Hello to the group.  It's been a while since I've been active so sorry for the silence.

 

1.  I have two Hitachi 2 TB drives HDS722020ALA330 that have to be from 2007 ish.. they seem to be plugging along fine, but.. pretty old as drives go and making me nervous.  Would you replace functional drives based on age?  I know occasionally hear some noisy drives in my server, but it would be pretty time consuming and difficult to narrow down where its' coming form. 

 

2.  I have 2 Seagate 3 TB drives in use and a 3rd warm spare ST3000DM001 (one is the parity drive).  Back when i bought them, they had solid reviews, but now seem to be failing for a number of people after 1 yr.  Would you keep those in service for long?  any good/bad experience?  anyone with them running strong after 1-2 years?  I have older drives in the array, but dont want to be sitting on a time bomb.

 

3.  Considering a Hitachi 3 TB H3IKNAS30003272SN as a replacement.  less sales as its a touch more expensive (not enough to worry about) but not many reviews either.  Good or bad here?

 

thanks again for the help

Hello to the group.  It's been a while since I've been active so sorry for the silence.

 

1.  I have two Hitachi 2 TB drives HDS722020ALA330 that have to be from 2007 ish.. they seem to be plugging along fine, but.. pretty old as drives go and making me nervous.  Would you replace functional drives based on age?  I know occasionally hear some noisy drives in my server, but it would be pretty time consuming and difficult to narrow down where its' coming form. 

No. I would suggest you run SMART reports on all your drives to look for signs of failure. Age is not a good indicator.

 

2.  I have 2 Seagate 3 TB drives in use and a 3rd warm spare ST3000DM001 (one is the parity drive).  Back when i bought them, they had solid reviews, but now seem to be failing for a number of people after 1 yr.  Would you keep those in service for long?  any good/bad experience?  anyone with them running strong after 1-2 years?  I have older drives in the array, but dont want to be sitting on a time bomb.

 

See answer above. Good or bad experiences by others is an important consideration when buying drives, but if you own them you need to monitor them for bad and pending sectors, and other sign of failure.

 

3.  Considering a Hitachi 3 TB H3IKNAS30003272SN as a replacement.  less sales as its a touch more expensive (not enough to worry about) but not many reviews either.  Good or bad here?

 

thanks again for the help

 

They may not be the very fastest, but Hitachi (now HGST) drives are among if not the most reliable drives you can buy. They are awesome IMO.

They may not be the very fastest, but Hitachi (now HGST) drives are among if not the most reliable drives you can buy. They are awesome IMO.

I agree completely.  They are the most reliable drives I have and I have WD Red 2,3,4TB, Seagate ST3000DM001, ST4000DM000, Toshiba DT01ACA300 3TB.

Thanks for the feedback.  I do now and then run a smart test, and haven't seen anything show up previously.  I will run them again tonight just to be sure. 

 

Reliability is 1st priority always, but on the speed front if I have a faster parity drive, would these being slower really affect my write time /parity check times to any consequential amount? 

 

What is the best drive out there to use for parity, ie speed and reliability?

 

 

Sent from my iPad using Tapatalk

 

They may not be the very fastest, but Hitachi (now HGST) drives are among if not the most reliable drives you can buy. They are awesome IMO.

I agree completely.  They are the most reliable drives I have and I have WD Red 2,3,4TB, Seagate ST3000DM001, ST4000DM000, Toshiba DT01ACA300 3TB.

 

Ever have issues with the Seagate ST3000DM001's?  How long have you had them?

 

 

Sent from my iPad using Tapatalk

 

They may not be the very fastest, but Hitachi (now HGST) drives are among if not the most reliable drives you can buy. They are awesome IMO.

I agree completely.  They are the most reliable drives I have and I have WD Red 2,3,4TB, Seagate ST3000DM001, ST4000DM000, Toshiba DT01ACA300 3TB.

 

Ever have issues with the Seagate ST3000DM001's?  How long have you had them?

 

 

Sent from my iPad using Tapatalk

Yes.  I have variously aged drives from 1 year+ to 1 month.  I have 7-10 of them sitting on my living room floor to be RMA'd if they are still in warranty - one to two may not be.  Thought they were going to be good drives but they lived up to the bad opinion I have had of Seagate drives for years.  Been buying WD, Samsung & Hitachi but it had been a while so I got some and they appeared to work well and no DOA's so thought the quality had improved and bought more.  But I should have remembered my problems from around 200GB or smaller drive days and the fact that they die without warning.  Smart report on some of the ST3000DM001's actually showed problems (pending and reallocated) and I removed them immediately but there was more than one ST3000DM001 that just died - no longer recognized by OS and/or PC bios and no smart errors stood out before it happened.  The Seagate's (around the 200GB days or before) is where I got such a low opinion of smart reports on drives.  I don't think I got a single warning before the drive died back then.  I will not be buying another ST3000DM001 even though when working they perform very well.  If I get those bad ones RMA'd to Seagate I will have plenty of backups available for the 15-18 drive array that is all Seagate ST3000DM001s.  Once I burn through those (if I do - maybe I've got ones that will last now) they will be replaced with HGST cool spins.

Thanks for the feedback.  I do now and then run a smart test, and haven't seen anything show up previously.  I will run them again tonight just to be sure. 

 

Reliability is 1st priority always, but on the speed front if I have a faster parity drive, would these being slower really affect my write time /parity check times to any consequential amount? 

 

What is the best drive out there to use for parity, ie speed and reliability?

 

 

Sent from my iPad using Tapatalk

 

If you are routinely writing to your array from multiple sources, and/or you have 7200 RPM drives in your array, you will get a small performance benefit from a 7200.RPM parity drive with high aerial density. It you want to get fancy you can create a RAID0 pair with a hardware controller that boosts unRaid performance a little more. The cache drive gives you a way to defer the write penalty regardless of your parity disk. But nothing you do is going to make unRaid writes to the protected array fast and mostly you just live with the performance hit knowing that is the price of unRaid's redundancy.

 

So my advice is to pick your parity drive the same way you pick other drives. Reliability is king.

short smart reports:

 

I went ahead and ran new ones on all my drives.  The first 2 are the older 7200 Hitachi's.  The next 3 are the 3TB Seagates, and then the rest.  I dont seen re-allocated, or pending counts on any of them, other seek error rates are high numbers but never understood how relevant a lot of the other values are. 

 

Anyone see anything to worry about?  Would you replace any of these? 

 

Thanks for any further advice.

 

smart_test_results.txt

With Seagates especially, when you see high raw values you need to look at the other values (VALUE, WORST & THRESHOLD) for that smart attribute to see if the value is approaching the threshold.  If not then that smart attribute is ok.  Or if you have seen significant changes in the value field towards the threshold recently.  But like you said there were no pending or reallocated sectors and I didn't see any other attributes approaching their respective threshold.  So your drives should be OK but I would trust bjp999's opinion more then mine.