disk as your backup data store is also a problem due...
-
DDerekG replied to the thread Intel NUC11 Nvidia 2060 proxmox gpu passthrough.How are you connecting the Nvidia 2060 to the NUC Box, are you connected via the Thunderbolt port? Have you enabled the device in the thunderbolt security settings? And in the BIOS.? There are 2 thunderbolt tools which need to be installed in...
-
Ssgp reacted to codingspiderfox's post in the thread Feature request: ask for Wake-On-LAN MAC address when not yet configured with Like.
Yes I know that it can be done this way but it would be more user friendly not having to look up that command. Also I think a good way is having a default interface name entered by the user which should always become the WoL interface. This...
-
Ggrubs replied to the thread question file level backup.Similarly - I'm doing a file system backup of cifs share 500GB in metadata mode takes 8-12 hours (but completes and works well for my purpose). I assume proxmox-backup-client in metadata mode is scaning the entire file contents over the network...
-
JJohannes S reacted to TErxleben's post in the thread SSD Backup mit ddrescue: Dateien wieder zurückspielen? with Like.
Da schwillt mir der Kamm! Du bist schon so ein echter Spezialist. Hast du wenigstens gelernt, dass im virtualisiertem Umfeld ein Virtualisierungs-Host in 15min neu aufgesetzt ist und dazu nur die Restore-Wartezeit einzelner VMs kommt? GENAU...