NASLite Network Attached Storage

www.serverelements.com
Task-specific simplicity with low hardware requirements.
It is currently Mon Dec 02, 2024 10:37 pm

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 1 post ] 
Author Message
PostPosted: Fri Sep 28, 2012 10:07 pm 
Offline

Joined: Fri Feb 25, 2005 11:50 pm
Posts: 139
Ha! Building my first 16TB NAS on an old Asus Terminator C3 POS :)


But Tony, Ralph....these things are so big now - it takes eons to prep/check em. Would be nice if we could break such tasks up into little pieces which run on any CPU cycles they can collect.

Then there are the issues of these WD power cycling drives. They seem to work - but at what cost to reliability? Load/unload counts are high.

Anyone who has actual read Google's drive research paper is now justifiably scared. So what can Server Elements do to mitigate our pain?

How about some better way to "surf" the disk pricing trends? Suspect a lot of us have drives of all sizes...so what can be done to extract the last bit of use out of these things to reduce our costs? Say after three years of use on the counter - turning them into backup drives?


Here's an example of new troubles:

After that 16TB NAS gets built - plan to use it to backup existing 8TB and 6TB servers. Piece of cake backing up the first unit. Just turn on the mirroring function - and copies appear. Now, after turning rsync off - how do I efficiently get the 6TB server copied over too? Especially since it contains one of those new fangled drives which is clearly failing? Sure I can do it with a client? Why would I want to? If you had a system which would rsync to an assigned folder - I'd never need to.

We need an easy scheme. One which takes into account surfing prices, warranties, and power use. And if you've taken that Google data to heart...you also want to identify then kill off any bad disk specimens vendors sell you. A primo disk will last almost a decade according to the counter - but many will fail after just a few thousand hours. What can server elements do to help us find these critters? Then perhaps isolate the bad block areas and keep that head on top of them.....so the manufacturer supplied warranty test stays happy....

These issues have become a sufficient pain in the ass for me to start manually collecting model, serial, hour , reallocated counts for a spreadsheet. Sure would be useful if that report generated itself.


Big drives fail. The simplest Google search will demonstrate that. Help us!


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 1 post ] 

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group