1 |
On Thu, Dec 07, 2017 at 09:52:55AM +0000, Richard Bradfield wrote: |
2 |
> On Thu, 7 Dec 2017, at 09:28, Frank Steinmetzger wrote: |
3 |
> > > I incorporated ZFS' expansion inflexibility into my planned |
4 |
> > > maintenance/servicing budget. |
5 |
> > |
6 |
> > What was the conclusion? That having no more free slots meant that you |
7 |
> > can just as well use the inflexible Raidz, otherwise would have gone with |
8 |
> > Mirror? |
9 |
> |
10 |
> Correct, I had gone back and forth between RaidZ2 and a pair of Mirrors. |
11 |
> I needed the space to be extendable, but I calculated my usage growth |
12 |
> to be below the rate at which drive prices were falling, so I could |
13 |
> budget to replace the current set of drives in 3 years, and that |
14 |
> would buy me a set of bigger ones when the time came. |
15 |
|
16 |
I see. I'm always looking for ways to optimise expenses and cut down on |
17 |
environmental footprint by keeping stuff around until it really breaks. In |
18 |
order to increase capacity, I would have to replace all four drives, whereas |
19 |
with a mirror, two would be enough. |
20 |
|
21 |
> I did also investigate USB3 external enclosures, they're pretty |
22 |
> fast these days. |
23 |
|
24 |
When I configured my kernel the other day, I discovered network block |
25 |
devices as an option. My PC has a hotswap bay[0]. Problem solved. :) Then I |
26 |
can do zpool replace with the drive-to-be-replaced still in the pool, which |
27 |
improves resilver read distribution and thus lessens the probability of a |
28 |
failure cascade. |
29 |
|
30 |
[0] http://www.sharkoon.com/?q=de/node/2171 |