My recollection is VMware doesn't do LACP unless you use the vSphere Distributed Switch.
I would hazard to guess that you are using vSphere Standard Switches, which will NOT negotiate LACP. You need to do a static trunk. I spend most of my time in the Cisco world, and what you are looking for is the HP equivalent of a "static EtherChannel". I this VMware KB article will help: http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1004048 At several points, that KB article has this note: Note: LACP is only supported in vSphere 5.1 and 5.5, using vSphere Distributed Switches (VDS) or the Cisco Nexus 1000v. The part for HP switches is at the bottom; if you know anything about Cisco, the Cisco configuration examples have you use a static EtherChannel rather than a dynamic (AKA LACP) EtherChannel. -- Phil Brutsche [email protected] From: [email protected] [mailto:[email protected]] On Behalf Of Jake Gardner Sent: Monday, May 12, 2014 3:28 PM To: [email protected] Subject: RE: [NTSysADM] VMware, HP, and Trunking Those are the vSwitch settings I have. My 5400zl has the three trunks configured manually for the esxi servers and each assigned a trk# group. I have 4 other 4GB Dynamic agg links to 2900 series switches. Here’s a snippet of a SHOW LACP from my 5400. B1 Active Dyn5 Up Yes Success 0 0 B2 Active Dyn5 Up Yes Success 0 0 B3 Active Dyn5 Up Yes Success 0 0 B4 Active Dyn5 Up Yes Success 0 0 B5 Active Trk4 Up No Success 0 293 B6 Active Trk4 Up No Success 0 293 B7 Active Trk5 Up No Success 0 294 B8 Active Trk5 Up No Success 0 294 Thanks, Jake Gardner IT Administrator 267-352-2020 Ext. 246 www.ttcdas.com<http://www.ttcdas.com/> From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of Matthew W. Ross Sent: Monday, May 12, 2014 2:12 PM To: [email protected]<mailto:[email protected]> Subject: [NTSysADM] VMware, HP, and Trunking Okay, I've hit a wall. Short story setup: We have 3 VMware ESXi hosts. They connect to our network via a HP Procurve 2510G-48 at the top of the rack. Each has a pair of Gigabit ethernet connection connections going to that switch, setup as a trunk. We have been dealing with some minor latency issues, mostly the occasional timeout on our Mail server's web interface. My current thought is that the 2510G-48 might not be able to handle the load... So I've moved the three servers' uplinks to our 5308xl as a test to see if these timeouts disappear. The move went smoothly, except on the 3rd host: If I have both Gigabit connections enabled on the 5308xl, I cannot ping Host 3. As soon as I disable one of the two ports on the 5308xl (either one), it pings. Switch Configuration: Port H1-H2 Trnk-1 TRUNK Port H3-H4 Trnk-2 TRUNK Port H5-H6 Trnk-3 TRUNK Each has Spanning Tree Priority 4. The vSwitch on all the hosts are setup the same way: Load Balancing: Route based on IP hash Network Failure detection: Link Status only Notify Switches: Yes Failback: Yes If I can't figure this out, I'll be contacting my vendor for support. Thanks for any ideas. --Matt Ross Ephrata School District ***Teletronics Technology Corporation*** This e-mail is confidential and may also be privileged. If you are not the addressee or authorized by the addressee to receive this e-mail, you may not disclose, copy, distribute, or use this e-mail. If you have received this e-mail in error, please notify the sender immediately by reply e-mail or by telephone at 267-352-2020 and destroy this message and any copies. Thank you. *******************************************************************
