Overclock.net › Forums › Components › Hard Drives & Storage › RAID Controllers and Software › PERC 5/i RAID Card: Tips and Benchmarks
New Posts  All Forums:Forum Nav:

PERC 5/i RAID Card: Tips and Benchmarks - Page 470

post #4691 of 7150
Quote:
Originally Posted by dantm View Post
For example this:

http://cgi.ebay.com/DELL-MY412-SAS-P...item1c151af7e8

I don't think this is a PERC, it's probably a SAS as you mentioned?

Thanks!
yep - that's a SAS 5/iR.

There is also a SAS 6/iR that looks similar to the PERC 6/i - the easy way to tell them apart is the 6/iR has no heatsink.
post #4692 of 7150
Quote:
Originally Posted by the_beast View Post
yep - that's a SAS 5/iR.

There is also a SAS 6/iR that looks similar to the PERC 6/i - the easy way to tell them apart is the 6/iR has no heatsink.
Thanks, you just helped me from making a moderately-priced mistake ;-)
post #4693 of 7150
double checking:

http://cgi.ebay.com.au/32-Pin-SAS-Co...item2c567b0106

^ this is correct cable? (item #: 190429462790)
post #4694 of 7150
Quote:
Originally Posted by tiro_uspsss View Post
double checking:

http://cgi.ebay.com.au/32-Pin-SAS-Co...item2c567b0106

^ this is correct cable? (item #: 190429462790)
Yep, if you want regular SATA and not eSATA.
PWNzershreck
(15 items)
 
  
CPUMotherboardGraphicsRAM
4930K @ 4.6 GHz ASUS Rampage IV Black Edition MSI GTX 1080 FE Heatkiller Acetal 16 GB Corsair Vengeance 1600C9 
Hard DriveOptical DriveCoolingOS
2x Samsung 840 Pro  ASUS DVD-RW SATA Koolance 380i & 2x HW Labs 480GTX Arch Linux x86_64, Windows 7 x64 
MonitorKeyboardPowerCase
LG UC88-B Ultrawide, ASUS VS278Q Ducky Corsair AX1200i Caselabs STH10 
MouseMouse PadAudio
Logitech G500 Func 1030 ASUS Xonar Essence STX 
  hide details  
Reply
PWNzershreck
(15 items)
 
  
CPUMotherboardGraphicsRAM
4930K @ 4.6 GHz ASUS Rampage IV Black Edition MSI GTX 1080 FE Heatkiller Acetal 16 GB Corsair Vengeance 1600C9 
Hard DriveOptical DriveCoolingOS
2x Samsung 840 Pro  ASUS DVD-RW SATA Koolance 380i & 2x HW Labs 480GTX Arch Linux x86_64, Windows 7 x64 
MonitorKeyboardPowerCase
LG UC88-B Ultrawide, ASUS VS278Q Ducky Corsair AX1200i Caselabs STH10 
MouseMouse PadAudio
Logitech G500 Func 1030 ASUS Xonar Essence STX 
  hide details  
Reply
post #4695 of 7150
Quote:
Originally Posted by binormalkilla View Post
Yep, if you want regular SATA and not eSATA.
thanks muchly!
post #4696 of 7150
I just need suggestion for 1.5TB drives to expand my array. What's the best right now ?

Thanks
post #4697 of 7150
Quote:
Originally Posted by fayed View Post
I just need suggestion for 1.5TB drives to expand my array. What's the best right now ?

Thanks
Western Digital Blacks

Seagate has the 7200.11's, but the WD's use larger platters (and thus are faster and create less heat.)

Everything else is 5400-5900RPM drives.
Test Bed
(24 items)
 
TTTbox
(15 items)
 
 
CPUMotherboardGraphicsRAM
Core i5 3570K  Biostar TZ774XE Power Color AX5770 1GB 4x4GB 1866Mhz 9-10-9-27-2T 
Hard DriveHard DriveHard DriveHard Drive
Samsung Spinpoint F3 1TB Samsung Spinpoint F3 1TB Western Digital Green 2TB Samsung Spinpoint M4 500GB 
Hard DriveOptical DriveCoolingOS
Intel X-25M G2 80GB Lite-On BluRay Drive Arctic Cooling Freezer 13 Pro Windows 7 Ultimate 
MonitorKeyboardPowerCase
Dell U2212HM Deck Legend Frost Tactile Silverstone Strider Plus 750w Custom Acrylic Test Bed 
MouseMouse PadAudioAudio
Logitech G500 Cooler Master Storm Tactics FPS Beyer Dynamics DT-990 600ohm Samson MediaOne 3A 
AudioAudioOtherOther
FiiO E09k QOGIR FiiO E17 ALPEN Microsoft Lifecam Studio Blue Yeti Pro Microphone 
CPUMotherboardGraphicsRAM
AMD A10-A5800K ASRock FM2A85X-ITX Radeon HD7770 1GB 2x8GB 1866mhz C9 
Hard DriveHard DriveOptical DriveCooling
SanDisk Ultra Plus 128GB 2TB 5400 RPM  24x DVD Drive Noctua L9A 
OSMonitorKeyboardPower
Windows 7 Home Premium 32" IPS TV Logitech K400 Seasonic G360 
CaseMouseAudio
Cooler Master Elite 120 Advanced Logitech K400 Trackpad Custom 2.1 Home Theater Setup 
  hide details  
Reply
Test Bed
(24 items)
 
TTTbox
(15 items)
 
 
CPUMotherboardGraphicsRAM
Core i5 3570K  Biostar TZ774XE Power Color AX5770 1GB 4x4GB 1866Mhz 9-10-9-27-2T 
Hard DriveHard DriveHard DriveHard Drive
Samsung Spinpoint F3 1TB Samsung Spinpoint F3 1TB Western Digital Green 2TB Samsung Spinpoint M4 500GB 
Hard DriveOptical DriveCoolingOS
Intel X-25M G2 80GB Lite-On BluRay Drive Arctic Cooling Freezer 13 Pro Windows 7 Ultimate 
MonitorKeyboardPowerCase
Dell U2212HM Deck Legend Frost Tactile Silverstone Strider Plus 750w Custom Acrylic Test Bed 
MouseMouse PadAudioAudio
Logitech G500 Cooler Master Storm Tactics FPS Beyer Dynamics DT-990 600ohm Samson MediaOne 3A 
AudioAudioOtherOther
FiiO E09k QOGIR FiiO E17 ALPEN Microsoft Lifecam Studio Blue Yeti Pro Microphone 
CPUMotherboardGraphicsRAM
AMD A10-A5800K ASRock FM2A85X-ITX Radeon HD7770 1GB 2x8GB 1866mhz C9 
Hard DriveHard DriveOptical DriveCooling
SanDisk Ultra Plus 128GB 2TB 5400 RPM  24x DVD Drive Noctua L9A 
OSMonitorKeyboardPower
Windows 7 Home Premium 32" IPS TV Logitech K400 Seasonic G360 
CaseMouseAudio
Cooler Master Elite 120 Advanced Logitech K400 Trackpad Custom 2.1 Home Theater Setup 
  hide details  
Reply
post #4698 of 7150
Hi everyone,

I've had a RAID-5 array consisting of 3 2TB Samsung HD203WI drives for a few months running just fine. I took the plunge and got 5 more drives as I know that they will become a rarity soon (already are).

I didn't want to expand the array with more than two drives as I wouldn't need 14TB of storage straight away. So I plugged the two new drives in (surface-tested already) and started expanding the array from 3 to 5 disks.

During the expansion process the PC would bluescreen and reboot on several occations. This was a bit scary, and I'm not too sure why it happened. I thought it might have to do with having an underpowered PSU, so once it finished (3 days later) I connected the 5 drives to a separate PSU.

It turned out that the separate OS drive had gotten a bunch of bad sectors during the expansion process (probably caused by the bluescreens/reboots). To make a long story short, I managed to get everything up and runing again and once in Windows 7 it the progress indicator in MegaRAID storage manager indicated that the arraw was initializing in the background.

I was happy and started watching a movie on the PC. This is when I first dropped one disk and then about 30 mins later the second drive dropped out leaving me with an offline VD! The two dropped drives were 2 of the "old" drives that had been working perfectly up until then if that makes a difference.

This is the point where I'm really freaking out and I'm asking for advice.

I have read in the past that you should be able to re-create the array with the same settings without initializing the array and that there would be a good chance it would be OK. Now I would like to know the specific steps of this so that I don't break something in the process.

Should I first delete the virtual disk (now containing the remaining 3 drives) in MegaRAID storage manager and then create a new virtal disk with the same settings as the old one? If yes, and it works, should I then try to initialize the drive (as this is what it was doing when it dropped the drives) or should I just leave it and it will initialize in the background?

Before trying this, is it worth trying to power down the system, unplugging everything for 30 min or so and then reconnect it all and see if it magically sees all the drives and the array with it?

Sorry for the long post.

Any tips and advice is truly appreciated.

EDIT: This has now been resolved. See this post
Edited by sireddie - 11/11/10 at 6:01pm
post #4699 of 7150
A couple of hours reading up on the issue and I've gotten wiser and have come up with the following plan of action:
  1. Power up and go into the PERC BIOS and import a foreign configuration. This should in theory put all drives back on and start recreating the array from 4 to 5 disks (as it was initially degraded before dropping the second drive). Before doing this I will make sure that the foreign config contains all 5 disks and that these are online.
  2. If the above fails, then the plan is to create a new array (in the PERC BIOS) with the exact same settings as the old array had and without initializing the array afterwards.
  3. If the above fails, I will then try using third party software to recover the data: http://www.data-recovery-software.net/
  4. Once the array has been rebuild by any of the means above, I will run a consistency check (again in the PERC BIOS) which basically is the same as a background initialization only that it is manually started.

Lesson learned so far: Don't expand your array in Windows (or any OS for that matter). Stick to the PERC BIOS for these things.

I've also managed to find the Dell Perc 5/i manual which is very handy in situations like these.

A few success stories from opeople in similar situations: 1 2 3

Sugestions/comments are welcome as I won't be trying anything for another 14 hours or so.
post #4700 of 7150
Quote:
Originally Posted by sireddie View Post
Hi everyone,

I've had a RAID-5 array consisting of 3 2TB Samsung HD203WI drives for a few months running just fine. I took the plunge and got 5 more drives as I know that they will become a rarity soon (already are).

I didn't want to expand the array with more than two drives as I wouldn't need 14TB of storage straight away. So I plugged the two new drives in (surface-tested already) and started expanding the array from 3 to 5 disks.

During the expansion process the PC would bluescreen and reboot on several occations. This was a bit scary, and I'm not too sure why it happened. I thought it might have to do with having an underpowered PSU, so once it finished (3 days later) I connected the 5 drives to a separate PSU.

It turned out that the separate OS drive had gotten a bunch of bad sectors during the expansion process (probably caused by the bluescreens/reboots). To make a long story short, I managed to get everything up and runing again and once in Windows 7 it the progress indicator in MegaRAID storage manager indicated that the arraw was initializing in the background.

I was happy and started watching a movie on the PC. This is when I first dropped one disk and then about 30 mins later the second drive dropped out leaving me with an offline VD! The two dropped drives were 2 of the "old" drives that had been working perfectly up until then if that makes a difference.

This is the point where I'm really freaking out and I'm asking for advice.

I have read in the past that you should be able to re-create the array with the same settings without initializing the array and that there would be a good chance it would be OK. Now I would like to know the specific steps of this so that I don't break something in the process.

Should I first delete the virtual disk (now containing the remaining 3 drives) in MegaRAID storage manager and then create a new virtal disk with the same settings as the old one? If yes, and it works, should I then try to initialize the drive (as this is what it was doing when it dropped the drives) or should I just leave it and it will initialize in the background?

Before trying this, is it worth trying to power down the system, unplugging everything for 30 min or so and then reconnect it all and see if it magically sees all the drives and the array with it?

Sorry for the long post.

Any tips and advice is truly appreciated.
ok, i've done this like a million times now (mostly because of UREs.. and when you have such large drives (>1TB), you really need to start thinking about how to deal with UREs....

here's the deal, the background initialization, just like a consistency check, re-creates all the parities. in order to do this, it needs to read N-1 parts of each stripe (N being the number of drives in the VD). If any of the N-1 parts isn't readable (like with a URE), it can't create the parity bits. I suspect something like this happened and it just so happened that some of those unreadable parts might have been on those older drives. The only way to know for sure is to dump the controller's internal log and read through that.

but, that said, here is how to recover.. and you better know which drive failed first, and which second. this is also assuming the drives marked as "failed" will spin up and are accessible. when the 1st drive failed, the VD is in degraded, but still accessible and *writable* ... meaning that data can still change on it... this means the 1st failed drive's data is no longer consistent with the rest of the array... so, put that drive away for now.... disconnect it to be safe. when the 2nd drive fails, the VD goes into "offline"... which means that at the moment drive #2 fails, the VD is no longer accessible. This is sort of good, because that means the rest of the VD is no longer changing at this point. So, what you can do, and still have a hope that the drives are consistent with each other, is to "force online" the #2 failed drive.... forcing online means that you're telling the controller that you know better than it does that failed drive #2 and the remaining RAID-5 disks (minus the failed drive #1) are consistent with each other, so just assume so and move on with life.

once you "force online" #2, the VD will be in degraded state again. but, at this point, some of the data is readable. i recommend that you consider backing up your data at this point to an external drive or over the network to another system. I say *some* because you might have a punctured stripe.. meaning that more than 1 component of some stripes might not be usable, so that means some of the VD blocks cannot be reconstructed. if you can, avoid reading those blocks... too many I/O errors trying to read bad blocks will take the VD offline again. Of course, if that happens, you can just force the failed drive online again.

there is no guarantees, but i've recovered a lot of RAID-5 arrays with 8 to 15 drives this way when i've gotten more than 1 failed drive.

once you've gotten your data safely off, then i would look into dumping the controller debug log and see exactly what caused the issue. then see if you can address the issue.

if you end up re-creating the array, run a "slow initialization" and let that finish before you start using the re-created array. this process writes 0 bits to the entire drive and associated parity bits. during the "write" process, it can sometimes detect bad blocks on drives and hopefully get them reallocated so the drives are more reliable. you can see if that happens by looking at the SMART data before and after.
TAIPEI
(10 items)
 
AURORA
(13 items)
 
 
MotherboardGraphicsRAMHard Drive
ASRock X99 Extreme11 EVGA GTX 980 Superclocked 32GB 8x4GB Corsair LPX Samsung XP941  
Hard DriveCoolingOSMonitor
Western Digital 3TB RE Noctua NH-D15 Fedora 21 Linux Samsung S27D590C 
PowerCase
Seasonic SS-1200XP Cooler Master Cosmos II 
CPUMotherboardGraphicsRAM
Dual Quad-core L5430 2.66Ghz 12mb cache Intel 5000 chipset ATI ES1000 64GB FBDIMM DDR2 PC2-5300 667Mhz 
Hard DriveOSPower
WD3000FYYZ PERC H700 w/ 512MB cache CentOS 7.2.1511 950W x2 
  hide details  
Reply
TAIPEI
(10 items)
 
AURORA
(13 items)
 
 
MotherboardGraphicsRAMHard Drive
ASRock X99 Extreme11 EVGA GTX 980 Superclocked 32GB 8x4GB Corsair LPX Samsung XP941  
Hard DriveCoolingOSMonitor
Western Digital 3TB RE Noctua NH-D15 Fedora 21 Linux Samsung S27D590C 
PowerCase
Seasonic SS-1200XP Cooler Master Cosmos II 
CPUMotherboardGraphicsRAM
Dual Quad-core L5430 2.66Ghz 12mb cache Intel 5000 chipset ATI ES1000 64GB FBDIMM DDR2 PC2-5300 667Mhz 
Hard DriveOSPower
WD3000FYYZ PERC H700 w/ 512MB cache CentOS 7.2.1511 950W x2 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: RAID Controllers and Software
Overclock.net › Forums › Components › Hard Drives & Storage › RAID Controllers and Software › PERC 5/i RAID Card: Tips and Benchmarks