ember raptor
After we added 4 new C800 nodes, following errors on all new nodes have started in event log:
cpeer.xcm.update.warn: Periodic update of peer network information failed. The following operations are incomplete: node/ip map update failed.
Furhter, "cluster peer health show -bypass-cache true" comand shows that new nodes cannot recognize any peer clusters/nodes, encrypted or not.
Worked with Support for a week or so, but still no solution.
If any experts here please can provide a clue.
pale ingot
Silly question but do so all nodes on each side of the cluster peer all have an intercluster ip address?
And on each side of the peer, the nodes in each cluster, are the intercluster lifs in the same ipspace?
ember raptor
Yes, just double checked, intercluster lif's on the source and destinations are all in the same ipspace "Default"
eternal gorge
and you can ping the peers? 'cluster peer ping'
...('cluster peer connection show' ... 'cluster peer offer show' ...)
static breach
wow, a week.. I know the feeling.
and yea, down to bare basics.
ping each lif to/from each node, probably already done that with support.
and then break it down to one node at a time, uncable all but one if possible, start with just adding that one, go from there
ember raptor
cluster peer ping -originating-node new-c800 -destination-cluster xyz
shows nothing. If originating-node is pre-existing one, then it works fine.
So, something is worng here.
network ping -node new-c800 new -destination IP-ic-in-destination cluster
works fine.
cluster peer connection show
shows UNIDENTIFIABLE on all new C800 nodes.
static breach
check network ports on the switch, vlan tags, blocking, etc
ember raptor
On the switch? You meant back-end switches? Not sure of what exactly need to be checked on the switch side. All vlan's for IC, client connections(whch are all fine) etc. are running on the same LACP group. Only new inter cluster connections have issues.
static breach
yes, on the switch side.
just verify port/vlan/etc config.
We've had issues where adding new systems the ports did not get set correctly, or were having issues and had to be shutdown and back up, etc.
If you can't ping from lif to lif, something has to be blocked/broken.
back to the bare minimum/basics.
ember raptor
ping from lif to lif is working fine.
And, those pre-existing nodes in the same vlan work fine, if vlan or the port on the switch is wrong, then all IC or LACPwon't work, so, I would say it looks not switches related.
unborn inlet
Does this sound like your issue?
ember raptor
Not exactly the same.
The error in event log seems the same.
But, in my case it is worsen, because from the source cluster, "cluster peer health show" the new nodes cannot see the peer clusters, pre-existing nodes can.
eternal gorge
netmask correct?
ember raptor
Just double checked, yes, all the same across IC lif's
eternal gorge
i'm with Ty then... make sure vlans on the switchports for the new nodes are correct... ACI policies too, if applicable
ember raptor
Well, I have asked the networking guy to check out switch port. He din't find anything. He didn't need to make any changes on vlan configurations. He just need to trunk over the same vlans as the other nodes. ACL is Not blocked.
As I said, the other nodes in the same vlan works fine, but the new nodes.
static breach
not sure, something has a break in it.
network/port or the netapp config.
pull all the nodes but one if you can
one network port/cable at a time
ping each lif to each lif from node to node, etc
pale ingot
If there is a firewall in place a rule may be blocking important traffic