Unsolved
This post is more than 5 years old
3 Posts
0
1817
February 28th, 2017 02:00
Dell N4064f DCB setup for SMB
Hello
We are having a big problem enabling Datacenter Bridging for SMB traffic for Storage Spaces Direct.
This is the commands we run
conf t
classofservice traffic-class-group 0 1
classofservice traffic-class-group 1 1
classofservice traffic-class-group 2 1
classofservice traffic-class-group 3 0
classofservice traffic-class-group 4 1
classofservice traffic-class-group 5 1
classofservice traffic-class-group 6 1
traffic-class-group max-bandwidth 50 50 0
traffic-class-group min-bandwidth 50 25 25
traffic-class-group weight 50 50 0
interface range ten 1/0/41,ten1/0/42,ten1/0/43,ten1/0/44,ten2/0/41,ten2/0/42,ten2/0/43,ten2/0/44
lldp dcbx version auto
lldp tlv-select dcbxp pfc
datacenter-bridging
priority-flow-control mode on
priority-flow-control priority 3 no-drop
As soon as we enable priority-flow-control priority 3 no-drop i loose connectivity to alot of servers and blade chassies connected on port channels.
If i shutdown the ports, do the config all is fine. If i enable the ports same result, loosing connectivity to alt of servers. I can see spanningtree doing all sorts of things. If i disable spanningtree same issue. But then i loose connectivity with our managment switch wich is a cisco 2960 switch.
Firmware is 6.3.1.11 i see there is a new version out, but the release note only has 2 fixes and not for this.
Now another thing that happnes now is that i lost SNMP connectivity to the switch. As it won't return any snmp statistics. We monitor bandwith on the switches for our wan traffic.
Please help, as we need to get RDMA working for Storage Spaces Direct.
0 events found
jantorep
7 Posts
0
February 28th, 2017 12:00
Yeah followed that guide. Found the cli manual as well and figured out the diffrent settings based on how i set it up on the S4810p switch we in another datacenter.
All config should be correct. But as soon as i turn on priority-flow-control priority 3 no-drop i loose connectivity to our blade chassies wich are connect via port-channel and some dell rx730 rack servers. I also loose connectivity to our managment server. But i can rdp vis rdpgw in to vm's running one blade servers. From there i can remote to all blade servers on 3 chassies. And the managment server that i can't reach from another subnet.
As soon as i turn this off all is back to normal. Will post some output from console logging tomorrow. On phone now.
I do belive this is a bug, as when i turned off spanningtree i lost connectivity to our managment switch. We used to have a dual ethernet link setup towards our office on the n4000's but can't anymore as the n4000's dont like spanning-tree port priority. So had to move them to some older cisco switches we have.
The problem feels like certain vlans get blocked on certsin ports. Between certain ports
As i can ping some servers on same vlan. The the servers i cant ping from some servers can ping the servers that i cant ping. Sonlets say sever 1-4 can ping eachother and 5-8 can ping eachother. But 1-4 cant ping 5-8 and vise versa. And they are on the same switch and on same vlan.
So very strange
Infront
3 Posts
0
March 1st, 2017 02:00
I have added a log from the switch where i enable it on the servers and the ports for the port-channels for blade switches. But same effect as when just doing it on the servers ports.
The spanningtree does alot of things when i enable
priority-flow-control mode on
But still things work, it's when i enable
priority-flow-control priority 3 no-drop
That's when connectivity stops working
1 Attachment
dcb.txt