Arista Switches rebooting

Saku Ytti saku at ytti.fi
Mon May 4 19:35:47 UTC 2020


On Mon, 4 May 2020 at 21:57, Matthew Petach <mpetach at netflight.com> wrote:

> As the process size in fabrication gets smaller and smaller, it takes less and less energy hitting a device to cause spurious events like these.

Agreed. However in this case it was not a single-event upset (or
multi-event upset) but timing problem, fixable with new software on
the memory controller.  Single-event upsets should be benign with ECC
memories.

-- 
  ++ytti



More information about the NANOG mailing list