[svlug] SSD was: Slides from the GoLUG meeting

Rick Moen rick at linuxmafia.com
Fri Jun 3 02:05:53 PDT 2016


Quoting Steve Litt (slitt at troubleshooters.com):

> For the last 5 years, I reboot every 2 to 4 weeks. Before then, as a
> power saving measure, I shut down every night and powered up every
> morning.

Those reboots didn't stress the hardware, of course -- whereas a
power-down followed by a power-up always does (the stress being applied
during power-up).  But I'm sure you're well aware of that.  I've heard
other people also observe that hard drives kept spinning as close to
24x7 as feasible tend to have relatively very long service lives (if not
overheated, etc.).

> The preceding is definitely a factor in my case. For 2 years we had no
> air conditioning (in central Florida). Also, I don't open my computers
> and vacuum them very often, so when I do, there's a lot of that
> dust-cotton. I now either keep the computers off the floor, or
> separate them from the carpet with wooden planks. 

I had to learn the hard way (but before catastrophic loss) to just get
them well off the floor.  Dust bunnies blocking vents or clogging fans
becomes yet another quick way to kill components.  And then you'd feel
really stupid for letting that happen.

> Also, before 2 years ago, I had nothing but 80mm fans. Everything ran
> hot.

The Antec and similar 80mm fans actually fail to suck, FWIW.  But indeed
they're sort-of worst-case cooling dictated by case height of 1U and 2U
units -- obliged to spin really fast, which alone makes them both
failure-prone (friction) and noisy (turbulence).

> And that's just the point. Except for Seagates and ancient IBM
> Deathstars, the majority of my hard drives retired in perfect working
> order.

Well done, sir.

I should have done the same, and have kept some inexcusably old, tiny,
and decrepit SCSI drives running in my servers for inadequate reasons,
over the years.




More information about the svlug mailing list