14 Oct
2022
14 Oct
'22
10:19 p.m.
On 14/10/2022 21:22, Andy Smith via BitFolk Users wrote:
I wouldn't do single parity with devices as big as
6TB, especially
not at Hetzner who are very likely using consumer HDDs (not ones
recommended to enterprise/NAS use), and often you're not the first
user of them (i.e. they took them out of a server that was rented by
others previously).
Their business model is to auction off previously-used servers,
it being cheaper to flog it off as-is rather than send a techie
into the warehouse to pull disks out. Hence the availability of
machines with "15 x 4.0 TB Enterprise HDD" at ~90 euros per
month.
FWIW - current machine uses "HGST Ultrastar 7K6000" which appear
to be "proper". However it looks like they have been running
for almost six years. So I am advising my customer to move.
The issue is that if you lose one device and then
replace it, it
will take a really long time to read 6TB of data off of the other
devices and during that whole time there is no redundancy at all, so
if there are any further unreadable bits on other devices, the data
is lost.
Well yes, I have sat through a rebuild and it was not pleasant.
The risk is a little bit lower if you have been
regularly scrubbing
your array as that forces it to read everything and find any
unreadable bits. Most Linux distributions using MD RAID do schedule
this for once per month.
Could that be why it was doing an unexpected rebuild the other day?
Nothing in /etc/cron.monthly.
<snip>
So if I convince my customer to move server to something with more
than, say, seven, spinny things what RAID should I be looking at?
Massive thanks all, in particular Andy Bennett, I have read your emails,
Andy Smith: advising me when it's not even your hosting I am talking
about!
thanks
Dom