> I'm inclined to agree that it's something about the Panda itself > rather than the software. I can't otherwise explain apparently random > dropouts unrelated to what is being done at the time. Today I returned > to the m/c after a couple of hours away to find a 'NetFetch ... must > exit' message. Yesterday it was a 'Message from H' in similar vein. > Also yesterday, a message "Error outside Impression Publisher: > Internal error: abort on data transfer at &20310D08". So it goes on, > generally trivial but sometines requiring a reboot. By contrast, the > Pi seems free of this behaviour. I would make a small comment that both you and Chris are, I believe, using CJE supplied Pandaboards. As far as I know, they were all old boards, not the more recent revisions that we've been using for about a year. Now, I'm not saying that ARMiniX is necessarily any different (I occasionally see spurious errors), but I've never heard anyone suffering data loss with an ARMiniX with a crash. For me, it's always been the case that (very) occasionally, I'll double click something, get a spurious error, then run it again without issue. Note that in terms of stability, our new i.MX6 machines are being deployed already as *servers*, ie. 24/7 operation. Andrew -- R-Comp 22 Robert Moffat, High Legh, Knutsford, Cheshire WA16 6PS Tel: 01925 755043 Fax: 01925 757377 http://www.rcomp.co.uk --- To alter your preferences or leave the group, visit //www.freelists.org/list/armini-support List-related queries to info@xxxxxxxxxxxx