1 |
On Oct 6, 2011 9:06 PM, "Mark Knecht" <markknecht@×××××.com> wrote: |
2 |
> |
3 |
> On Thu, Oct 6, 2011 at 1:28 PM, Michael Mol <mikemol@×××××.com> wrote: |
4 |
> > On Thu, Oct 6, 2011 at 4:21 PM, Mark Knecht <markknecht@×××××.com> |
5 |
wrote: |
6 |
> >> On Thu, Oct 6, 2011 at 1:03 PM, Paul Hartman |
7 |
> >> My worry was that if the mdraid daemon saw one drive gone - either |
8 |
> >> when starting to spin down or when one spins up slowly - and if mdraid |
9 |
> >> didn't understand that all this stuff was taking place intentionally |
10 |
> >> then it might mark that drive as having failed. |
11 |
> > |
12 |
> > Does mdraid even have an awareness of timeouts, or does it leave that |
13 |
> > to lower drivers? I think the latter condition is more likely. |
14 |
> > |
15 |
> > I suspect, though, that if your disk fails to spin up reasonably |
16 |
> > quickly, it's already failed. |
17 |
> > |
18 |
> |
19 |
> In general I agree. However drives that are designed for RAID have a |
20 |
> feature known as Time Limited Error Recovery (TLER) which supposedly |
21 |
> guarantees that they'll get the drive back to responding fast enough |
22 |
> to not have it marked as failed in the RAID array: |
23 |
> |
24 |
> http://en.wikipedia.org/wiki/Time-Limited_Error_Recovery |
25 |
> |
26 |
> When I built my first RAID I bought some WD 1TB green drives, built |
27 |
> the RAID and immediately had drives failing because they didn't have |
28 |
> this sort of feature. I replaced them with RAID Edition drives that |
29 |
> have the TLER feature and have never had a problem since. (Well, I |
30 |
> actually bought all new drives and kept the six 1TB drives which I'd |
31 |
> mostly used up for other things like external eSATA backup drives, |
32 |
> etc...) |
33 |
> |
34 |
> Anyway, I'm possibly over sensitized to this sort of timing problem |
35 |
> specifically in a RAID which is why I asked the question of Paul in |
36 |
> the first place. |
37 |
|
38 |
My first RAID was with three Seagate economy 1.5TB drives in RAID 5, shortly |
39 |
followed by three 1TB WD black drives in RAID 0. I never had the problems |
40 |
you describe, though I rebuit the RAID5 several times as I was figuring |
41 |
things out. (the 3TB RAID0 was for some heavy duty scratch space.) |