[SOLVED] Problems after Upgrade

[r4a5a88]

Hi

Sorry for the doublepost ( I pisted one like this in the german section)

I upgraded from wheezy to jessie (proxmox 3.x to 4.x)
I tried it with apt-get dist-upgrade and didn't read the upgrade section on the proxmox homepage
this lead to to a few errors. Afterwards I installed the server anew with the 4.2 image

the Problem is , I cannot add my Node back into my cluster
I have a cluster consisting of 5 servers
the server I upgraded was my main node at he time.

when I try to add it back into the cluster I get a authentication key already exists message.

what should I ?

 

[wosp]

Not sure, but try to run this on all nodes:

# mv /root/.ssh/known_hosts /root/.ssh/known_hosts.BAK

And try again (and start/test a SSH session from each node to all other nodes).
If this doesn't help, just run the following to restore the old situation:

# mv /root/.ssh/known_hosts.BAK /root/.ssh/known_hosts

 

[dietmar]

You can use the --force flag if you want to re-add a node.

 

[r4a5a88]

I tried it and got
unable to copy ssh ID

the I deleted the host from /etc/ssh/ssh_known_hosts

then I got

copy corosync auth key
rsync: link_stat "/var/lib/pve-cluster/corosync.authkey" failed: No such file or directory (2)
rsync: link_stat "/etc/pve/cluster.conf" failed: No such file or directory (2)
rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1536) [Receiver=3.0.9]
can't rsync data from host 'IP'

 

[r4a5a88]

I tried it form another member

copy corosync auth key
rsync: link_stat "/var/lib/pve-cluster/corosync.authkey" failed: No such file or directory (2)
rsync: link_stat "/etc/pve/cluster.conf" failed: No such file or directory (2)
rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1536) [Receiver=3.0.9]
can't rsync data from host 'x.x.x.x'

t

 

[r4a5a88]

it seems these files are missing on the node I am trying to add

 

[r4a5a88]

it seems I did it the wrong way I sould have ttried it on the node I wanted to add
but htere he tells me the corosync directory is not there

 

[r4a5a88]

I solved it
I just created a new cluster and will ad the other nodes after the upgrade

 

[Norberto Iannicelli]

For me it was worse, after updating all my cluster has a problem.
load balance does not work anymore, I have to shut down a network interface to the cluster work again, if I leave the two network interfaces connected the nodes are restarting without stopping and no quorum occurs. Someone from the staff could help me? Shipping details.