08-12-2014 07:48 AM
My environment consists of (4) 300 series switches (2) pair connected via ISL. The backend storage is NetApp Storage. the end points are various HP servers and UCS, tape drives and ESX farm. At random times (which always seem to be in the evening something happens where by the switch seems like it freezes (again I am calling it a switch freeze but I am certain that is not he term or perhaps not the switch). When I select the fast reboot then whatever was connected to that side goes back on lines such as VM's etc. Which is another thing. why are the machines and servers that go offline not route over to the ISL and utilize the "B' side.
Has anyone ever expierenced this before? This is becoming quite a pain to trouble shoot the root cause. I am not certain if its a server causing this issue and masquerading it as a switch issue etc. I am grasping in the wind.
If more info is needed to help diagnose let me know.. Any other hints, tips or thoughts would greatly be appreciated.
Fabric OS: v6.3.1a
Made on: Fri Feb 26 18:37:16 2010
Flash: Tue Dec 14 19:59:32 2010
08-13-2014 02:47 AM
First thing you should do is to upgrade FOS code to the latest 6.4 version at least, because the one installed on that switch is pretty old. Once done that we'll see how the switch behaves.
08-13-2014 01:14 PM - edited 08-13-2014 01:14 PM
According to the compatibility 6.4 is not compatible. We would have to jump to the 7.0 family.. argh. But why all these years it has been fine and for the past few months its having this issue.
08-14-2014 12:47 AM
maybe the hardware is starting to fails (check the output of 'errdump' and 'errdumpall') or maybe something in the config has changed recently or some feature of the switch has reached a certain threshold that makes it fail even though nothing had been modified...
In any case, a FOS upgrade is always a good start (btw, Brocade 300 does support 6.4). And when it comes to FOS 7, I am pretty happy with its stability (7.0.2c is the one have).