Yes, you can use the "singleport lag enable" command to keep the lag up if only one link is left, this works well in the field. Without it you will need to set appropriate VLAN egress on the base ports forming the lag. Make sure all your firmware is right up to date though before you start.
When I first encountered LACP I also felt that I needed to use static lag's to achieve control over how they would form but couldn't get it to operate 100% predictably in the field, but was fine in the lab. I would strongly recommend following the procedure in the firmware feature guide on LACP for configuration. We make extensive use of this method across N,B,C,D and G series switches and with the "singleport" feature enabled and it works very well. I agree that you should always test the configuration by unplugging links too, and checking things fail over as you intended. My opinion is the static configuration should only be used when you need to form a link aggregation with another vendor's kit. On a side note I have to say the feature guides are excellent, keep up the good work Enterasys. Mark Lamond _____ From: Matt Clark [mailto:[email protected]] Sent: 20 September 2010 10:56 PM To: Enterasys Customer Mailing List Subject: RE: RE:[enterasys] Lacp config I agree. I have also seen instances where if there was a 2 port lag and one port died the lag would go away and if you had only tagged the lag traffic would stop. That is why I on the config I sent I tagged the ports in the lags and the lags themselves. However I have been told this has been fixed in recent firmware. Anybody test that? Matt Clark IT Engineer Pinal County Information Technology Office: 520.866.6650 | Mobile: 520.705.1575 | FAX: 520.866.2911 _____ Please take a moment to fill out the survey regarding this inquiry / experience with our department. http://pinalcountyaz.gov/Departments/InformationTechnology/Surveys/NetworkIn frastructure-CustomerSurvey.aspx --- To unsubscribe from enterasys, send email to [email protected] with the body: unsubscribe enterasys [email protected]
