API poor performance

[Guy]

Hello,

I've reinstall proxmox 4.1 on a 5 node cluster. I've fixed the clustering so igmp-snooping is working. Well via pvecm status I can see all the nodes are good.

Quorum information
------------------
Date: Wed Mar 16 08:45:34 2016
Quorum provider: corosync_votequorum
Nodes: 5
Node ID: 0x00000001
Ring ID: 392
Quorate: Yes

Votequorum information
----------------------
Expected votes: 5
Highest expected: 5
Total votes: 5
Quorum: 3
Flags: Quorate

Membership information
----------------------
Nodeid Votes Name
0x00000001 1 10.250.0.101 (local)
0x00000004 1 10.250.0.102
0x00000003 1 10.250.0.103
0x00000005 1 10.250.0.104
0x00000002 1 10.250.0.105


Seems happy right? And so far good, however when I try and point to one of the heads and ask questions of the others it takes forever to return via the webUI or via the API. I've tried rebooting all the cluster heads. I'm guessing I still have something wrong with my cluster but I'm not sure what now.

They are plugged into a juniper switch, and I have this configuration:

protocols {
igmp {
interface ge-0/0/3.0 {
version 2;
}
}
igmp-snooping {
vlan Operations;
}
rstp {
interface ge-1/0/24.0 {
edge;
}
}
vstp {
vlan all;
}
lldp {
interface all;
}
}

show igmp-snooping vlans
VLAN Interfaces Groups MRouters Receivers RxVlans
Operations 21 1 1 5 0


show igmp-snooping membership detail
VLAN: Operations Tag: 100 (Index: 7)
Router interfaces:
ge-0/0/47.0 dynamic Uptime: 07:21:36 timeout: 159
Group: 239.192.242.39
ge-0/0/5.0 timeout: 171 Last reporter: 10.250.0.105 Flags: <V2-hosts>
ge-0/0/4.0 timeout: 165 Last reporter: 10.250.0.104 Flags: <V2-hosts>
ge-0/0/3.0 timeout: 172 Last reporter: 10.250.0.103 Flags: <V2-hosts>
ge-0/0/2.0 timeout: 168 Last reporter: 10.250.0.102 Flags: <V2-hosts>
ge-0/0/1.0 timeout: 170 Last reporter: 10.250.0.101 Flags: <V2-hosts>