What did you do on your Synology NAS today?

[Pauljw]

Data Scrubbing completed! Oh my.

How often do you do the data scrubbing? I've done it once and it took days so I'm thinking quarterly, is that often enough do you think?

 

[Rusty]

Did some 10G speed tests with DS723+... Not bad for a 2core (4-thread) CPU with 2GB ram. copy transfers are even faster in write by about 200MB/s.

It is impressive to see such a small and silent device being so fast.

 

[Tedster]

How often do you do the data scrubbing? I've done it once and it took days so I'm thinking quarterly, is that often enough do you think?

I do it every 6 months, I don't like how it hammers the drives!
It'd wouldn't be so bad if it told you what its done! If anything?

 

[Pauljw]

Okay, thanks. I sort of understand what it's doing, but I agree, it doesn't seem to inform you if it found any issues along the way. As long as it ends on a positive note, I'll just assume it's doing it's job.

 

[Telos]

You should not need to scrub often. Error checking is performed each time a file is accessed. Only files rarely accessed require forced scrubs... particularly in such cases as just before you replace a failing drive, or upsize a drive, as these processes could end with a degraded volume. Once a drive fails, scrubbing is too late and one can only hope that the remaining copy (or parity version) is good. If drives never failed, you would not need to schedule scrubs.

 

[Gerard]

About a month or two ago, while poking around, I had noticed that one system was set to data scrub every 3 months, and check another which was set to every month. I changed that one to every 3 months too. Is the general consensus here that I can push this to every 6 months? Or is every 3 ok too. TIA

 

[Pauljw]

You should not need to scrub often. Error checking is performed each time a file is accessed. Only files rarely accessed require forced scrubs... particularly in such cases as just before you replace a failing drive, or upsize a drive, as these processes could end with a degraded volume. Once a drive fails, scrubbing is too late and one can only hope that the remaining copy (or parity version) is good. If drives never failed, you would not need to schedule scrubs.

Thank you, very informative.

 

[Telos]

the general consensus here that I can push this to every 6 months? Or is every 3 ok too

Ideally, for one drive redundancy arrays, you would complete a scrub just before drive failure... So your scrub frequency should be consistent with the thickness of your tinfoil hat (joking... but some truth there).

If you use SHR2 or RAID6, I'd think that an annual check would be more than sufficient.

 

[Tedster]

I've had it disabled for about 6 years!

 

[Gerard]

If you use SHR2 or RAID6, I'd think that an annual check would be more than sufficient.

Ok with that being said I have two setups. A rs820+ 4 drives running raid 10. Also have a ds720+ 2 drives running shr 1

 

[Telos]

Ok with that being said I have two setups. A rs820+ 4 drives running raid 10. Also have a ds720+ 2 drives running shr 1

There's no hard and fast rule. Factors that enter into that decision include the condition/age of your drives, their basic quality (Enterprise, NAS, PC, shucked...), and the length of time it takes to run the scrub. For me, it's monthly, but for you, it could be weekly, bimonthly, quarterly...

 

[Tedster]

Turned off the fan, that I modded onto the cpu of my 1515+. It's just too darn cold here !

 

[Tedster]

I repaired my 1515+ again! The replacement transistor failed again. It's been about a year since last time, so I can live with that.