Sometimes it's the hardware

My desktop has been extra “crashy” the last few weeks and I finally relented and started a MemTest86 this morning…


Time to buy new memory :grimacing:

Remember to buy ECC memory if you’re building a server for Ignition…


6th Gen Intel? New RAM justifies a thread-ripper too :smiley:

Gonna hold out for Zen 4 / Ryzen 6000… hopefully next year? AMD supports ECC memory too :laughing:

1 Like

I recently read some articles about people being pissed off at Intel for not supporting ECC, and how much productivity probably has been lost over the years because of it.

I’m planning on upgrading my TrueNAS RAID system this year, but even I’m torn between ECC vs non-ECC systems (mainly because of the price). Currently I’m running on a non-ECC HP Proliant micro server, and that hardware is sweet for what it does (it just do what I want it to do) and I haven idea if non-ECC has caused me any issues :thinking:

A long time ago (but in this galaxy), I ran a NetWare server on a non-ECC workstation platform, with the mobo packed to the gills with all the RAM it could take. After a couple years of random crashes at long intervals, about three per year, I looked up the DRAM specs. (Back then you could actually find the transistor versus die dimensions.) Anyways, in that fragile age, I calculated the local natural background radiation (gamma) would flip a bit at least that often. With just parity checking, the mobo response was die! ):

ECC for me thereafter, at least for servers.

1 Like

I live at 7000 feet, so my radiation levels must be higher by some degree :thinking:. And I also live in an area that has Radon issues (although I had my house thoroughly checked before I bought it)