Search results

I suppose you want to assign these IP addresses to Guest Systems ( VM, Containers) ? Then you should configure the IP address inside each guest, not on the PVE host.

@helloworld: > I had the same issue last two weeks, suddenly arp packets are not received by the virtual machine. * can you see the arp packets when running tcpdump against the tap device of the VM ? * also which kind of load would make the problem reproduciblle ? does the problem happens...

you cannot use the same IP to connect multiple computers on the internet, each computer needs to have its own IP If you only have one IP and still want to give access to multiple VMs you need to use "masquerade" so your VMs have internet access and "port forwarding" if you need the rest of the...

Try the following command sequence and post their output: systemctl stop pveproxy systemctl start pveproxy ps faxl | grep pveproxy executed on your "sumo" node

@czechsys: How do you get the list of exports in that case ? (NFSv4)

You should be a bit more specific about your problem, but I would start with the following checks: * are you seeing all nodes of your cluster with the command pvecm nodes * is the storage of your containers OK ? pvesm status

showmount --exports asks the NFS server to provide a list of exported filesystems from the NFS server, and yes we use that to check NFS server avaibility on the server side this service is provided by the rpcbind service, which is listening on port 111 so you should: * check that this...

this could happen if the disks of your pool are a bit slow to initialize if i am not mistaken. When this happens: * verify that the kernel has detected all your disks ls /dev/sd[a-z] this should give you a list of devices, from which the quantity should match with the number of physical disks...

> Komisch ist jedoch, dass die Qcow2-Platten auf derjenigen NFS liegen, die den Schreibcache deaktiviert hat. Und diese sind defekt nach dem Backup. (so zumindest unsere Beobachtung) Die theorie, ist das wenn der schreib cache von einer platte deaktiviert ist, die I/O werden langsamer, aber...

Was betrefft der file server: Qcow2 corruption über NFS sind bis jetzt unbekannt. Von mir das problem stamm von einer NFS verbindung problem. Was zu achten: * ist jeder disk vom der VM in cache mode 'nocache' ? * welche NFS mount options sind benuzt ? Die nfs mount 'hard' option ist...

are you sure the VM is running ? what is the output of qm status my_vm_id or pct status my_ct_id if you have a container here

Maybe srvadmin-itunnelprovider has dependencies to jessie-only packages ? You can check that with apt-cache show srvadmin-itunnelprovider You should ping your reseller to Dell add a new repository for Debian Stretch.

Yes you can edit these configuration files manually, PVE does not have an intermediate state between the underlying configuration files and the GUI. concerning your setup: if you want to have your disk image as visible files in a zfs dataset, make sure you keep the format to raw, as qcow2 on...

Yes, this is possible, see https://pve.proxmox.com/pve-docs/chapter-qm.html#_importing_virtual_machines_from_foreign_hypervisors for example the only difference is that you the disk format will be qcow2 or raw if you're importing from KVM when using a ZFS pool as target storage, PVE will...

a pve cluster can have v4 and v5 running altogether however in you case I would do the v4->v5 upgrade and then add the node to the cluster follow the upgrades notes from https://pve.proxmox.com/wiki/Upgrade_from_4.x_to_5.0

Why do the tagged packets of VLAN 60 do not reach the trunk port ?

does the PVE guest or the host has no network access ?

so I understand you right, you tring to ping 10.152.152.10 from the VM with IP 10.152.152.12 ? are you sure the NICS using these IP are connected, and on the same bridge ? also try to create a new ubuntu VM and make it ping your firstubuntu VM, so you will know if have a network...

> IPAM in SolusVM isn't perfect but it fits our use case. It allows us to assign IP addresses and we can manage reverse DNS through its integration with PowerDNS. Both are powerful assets for us and we are looking for similar functionality in Proxmox. maybe you should describe a bit more what is...

Do you have any benchmarks of the storage performance problem this can solve ? Last time I checked a MTU of 9000 would provide a 5-10% performance over ISCSI when using 8k transfer blocks, and a lot of configuration headaches.