UNSOLVED
Performance issues and "foreign PDU" error
Hi,
We're using a M100e chassis with 2 M6220 switches. The M6220 aren't connected with eachother, but are connected to 2 Cisco 2970G's:
M6220 <--> C2970G <--> C2970G <--> M6220
We're experiencing performance issues, like 1% packetloss and slow responses.
Both Cisco switche do not show any error or warning in the logs, but both powerconnects are flooded with these messages:
<189> APR 23 17:19:12 10.0.250.22-1 ISDP[122766080]: isdp_txrx.c(169) 459307 %%Foreign pdu on interface Po1
I've been searching the whole net, but haven't found an explanation of these errors.
Does anyone know where these errors come from?
Config on Powerconnects:
interface Gi1/0/17
channel-group 1 mode active
description "C2970G-a"
mtu 9216
switchport mode trunk
exit
!
interface Gi1/0/18
--More-- or (q)uit
channel-group 1 mode active
description "C2970G-a"
mtu 9216
switchport mode trunk
exit
!
interface Gi1/0/19
channel-group 1 mode active
description "C2970G-a"
mtu 9216
switchport mode trunk
exit
!
interface Gi1/0/20
channel-group 1 mode active
description "C2970G-a"
mtu 9216
switchport mode trunk
exit
!
interface port-channel 1
switchport mode trunk
mtu 9000
Config on Cisco's:
interface Port-channel2
description PCM6220-a
switchport trunk encapsulation dot1q
switchport trunk allowed vlan 1-1001,1006-4094
switchport mode trunk
spanning-tree bpdufilter enable
!
interface GigabitEthernet0/21
description EC2-PCM6220-a
switchport trunk encapsulation dot1q
switchport trunk allowed vlan 1-1001,1006-4094
switchport mode trunk
channel-protocol lacp
channel-group 2 mode active
!
interface GigabitEthernet0/22
description EC2-PCM6220-a
switchport trunk encapsulation dot1q
switchport trunk allowed vlan 1-1001,1006-4094
switchport mode trunk
channel-protocol lacp
channel-group 2 mode active
!
interface GigabitEthernet0/23
description EC2-PCM6220-a
switchport trunk encapsulation dot1q
switchport trunk allowed vlan 1-1001,1006-4094
switchport mode trunk
channel-protocol lacp
channel-group 2 mode active
!
interface GigabitEthernet0/24
description EC2-PCM6220-a
switchport trunk encapsulation dot1q
switchport trunk allowed vlan 1-1001,1006-4094
switchport mode trunk
channel-protocol lacp
channel-group 2 mode active
DELL-Josh Cr
Community Manager
•
9784 Posts
•
45074 Points
3055
0
Posted April 24th, 2014 09:00
Either one of those could be the PDUs that it is seeing and causing the message in the logs, you could try turning them off temporarily and seeing if the messages stop, it probably is not the cause of the performance issues though.
I've disabled spanning-tree, but it didn't change a thing. The error logs still keep coming and the performance is still the same.
The VTP-mode cannot be changed, since there are already VLANs set. We highly rely on VLANs, so I cannot easily start over with a clean config.
DELL-Josh Cr
Community Manager
•
9784 Posts
•
45074 Points
3055
0
Posted April 25th, 2014 08:00
That’s fine, I don’t think the error messages are the cause of the performance issues. Are the switches up to date on firmware? Are there certain interfaces that are having more problems than others?
DELL-Josh Cr
Community Manager
•
9784 Posts
•
45074 Points
3055
0
Posted April 23rd, 2014 10:00
Hi,
The PDU they are referring to is Protocol data unit. So it is not recognizing something that is being sent, are there any proprietary Cisco features being used on the 2970Gs?