pzozulka
asked on
HP ProCurve Spanning Tree
We have two core procurve switches (5304xl), and a server that does NIC Teaming.
NIC Team:
Server NIC1 - connects to switch 1 (port C10)
Server NIC2 - connects to switch 2 (port A4)
The two switches are connected via a 4-port trunk, and have spanning tree enabled.
I'm assuming that since the switches are not aware of any "link aggregation" going on, they are using Spanning Tree to block one of these ports.
How do I find out on the switch side if either port C10 on switch1 is being blocked by spanning tree, or if the other port on switch 2 is being blocked?
NIC Team:
Server NIC1 - connects to switch 1 (port C10)
Server NIC2 - connects to switch 2 (port A4)
The two switches are connected via a 4-port trunk, and have spanning tree enabled.
I'm assuming that since the switches are not aware of any "link aggregation" going on, they are using Spanning Tree to block one of these ports.
How do I find out on the switch side if either port C10 on switch1 is being blocked by spanning tree, or if the other port on switch 2 is being blocked?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
What's the difference between "forwarding" and "disabled"?
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
show int brief
will show port status as up, down
a port can be up -but in a blocking state, ready to change state, when the topology changes
will show port status as up, down
a port can be up -but in a blocking state, ready to change state, when the topology changes
ASKER
Hmmm...this still doesn't answer my question on how to check if spanning tree is blocking one of the two ports.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
OK, so it appears that "show span" does not show any "Blocked" ports on either switch. All of them either say Disabled or Forwarding.
However, the logs seem to show something a bit different. The below ports are part of the NIC team.
Switch 1:
I 09/23/13 16:03:55 ports: port B10 is now off-line
I 09/23/13 16:04:03 ports: port B10 is Blocked by LACP
I 09/23/13 16:04:06 ports: port B10 is Blocked by STP
I 09/23/13 16:04:09 ports: port B10 is now on-line
I 09/23/13 16:05:12 ports: port B10 is now off-line
I 09/23/13 16:05:15 ports: port B10 is Blocked by LACP
I 09/23/13 16:05:18 ports: port B10 is Blocked by STP
I 09/23/13 16:05:20 ports: port B10 is now on-line
I 09/23/13 16:06:37 ports: port B10 is now off-line
I 09/23/13 16:06:40 ports: port B10 is Blocked by LACP
I 09/23/13 16:06:43 ports: port B10 is Blocked by STP
I 09/23/13 16:06:46 ports: port B10 is now on-line
I 09/23/13 16:06:53 ports: port B10 is now off-line
I 09/23/13 16:06:56 ports: port B10 is Blocked by LACP
I 09/23/13 16:06:59 ports: port B10 is Blocked by STP
I 09/23/13 16:07:02 ports: port B10 is now on-line
Eventually port B10 is on-line...
Switch 2:
I 09/23/13 17:03:52 ports: port A6 is now off-line
I 09/23/13 17:03:54 ports: port A6 is Blocked by LACP
I 09/23/13 17:03:57 ports: port A6 is now off-line
I 09/23/13 17:04:00 ports: port A6 is Blocked by LACP
I 09/23/13 17:04:00 ports: port A6 is Blocked by STP
I 09/23/13 17:04:03 ports: port A6 is now on-line
I 09/23/13 17:06:32 ports: port A6 is now off-line
I 09/23/13 17:06:35 ports: port A6 is Blocked by LACP
I 09/23/13 17:06:38 ports: port A6 is Blocked by STP
I 09/23/13 17:06:41 ports: port A6 is now on-line
Eventually port A6 is on-line...
What do these logs tell you?
However, the logs seem to show something a bit different. The below ports are part of the NIC team.
Switch 1:
I 09/23/13 16:03:55 ports: port B10 is now off-line
I 09/23/13 16:04:03 ports: port B10 is Blocked by LACP
I 09/23/13 16:04:06 ports: port B10 is Blocked by STP
I 09/23/13 16:04:09 ports: port B10 is now on-line
I 09/23/13 16:05:12 ports: port B10 is now off-line
I 09/23/13 16:05:15 ports: port B10 is Blocked by LACP
I 09/23/13 16:05:18 ports: port B10 is Blocked by STP
I 09/23/13 16:05:20 ports: port B10 is now on-line
I 09/23/13 16:06:37 ports: port B10 is now off-line
I 09/23/13 16:06:40 ports: port B10 is Blocked by LACP
I 09/23/13 16:06:43 ports: port B10 is Blocked by STP
I 09/23/13 16:06:46 ports: port B10 is now on-line
I 09/23/13 16:06:53 ports: port B10 is now off-line
I 09/23/13 16:06:56 ports: port B10 is Blocked by LACP
I 09/23/13 16:06:59 ports: port B10 is Blocked by STP
I 09/23/13 16:07:02 ports: port B10 is now on-line
Eventually port B10 is on-line...
Switch 2:
I 09/23/13 17:03:52 ports: port A6 is now off-line
I 09/23/13 17:03:54 ports: port A6 is Blocked by LACP
I 09/23/13 17:03:57 ports: port A6 is now off-line
I 09/23/13 17:04:00 ports: port A6 is Blocked by LACP
I 09/23/13 17:04:00 ports: port A6 is Blocked by STP
I 09/23/13 17:04:03 ports: port A6 is now on-line
I 09/23/13 17:06:32 ports: port A6 is now off-line
I 09/23/13 17:06:35 ports: port A6 is Blocked by LACP
I 09/23/13 17:06:38 ports: port A6 is Blocked by STP
I 09/23/13 17:06:41 ports: port A6 is now on-line
Eventually port A6 is on-line...
What do these logs tell you?
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Are you saying that there is no "loop" on the network NOT because spanning tree is blocking something, but perhaps because the vm-NICs are "blocking" on their end?
I will not jump to conclusions on the vm-side.
In a stable network with phy-switches a potential loop and STP blocking somewhere to prevent this loop, there would not be this high freqency of flapping:
-On a 'spanning tree accessport' (portfast) link there would be a repeated pattern of short broadcast-storm - then blocking for perhaps a few minutes.
-On a 'no spanning tree accessport' (that is a link between switches) the port would be in a stable blocking state.
In a stable network with phy-switches a potential loop and STP blocking somewhere to prevent this loop, there would not be this high freqency of flapping:
-On a 'spanning tree accessport' (portfast) link there would be a repeated pattern of short broadcast-storm - then blocking for perhaps a few minutes.
-On a 'no spanning tree accessport' (that is a link between switches) the port would be in a stable blocking state.
ASKER
Hmm...then not sure where else to look to determine how this NIC Team is working the way it is.
I mean everything is working fine -- and I don't think there are any problems with this NIC Team connecting to different switches. Afterall, I would think this would be a recommended strategy so that if one switch fails traffic would continue flowing uninterupted. In fact we tested this and it worked fine.
The reason I'm trying to get to the bottom of this is I NEED to understand how this is working. Either STP is blocking one the ports which would explain a lot. Or my only ohter conslusion would be the NIC team is isolating one of the NICs -- even though they are both setup as Active NICs.
Where else do I look to get to the bottom of this. Tried the logs and tried "show span".
I mean everything is working fine -- and I don't think there are any problems with this NIC Team connecting to different switches. Afterall, I would think this would be a recommended strategy so that if one switch fails traffic would continue flowing uninterupted. In fact we tested this and it worked fine.
The reason I'm trying to get to the bottom of this is I NEED to understand how this is working. Either STP is blocking one the ports which would explain a lot. Or my only ohter conslusion would be the NIC team is isolating one of the NICs -- even though they are both setup as Active NICs.
Where else do I look to get to the bottom of this. Tried the logs and tried "show span".
I am not an expert on vm's and cannot tell where to look inside these.
But as far as I hear the STP is not blocking anything on the outside.
But as far as I hear the STP is not blocking anything on the outside.
ASKER
Open in new window
Also, the image below is showing our NIC-team config. Although these NIC-Teams are not able to take advantage of bandwidth improvements, they are able to take advantage or redundant switches. In case one switch goes down, the other will continue to server traffic with zero environment down time.