Hi,
I have set up a Proxmox VE 3.3 cluster with three nodes (HP ProLiant DL380 G6), with a ceph cluster on the same machines (all three nodes currently have two disks used as OSDs).
I am using 1 NIC for the public-facing side (vmbr0 with eth0) and three NICs with bonding for the storage and cluster communication (vmbr1 on bond0). The /etc/network/interfaces looks like this:
In the storage network, the nodes have the addresses .5, .10 and .15.
In /etc/hosts, the storage network IPs are connected to the hostname and AFAICT, all cluster and storage traffic is going through the bond.
I was able to set up the OSDs and the pool via the GUI and I could also add the RBD storage that way. My /etc/pve/storage.cfg looks like this:
The Ceph storage is properly shown on each node in the GUI. However, if I select the storage and click the "Content" tab, I cannot do anything, but rather get the error message
I am rather new to PVE, so I was not entirely sure where to look for the reason (of course I searched the web, but this error seems to be not very common). However, I tried to look through the logs in /var/log, where I couldn't find anything suspicious.
Any hints on a possible reason or a place to look at are highly appreciated. If you need more information on the setup, I will provide them ASAP.
Thank you very much!
I have set up a Proxmox VE 3.3 cluster with three nodes (HP ProLiant DL380 G6), with a ceph cluster on the same machines (all three nodes currently have two disks used as OSDs).
I am using 1 NIC for the public-facing side (vmbr0 with eth0) and three NICs with bonding for the storage and cluster communication (vmbr1 on bond0). The /etc/network/interfaces looks like this:
Code:
auto lo
iface lo inet loopback
auto vmbr0
iface vmbr0 inet static
address PUBLIC_IPv4
netmask 255.255.255.192
gateway GATEWAY
bridge_ports eth0
bridge_stp off
bridge_fd 0
iface vmbr0 inet6 static
address PUBLIC_IPv6
netmask 64
up route -A inet6 add v6PREFIX::/48 dev vmbr0
up route -A inet6 add default gw v6PREFIX::1 dev vmbr0
down route -A inet6 del default gw v6PREFIX::1 dev vmbr0
down route -A inet6 del v6PREFIX::/48 dev vmbr0
auto bond0
iface bond0 inet manual
slaves eth1 eth2 eth3
bond_mode 4
bond_miimon 100
bond_lacp_rate 1
bond_xmit_hash_policy 2
bond_downdelay 200
bond_updelay 200
mtu 9000
auto vmbr1
iface vmbr1 inet static
address 172.30.254.5
netmask 255.255.255.0
bridge_ports bond0
bridge_stp off
bridge_fd 0
mtu 9000
up route add -net 224.0.0.0 netmask 240.0.0.0 dev vmbr1
down route del -net 224.0.0.0 netmask 240.0.0.0 dev vmbr1In the storage network, the nodes have the addresses .5, .10 and .15.
In /etc/hosts, the storage network IPs are connected to the hostname and AFAICT, all cluster and storage traffic is going through the bond.
I was able to set up the OSDs and the pool via the GUI and I could also add the RBD storage that way. My /etc/pve/storage.cfg looks like this:
Code:
rbd: pool-ceph01
monhost 172.30.254.5 172.30.254.10 172.30.254.15
pool pool-ha01
content images
username adminThe Ceph storage is properly shown on each node in the GUI. However, if I select the storage and click the "Content" tab, I cannot do anything, but rather get the error message
I am rather new to PVE, so I was not entirely sure where to look for the reason (of course I searched the web, but this error seems to be not very common). However, I tried to look through the logs in /var/log, where I couldn't find anything suspicious.
Any hints on a possible reason or a place to look at are highly appreciated. If you need more information on the setup, I will provide them ASAP.
Thank you very much!
But at least now any search engine will find this thread when searching for the error message, so people who make the same mistake, will find the solution to the problem faster