Search results

just a note you dont need a bridge for the raw host access, you can just leave it as eth0.1+eth1.1 ---> bond0 and iptables restrict on bond0 device note you would have to either not assign an IP for vmbr1 or manually reconfigure pveproxy server to only listen on the IP assigned to bond0/vmbr0

you need to bridge them with a physical interface thats linking the servers

i run openvpn inside kvm vm's no problems and no major packet loss to report (at least not due to openvpn) remember that default openvpn goes over UDP so if the network between your 2 machines is bad then you will get packet loss, for more unreliable networks you might consider switching to TCP...

I'm having same version running latest (non-enterprise) packages: root@rsw1:~# pveversion -v proxmox-ve-2.6.32: 3.1-114 (running kernel: 2.6.32-26-pve) pve-manager: 3.1-24 (running version: 3.1-24/060bd5a6) pve-kernel-2.6.32-20-pve: 2.6.32-100 pve-kernel-2.6.32-26-pve: 2.6.32-114...

DotDeb are also maintaining a current Zabbix Agent version in there repo - more info at http://www.dotdeb.org

Re: sheepdog storage only available on some pve nodes looks like this is due to updating qemu and not rebooting instances first rather than any issue

Re: sheepdog storage only available on some pve nodes perfect thanks i can live migrate with it nicely :) only thing to note is that i couldnt migrate storage to it without first turning the VM off - i got the following error if i tried with the node online: TASK ERROR: storage migration...

i setup my nfs store for some vm backups via proxmox - its configured to backup automatically and retains the latest 2 backups manually i cant manually create a backup without deleting one the retained auto backups or i get this error INFO: starting new backup job: vzdump 101 --remove 0 --mode...

So wait a law firm is wanting to knowingly put its data in VM's with an outdated and likely very insecure guest OS thats poorly optimised for running in a virtualised environment (tip make sure you install virtio drivers) i know this is a proxmox forum but Windows Terminal Services is the way...

i have a 5 node cluster (not HA yet) thats slowly growing i've been looking at shared storage without requiring SAN's and am all but settled on sheepdog as an option 2 of my servers are not really adequate for running sheepdog so i want to avoid installing on those nodes and not allow booting...

node addresses iirc is the IP's of the other nodes in the cluster? - actually no its the server address that you ran it on - and its the IP it has detected cluster comms is happening on are you by any chance using an openvpn tunnel? - you may need to use bridging/tap mode to get the ip to...

VMBR is only needed if you are using KVM VM's if you are just using OpenVZ containers its not required if you have shorewall on the host and both the external IP's on the host - you will have to configure your VM's with internal IP's that the hostnode can ping (you may have to set an internal...

agreed but for a quick and dirty test, i think you can setup a pseudo fencing device in the cluster config that doesnt do anything but allows the rest of the cluster actions to work, then you can just pull the power from the node running the VM and watch proxmox cluster on one of the other hosts

nothing you can do about this, backup and restore works fine though, just renumber them when you restore if your not worried about HA (and moving vm's around) this, isnt such a critical issue, clustering wont break your network configs if you dont move your VM's around and as long as all the...

did you install the intermediate cert or not? are you running the latest version of proxmox also - there was a bug that was fixed with intermediate certs for the java applet recently

you can create a vmbr interface thats not bridged to a physical eth interface, it will just be host only (that is you cannot access those IP's from outside the hostnode) but in short to do what you want you need something to do NAT - thats either 1) manual config in proxmox hostnode 2) using...

you need to change the UUID references in /etc/fstab to there new references as a new disk/partition will have a new UUID after format you can find your current disk UUID by running this command blkid

you have to resize the partition physically if the LVM physical disk is limited to 15GB first - a quick and dirty fix if the rest of the disk space is unpartitioned - create a new ext3/4 partition and mount it as something like /data then move/copy /var/lib/vz to /data/vz and then symlink...

the key thing to note is understand your application and its performance impact, and design your system around the app, i've never had issues with "soft" load balancers like haproxy - in many cases you get better control out of soft load balances over hardware ones, and better failover...

half of your errors seem like incorrect config of CSF to me, its a very powerful tool but if not setup right at the beginning it can cause more problems than it fixes