replacing hard disk

[prpr]

Why is this not then showing up in the diagnostics part of webif?

What version are you on? There was a fix in 1.5.2-6 for various SMART related things.

Also can just check that this drive still has plenty of life in it as it's only reallocated 112 sectors?

Possibly.
I recently had a 4TB drive (about 900MB in use) go from 0 bad sectors to 112 from one night's check to the next night's.
The following morning it had several thousands of bad sectors. I switched the machine off remotely as I was away.

When I got back I switched it on with a view to seeing what the damage was, and heard lots of clicking so quickly turned it off again.
It was clearly beyond my help, so I sent it for pro data recovery. The initial assessment was that they could get back 75% to 100% of the data.
Great I thought. Then they gave me the price. It was roughly £1200.
Having umm-ed and ahh-ed I decided to bite the bullet. Over two weeks later they told me they couldn't get anything back at all.
The only good part about it was that they didn't charge for the recovery attempt having failed to do it when they said they could.

I'm still mortified about the loss several weeks later (the first unrecoverable one I've ever had in well over 35 years), and I've only recovered a small amount of the material from other sources. I do know exactly what was lost as Remote Scheduling has the list of disk contents stuck in it.
I had multiple backup options available but, with everything else that's happened in the recent past, hadn't had time to make copies of stuff.

 

[bottletop]

..
Possibly.
I recently had a 4TB drive (about 900MB in use) go from 0 bad sectors to 112 from one night's check to the next night's.
The following morning it had several thousands of bad sectors. I switched the machine of remotely as I was away.
...

Out of curiosity, do you perform nightly checks? Are your regular checks looking at the SMART attributes or something else? Did you switch the machine of or off?

 

[prpr]

Out of curiosity, do you perform nightly checks?

The WebIf does it as part of automatic processing and RS uploads it to the server, which then generates alert emails in the morning.
Machines without RS just put it in the Disk stats. on the System Monitoring page and generate a banner alert, but not emails.

Are your regular checks looking at the SMART attributes or something else?

Yes, SMART.

Did you switch the machine of or off?

Off.

 

[bottletop]

...
Possibly.
I recently had a 4TB drive (about 900MB in use) go from 0 bad sectors to 112 from one night's check to the next night's.
The following morning it had several thousands of bad sectors. I switched the machine off remotely as I was away.
.......

Any idea what caused the jump from 0 to 112, then 112 to several thousands of bad sectors? Eg was the drive moved while spinning? Did you try running fixdisk to correct it?

 

[prpr]

No. As I said, I was away, so no chance of it being moved, unless by poltergeists or earthquakes.
It just suffered a catastrophic head crash for whatever reason.

 

[rodp]

What version are you on? There was a fix in 1.5.2-6 for various SMART related things.

Hmmm, I thought I had updated it but maybe not (I assume you are talking about Webif). Let me update that!



Is there an easy way to find out the difference between the two version?

 

[rodp]

Stats are back and working now. Current stats below. 120 hours of error free running - let's hope it will continue now. However I think it is time to see what dual drive NAS drives are going for a good price this Black Friday! Let me ask a quesiton about that in a different post!

(pictures as wouldn't let me copy and paste without it messing up the layout!)

 

[prpr]

I assume you are talking about Webif

It was in answer to your question "Why is this not then showing up in the diagnostics part of webif?", so I think it's fairly reasonable just to ask which version without having to qualify it further.

If you're going to ask questions about stuff, then 1) make sure you are up to date with upgrades or 2) at least quote the version you are using. I don't think this is too much to ask, but so many people don't do it.

Is there an easy way to find out the difference between the two version?

Depends on your definition of 'easy', but probably not. The changes are all on Git though.
1.5.1 is over 18 months old. There have been loads of fixes since then (106 commits to be precise). I don't know why people don't upgrade. If it's too much bother, then install the auto-update package and forget about it.