Initial research seems to indicate it may be something related to a recent network infrastructure change. As part of that discussion, the configuration options for LACP and PAGP were mentioned. Those configuration options were not previously familiar to me - so I spent a quick 15 minutes doing some research and found these resources to be of some interest:
Some background info on LACP and PAGP
VMware Virtual Networking Concepts
Link Aggregation Control Protocol (LACP) (802.3ad) for Gigabit Interfaces
For Catalyst 2950 and Catalyst 2955 Switch
Configuring EtherChannels
“Interfaces configured in the on mode do not exchange PAgP or LACP packets.”
“Interfaces can form an EtherChannel when they are in different PAgP modes as long as the modes are compatible”
Default EtherChannel Configuration
Table 30-2 shows the default EtherChannel configuration.
EtherChannel Configuration Guidelines
“Do not configure a GigaStack GBIC port as part of an EtherChannel.”
“Configure all interfaces in an EtherChannel to operate at the same speeds and duplex modes.”
“Do not configure a secure port as part of an EtherChannel.”
PAgP is cisco proprietary which has 2 implications:
1-Interoperability with other network gear from different vendors is not guaranteed.
2-More importantly, your vmware servers may not be able to leverage the etherchannel
1-Interoperability with other network gear from different vendors is not guaranteed.
2-More importantly, your vmware servers may not be able to leverage the etherchannel
“Cisco's recommandations for VMWare connection is to use the mode "on", not LACP or PAGP”
“ESX does not support PAgP or LACP, you need to set Etherchannel to NONEGOTIATE”
“I need to get in the habit of using LACP. I've been using "on" exclusively and it's bitten me a couple times.”
“Channeling unconditionally can cause a loop if the other side isn't configured.”
For ports to be members of the same EtherChannel, there are some restrictions. Ports must:
- Belong to the same VLAN
- Have identical STP settings
- Have identical speed/duplex settings
- Note: In addition, if the EtherChannel is to be used as a trunking interface, all ports must be in trunking mode, have the same native VLAN, and pass the same set of VLANs.
EtherChannel Negotiation: PAgP vs. LACP
There
are two EtherChannel negotiation protocols. Port Aggregation Protocol
(PAgP) is a Cisco-proprietary protocol, while Link Aggregation Control
Protocol (LACP) is standards based.
PAgP
dynamically modifies the EtherChannel if one of the ports’ VLAN, speed,
etc. is changed so that all of the links in the EtherChannel match.
PAgP can be configured in active mode (desirable), which actively
attempts negotiation. Passive mode (auto, the default) only negotiates
an EtherChannel if the far end initiates it.
LACP assigns roles to end points. The switch with the lowest system priority
makes decisions about what ports will participate in the EtherChannel
at any given time. If you’re familiar with STP, this is similar to the
way the Root Bridge is elected. Ports are selected and become active in
the EtherChannel according to their port priority. LACP Active mode (active) – actively negotiates, while passive mode (passive) negotiates only if the far end initiates it.
Etherchannel Performance Issues with….
“Recently there was a post on Linkedin
about port aggregation and the use of negotiation. It’s really
disturbing seeing how many people recommend not using LACP or PaGP to
configure aggregated ports (port-channels|etherchannels). Seems to be
common misconception that if two switches are back-to-back configuring
an on mode is acceptable.”
