On Fri, Jan 28, 2011 at 04:16:49PM -0600, Richard A Steenbergen scribbled: # On Fri, Jan 28, 2011 at 02:03:54PM -0800, Keith wrote: # > # > Currently the box is running 10.2R1.8. It has a MIC-3D 20 port card, # > MPC1, and RE-S-2000. # # Juniper just put out a tech bulletin this morning admitting the obvious, # that 10.2R1/R2/R3 and 10.3R1 for Trio (MPC) cards are massively broken # and shouldn't be used. Try 10.3R2, it's been mostly ok for us (not # counting the bug we hit the other day where all the MPCs crashed in an # endless loop after updating a prefix-list referenced in a firewall # filter on them, but at least so far this seems rare :P). Alas 10.3R3 # seems to be delayed, but you'll be far better off with 10.3R2 than you # will with 10.2R1 in the config above.
I actually find myself in the same situation as the OP, as I just powered up and started doing some initial configuration, testing, and most of all: some learning on our new MX480 (with MX-MPC2-3D), it seems to have shipped with 10.3R1.9 on it. Can you elaborate on the bustedness of this release at all, or bound to secrecy? I haven't run into anything weird yet, but I'm very early on in the process, only having some basic connectivity to a few LAN switches, and a logical system with a pair of IBGP sessions up inside it. Where should I expect oddity, as I was hoping to wait and get a better lab setup with it to try an ISSU to 10.3R2 to see how it goes (I also have GRES and NSR enabled..)? Thanks; -- Jonathan Towne _______________________________________________ juniper-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/juniper-nsp
