My current unraid setup is running on a system that is nearly 10 years old; but it's been rock solid. It's one of the first 3.2ghz XEON intel processors on an Intel desktop board.

 

I have:

4x500gb IDE disks attached to the motherboard.

2x500gb SATA disks attached to the motherboard

1x300gb SATA on PCI card

1x80gb SATA on PCI card (cache disk)

 

For some reason I have stacks of 500gb drives laying around. 

 

Am I going to commit storage suicide if I plan on moving all of these drives to another motherboard/PCI card? I don't have a problem moving everything off to another set of disks temporarily but that's a pain... What would you guys do?

Hey Aaron,

 

As long as everything is re-assigned back to its original position (take a screenshot of your unraid "disks" webui) then you should be able to upgrade your setup with minimal hassle. No data shuffling required.

 

What is the reason for upgrading? Performance, stability, expansion, cooling or just to modernize? I imagine a 10-year old setup is probably headed for disaster with swollen caps and a lint farm full of creepy-crawlies.

 

Can you supply some more specific information about your current setup? In particular:

- Case / PSU

- CPU/s model#

- Motherboard model#

- Memory type/amount

I'd build a whole new setup if all your stuff is 10 years old.

 

If the HDDs are about that old too I'd get new drives. To my calculations you have just under 3TB of data.

 

Get 3 2TB HDDs start from scratch and transfer.

 

Getting IDE support on some of the newer boards can be difficult.

 

Josh

 

And if you end up freeing the smaller drives, you could check to see if Rajahal is still trading a larger drive for several smaller ones

http://lime-technology.com/forum/index.php?topic=16753.0

I would also start budgeting for a new setup.

Only the mobo+cpu are that old. They're rocking. I've always kept them cool and clean. I don't expect them to die any time soon. Most of the drives are new (except a few of the IDE drives)

 

Sounds like a mobo swap should be minimal hassle (found a wiki article explaining it. No biggie.) I don't plan on ordering new drives any time soon. I literally have stacks of 500gb sata drives to use up.  (If I were going to go the 2tbx3 route I'd go with something FreeNAS to get better performance and ZFS compression)

 

Going to grab a few of these: http://www.monoprice.com/products/product.asp?c_id=104&cp_id=10407&cs_id=1040702&p_id=2530&seq=1&format=2

 

and a new mobo/proc on newegg.

Going to grab a few of these: http://www.monoprice.com/products/product.asp?c_id=104&cp_id=10407&cs_id=1040702&p_id=2530&seq=1&format=2

 

and a new mobo/proc on newegg.

 

Be careful with that controller, someone on here recently had issues when two drives were accessed simultaneously.

In that situation, the controller was not providing the correct data back to the host.

It only showed up when the drives were used simultaneously.

 

Instead of a few of those controllers consider the Rosewill RC-218.

Going to grab a few of these: http://www.monoprice.com/products/product.asp?c_id=104&cp_id=10407&cs_id=1040702&p_id=2530&seq=1&format=2

 

and a new mobo/proc on newegg.

 

Be careful with that controller, someone on here recently had issues when two drives were accessed simultaneously.

In that situation, the controller was not providing the correct data back to the host.

It only showed up when the drives were used simultaneously.

 

Instead of a few of those controllers consider the Rosewill RC-218.

 

I would be curious to see a link to that issue if you come across it again.  I've used countless SIL3132 controllers from monoprice and have only seen one or two DOAs, and never any other issues.

Going to grab a few of these: http://www.monoprice.com/products/product.asp?c_id=104&cp_id=10407&cs_id=1040702&p_id=2530&seq=1&format=2

 

and a new mobo/proc on newegg.

 

Be careful with that controller, someone on here recently had issues when two drives were accessed simultaneously.

In that situation, the controller was not providing the correct data back to the host.

It only showed up when the drives were used simultaneously.

 

Instead of a few of those controllers consider the Rosewill RC-218.

 

I would be curious to see a link to that issue if you come across it again.  I've used countless SIL3132 controllers from monoprice and have only seen one or two DOAs, and never any other issues.

 

http://lime-technology.com/forum/index.php?topic=17823.msg160418#msg160418

There was a bunch of diagnosing over a few days.

Thanks Weebo.  I skimmed the thread and didn't see any mention of the SIL3132 cards from monoprice.  Hoopster was using Syba and Rosewill cards.

Thanks Weebo.  I skimmed the thread and didn't see any mention of the SIL3132 cards from monoprice.  Hoopster was using Syba and Rosewill cards.

 

Ok that must be the wrong thread I was involved in.

There was another that was corrupting data while both drives were accessed.

One drive at a time was fine, two drives at a time would reveal corruption.

That thread, had a link to another thread which exhibited the same symptoms.

 

I have 2 SIL3132 cards and have not seen that problem, But they were SYBA cards, not the monoprice cards. I'll try and poke around s'more.

Look into this thread.

http://lime-technology.com/forum/index.php?topic=17641.msg158817#msg158817

 

Thanks again Weebo.  That thread does indeed point out the Monoprice SIL3132 controller as the culprit.  I would suggest that ixnu was unlucky and just got a bad one, but he does point to others having similar issues in different systems as well.  I always test every motherboard I use with the monoprice SIL3132 cards as well as the SASLP cards.  Since they generally pass my tests (which included parity checks), I will continue to trust them.  However, I will keep this issue in mind.  Thanks for bringing it to my attention!