07-15-2008 07:40 PM
This is a basic question - not too technical at all. I am in the midst of setting up fabric alerts. Firstly on a Hitachi software application (Hitachi's Tuning Manager) then later on the switch GUI's. Now I can set up alerts - no problems.
My alerts are looking for what I term the basic diagnotic checks - that is CRC errors, invalid word errors, link failures, delimeter errors,etc, etc. I got most of these condictions from the Brocade manual.
My question - as in what happened recently - is I got errors - a series of link reset counts, link failures, zero buffer credits, invalid order counts on some of my eneterprise 48k switches. Now what - where do you look up the errors that provide more than a description. In laymen terms - what does it mean to encounter a high level of zero buffer credits, or what does it mean to have delimeter errors, etc. One thing to trap the exception or alert - another thing to determine what it means and what to do.
I would appreciate anyone pointing me to a guide/reference material. I have a medium size SAN / fabric infrastructure including DR and several 48k switches. Alerting is important - but consequent actions are even more important.
07-16-2008 03:15 AM
07-16-2008 03:21 AM
Also check the Fabric OS message reference guide
Maybe if you are HDS client, try creating a extranet account on the HDS website and the tuning manager guide might also give you some additional references.
Visit www.hds.com -> Click on PartnerXchange Login and register with your official email id. Once done await confirmation from HDS (takes 1-2 business days).
Hope this helps.