RAID Clarifications

Hi,
I read in the Knowledge Base Answer ID 19641 steps to installing a used HDD into NMH-300 to rebuild the RAID setup.
Can anyone confirm if the steps apply to new HDD as well?
For instance, if RAID1 is already configured and the HDD is Bay 1 crashes, do I:
1. Insert a new HDD directly back to Bay 1 and begin the build, or
2. Switch the working HDD in Bay 2 to Bay 1 then insert the new HDD into Bay 2 and begin the build.
Any advice is appreciated, I spent countless hours trying different rebuild strategies but the NMH-300 RAID setup is not hot swappable.
Thanks in advance.
Regards

Hi,
If we base it on the Linksys article below regarding RAID 1, bay 1 of the mediahub should always contain the harddisk thats going to be backed up onto the harddrive on bay 2. If the harddisk on bay 1 gets defective then you may need to put the harddisk on bay 2 onto bay 1 then add the (new) replacement harddisk on bay 2. This sounds confusing for me as well ...
Creating RAID 1 on the Network Media Hub
I hope this helps.

Similar Messages

RAID - Clarification on Concept & Components

This forum is a great resource on RAID ... yet I still am foggy and need to better understand before making my controller choice.
Can someone shed more light on the various components of a controller to consider?
How do these variables affect video editing performance and future upgrades?
Does controller documentation give you all the info needed to install?
What are the main things to think about as one compares these the main elements on a RAID controller?
Type
External Connectors
Internal Connectors
Interface
Transfer Rate
Cache Memory
RAID (this has been explained well already)
OS Support
I edit mostly doc style with xdcam and dslr footage. My build will be the typical assortment posted many times - i7-950, X58 mobo, 12 GB RAM or more, GTX 470 or 580 and a non RAID system disc,and a 850+ PSU
My current plan is to RAID 5 - 4 1TB 7200 Samsung f3's. Since I do not see much info out there on RAID 3 ... I want to stick with RAID 5.
A fellow forum member has had great luck with 3ware 9750-4i. It's a great price and should meet my needs ... unless I should be planning further ahead.
http://www.newegg.com/Product/ImageGallery.aspx?CurImage=16-116-109-TS&SpinSet=16-116-109- RS&ISList=16-116-109-Z01%2c16-116-109-Z02%2c16-116-109-Z03%2c16-116-109-Z04%2c16-116-109-Z 05%2c16-116-109-Z06&S7ImageFlag=1&Item=N82E16816116109&Depa=0&WaterMark=1&Description=3war e%20Internal%209750-4i%20SATA%2fSAS%206Gb%2fs%20PCI-Express%202.0%20w%2f%20512MB%20onboard %20memory%20Controller%20Card%2c%20Single
3ware Internal 9750-4i SATA/SAS 6Gb/s PCI-Express 2.0 w/ 512MB onboard memory Controller Card, Single
3ware > Newegg Item#: N82E16816116109
THANK YOU.
--NAN

Actually the last benchmarks i saw for the 3Ware card with raid 5 or 10 were very good. I was actually surprised by them though their management utility is really lacking.
Eric
ADK

RAID / Storage Rebuild... requesting clarifications for all ;)

Hello All,
I'm replacing the 2 Seagate HDDs out of my home video editing system (CS5) that was previously configured as below, and think its a good time to get clarification on some items I never really resolved after many hours of research, to include where the different items should be installed/stored on a six-HDD system, and/or whether someone in my situation should completely re-think the system setup.
It was in the following configuration not out of original intention, but due to adding HDDs over time...
Intel RST/Matrix RAID via ASUS P6T Deluxe V2....
2 x 1TB Seagate HDDs w/ 3 Partitions
C: OS & Critical Programs in 100GB Matrix Raid1 Partition,
D: Saved Adobe Project Files, etc. in 250GB RAID0 Parition
E: Rendered Outputs, etc. in remaining GBs RAID0 Partition
2 x 1TB Samsung Spinpoint F3
F: Scratch files on 2GB RAID0
2 x 1TB Samsung Spinpoint F3
G: Original Video / Media Storage on 2GB RAID0
System Hardware:
Asus P6T Deluxe V2
i7 980
2 x 1TB Seagate 7200 HDDs (changing to something else)
4 x 1TB Samsung Spinpoint F3 7200 HDDs
1 x Asus GTX570 Direct CU II (exchanging for a 2-slot card, as this 3-slot won't snap secure due to sata port locations, etc)
1 x GTX275 for additional video out (haven't actually tried 3 monitors yet)
24GB Corsair Vengeance RAM
900w power supply
Cooler Master HAF Case
Various external HDDs for periodic backup via eSATA
Was Vista, ordering Windows 7 Pro
1) Assuming the hardware remains 6 x 7200RPM HDDs on a P6T Deluxe V2 (sort of want to stay on it, since just got an expensive lga1366 cpu), which only has 6 SATA ports available...
a) Am I even correct that the best editing configuration for six HDDs is 3 x RAID0 arrays?
b) If so, how should I distribute the following items across the 3 arrays - OS, Programs, Adobe Projects, Scratch, Outputs, Original Media? I could never figure out from forums which items are the most important to separate from each other on different discs.
2) Or should I really be modifying the system to allow more than the 6 HDDs at this point?
a) Now for the really nebulous question... if I need to change, should this be an internal RAID controller card with more ports, or does it need to be an external setup?
If I really need expanded equipment, I'm ok with things like buying "faster" HDDs (10k or whatever) to replace the 2 Seagates if that's a strong recommendation, spending a bit on better RAID equipment if it is actually worthwhile, but...
1. Can't spend a thousand to only slightly improve it
2. Need a setup that someone of moderate computer skill and experience can setup and maintain.
In particular, the question of how to distribute the various items/ tasks across existing discs is likely of interest to many readers. This is talked about a lot, but I'm really unclear on how best to distribute them and in what priority if you don't have separate arrays for every item/task.
Thanks to anyone interested!
P.S. NOTE to other amateurs w/ my RAID setup... trying to install a large graphics card and somehow resetting the BIOS in the process is what spurred my current possible re-build. Didn't realize it had reset, continued startup to find Intel RST not functioning, drives renamed and of course RAID0s inaccessible. Then changing back to RAID setup in BIOS not effective at that point. Since I have important stuff backed up and wanted to change out some drives anyway, not worth trying to restore the RAID and recover, but you might end up in worse shape. SUPPOSEDLY if you change the setting back to RAID before OS startup, your arrays and Intel RST will function normally even after the bios reset.

I swear you are a benevolent jaguar, always lurking and ready to pounce at a moment's notice!
Information received and agreed. That's exactly the simple breakdown I've needed to know for years. I assume the 1TB on C: & F: is not a mandate, but based on already having 1TB HDDs on-hand (and also assuming it should be large enough to keep the disc usage % low)?
Unless I'm advised otherwise, I suppose I'll just get the "fastest" large capacity non-ssd HDD's I can (maybe velociraptor 600gb?) for the non-RAID single discs then (C: & F: drives), since I want to replace the two seagate HDDs anyway. I guess both seek/read time and not just throughput/write stats are relevant for the C: OS drive, if not so much for the export F: drive. For that matter, guess I could just SSD the C: drive, and Velociraptor (or something with more capacity) the F: drive.
Many Thanks!
P.S.: Assumed the matrix partition wasn't a good idea... a hold-over from when I only had 2HDDs and was guessing my way along.

Clarification on how to use Xserve Raid and Fibre Channel without xsan.

First let me apologize for not responding earlier to your response, I tend to get busy and then forget to check back here.
Tod, the answer to your question is No, only one computer is accessing the xserve raid files at any one time and that is via Fibre Channel. However I do have the xserve raids set up as share points via ethernet.
Maybe I should turn that off and only access the files with the one computer that can connect via fibre channel.
I never thought of that. I will try that while I await for your answer, thanks again.
Todd Buhmiller
I have the following setup:
Xserve: 2x2Ghz Dual Core Intel Xeon, 5Gb of Ram, Running 10.5.8 Leopard Server
Xserve Raid with firmware version 1.5.1/1.51c on both controllers, and
Qlogic Sanbox 5600
Apple Fibre Channel Cards in Xserve, and Mac Pro Tower; Apple 2 Port 4Gbs Fibre Channel Card
Mac Pro Tower-Quad Core Intel Xeon 2.8Ghz, 16Gb of Ram, Running Snow Leopard 10.6.4
Here is the problem.
The directory for the xserve raids keep getting corrup, and I use disc warrior to rebuild them. Is there a way to keep the directories from getting corrupt? I am a few pieces of equipment before I can build an Xsan as that is the ultimate goal, but until then, I just need to be able to have the raids funciton as storage without having to rebuild the directories all of the time.
Anybody have any suggestions?
Thanks
Todd Buhmiller
Widescreen Media
Calgary, Alberta Canada
Tod Kuykendall
Posts: 1,237
From: San Diego
Registered: Oct 11, 2000
Re: Xserve Raid Mounts, Corrupt Directory tired of rebuilding directory
Posted: Jun 27, 2010 1:25 PM in response to: Todd Buhmiller
Are multiple computers accessing the same data on the RAID at the same time?
If so then NO. This is the source of your data corruption and I'm surprised if you were able to get all your data back every time if this is how you've been running your system. Each fibre channel assumes it has full and sole control of every volume it has mounted, no data arbitration is practiced and data corruption will occur if this assumption is wrong.
The only way this set-up will work is to use partitions or LUN masks so the volumes are accessed by one computer at any time. As long as one computer relinquishes control before another mounts it you will dodge arbitration issues but this is a dangerous game. If you screw up and mount an already mounted volume - and there is no easy way to tell if a volume is mounted - corruption will then occur. Sharing data simultaneous at fibre speeds is what XSAN does and to do this you need it.
HTH,
=Tod
Intel Xserve, G5 XServes, XRAID, Promise

+The xserve raids will mount automatically to any computer that I connect the qlogic fc switch to+
This is source of the corruption to your data. Any computer that attaches to a drive/partition via fibre channel assumes that it alone is in control of the drive and data corruption is inevitable.
+Is that the issue, should I disconnect the xserve from the fc switch and leave it connected via ethernet?+
Short answer: YES. The ethernet connections are fine because the server is controlling the file arbitration through the sharing protocol. Fibre channel connections assumes complete control over the partition and no arbitration of the file access is performed. It's like two people independently driving trying to drive the same car to different locations.
Depending on your set-up it is possible for the two machines to see and use different parts of the Xserve RAID storage but they cannot access the same areas without SAN doing the arbitration.
Hope that's clear,
=Tod

Clarification: Installing ARCH to RAID setup in BIOS

So I've been looking in to setting up ARCH on my laptop that has a RAID that's set up in the BIOS. I've been researching on the wiki as well as on the internet but it's only partially answered my question:
With a BIOS enabled RAID do I still need to use tools like MDADM to set up the raid inside of arch, or should arch see its own drive like windows does?
I'm sure I will have other questions as I delve in to installing on a RAID so I will ask what I have now as well:
Reading threads about doing this it seems like I will need a third drive something outside of the RAID to store the RAID configuration if I opt to use GRUB, however syslinux supports RAID during boot, does this still hold true and if so what do I need to do while installing syslinux to have it know that there's a RAID there and configure correctly?
Any other tips on this topic are very welcome.

So, is rEFIT able to boot GRUB but GRUB is not able to find Arch? Or is rEFIT not able to find GRUB?
For what it's worth, I recently installed Arch on my Macbook 5,1. I also used rEFIT (installed using OS X). My drive partitioning was
/dev/sda1: EFI/OS X boot partition
/dev/sda2: OS X
/dev/sda3: Linux Swap
/dev/sda4: Arch
As you mentioned, the new Arch installer doesn't give you the option of selecting primary partitions to install GRUB (the old installer did). So, when I reached the last stage of the installer (installation of the bootloader), I exited the installer, launched GRUB in interactive mode, and installed GRUB to my 4th partition (where /boot is). (It sounds like you did something similar.) When I rebooted rEFIT discovered that my 4th partition was bootable and displayed a Tux logo. Clicking on the Tux logo launched grub. From there I am able to launch Arch.
Some thoughts:
1. Did you make sure your partition tables are synced? I didn't need to do this, but I've heard of other needing to do this.
2. Did you make sure that the grub was configured properly; e.g. you've specified the correct path to the kernel, etc.
3. Did you try prayer and/or other forms of divinity?

Need urgent help with RAID failure on Graphics server (running OSX10)

Hey.
Here's the deal... Our graphics department has a MAC G4 (fully upgraded. newest patches, OS, etc.) that's got a 4 port SATA RAID controller. They use this with four 750GB hard drives as an external RAID (the OS is on a seperate, 65GB hard drive). The way it's set up, through Apple's OS, is we have the four 750GBs in pairs... those pairs are striped, and then the two sets of striped 750s are mirrored. This has worked just fine for a number of months, but we recently had a power surge, and now we're unable to mount "Monkey" (the name of the 1.4TB array).
See the below screenshots for more clarification.
http://www.amcdoors.com/lorenzo/raidsetup.JPG
This shows the RAID card, and the four SATA cables coming out of it.
http://www.amcdoors.com/lorenzo/leftside1.JPG
As you can see in this and the below screenshots, this shows the 1.4TB Striped set as 'online', but the two below (the two 750GB hard drives) show up as "offline".
http://www.amcdoors.com/lorenzo/leftside2.JPG
this shows the actual drive i click on. same thing
http://www.amcdoors.com/lorenzo/leftside3a.JPG
If i click the entire array, you can see how it's set up (the Mirrored RAID set consisting of the two Striped RAID sets). The RAID sets (all 3 of them) are showing up as online, yet the drives are offline. No idea why. The drives are all powered, and are spinning, and plugged in.
http://www.amcdoors.com/lorenzo/leftside3b.JPG
when you click "Monkey Drive", nothing shows up in it. i don't remember if anything ever used to. if you click 'mount', nothing happens.
http://www.amcdoors.com/lorenzo/leftside4.JPG
http://www.amcdoors.com/lorenzo/leftside5.JPG
this shows the exact same problem as the first RAID set.
http://www.amcdoors.com/lorenzo/verifyRAIDset1.JPG
If you try to Verify the RAID set, it shows that no repairs are necessary
http://www.amcdoors.com/lorenzo/verifymonkey.JPG
If you try to verify Monkey, you get a "Volume Needs Repair" error... but when you try to repair it, you get this:
http://www.amcdoors.com/lorenzo/repairmonkey.JPG
Really need help; this is a pretty critical problem. Someone that's familar with RAID solutions on MACs please help me out. We can't risk losing any data if at all possible.
Thanks
~Lorenzo

You might want to post this to the Server Products forums.

RAID controller on G4; problems... is RAID dead? Help please!

Hey.
Here's the deal... Our graphics department has a MAC G4 (fully upgraded. newest patches, OS, etc.) that's got a 4 port SATA RAID controller. They use this with four 750GB hard drives as an external RAID (the OS is on a seperate, 65GB hard drive). The way it's set up, through Apple's OS, is we have the four 750GBs in pairs... those pairs are striped, and then the two sets of striped 750s are mirrored. This has worked just fine for a number of months, but we recently had a power surge, and now we're unable to mount "Monkey" (the name of the 1.4TB array).
See the below screenshots for more clarification.
http://www.amcdoors.com/lorenzo/raidsetup.JPG
This shows the RAID card, and the four SATA cables coming out of it.
http://www.amcdoors.com/lorenzo/leftside1.JPG
As you can see in this and the below screenshots, this shows the 1.4TB Striped set as 'online', but the two below (the two 750GB hard drives) show up as "offline".
http://www.amcdoors.com/lorenzo/leftside2.JPG
this shows the actual drive i click on. same thing
http://www.amcdoors.com/lorenzo/leftside3a.JPG
If i click the entire array, you can see how it's set up (the Mirrored RAID set consisting of the two Striped RAID sets). The RAID sets (all 3 of them) are showing up as online, yet the drives are offline. No idea why. The drives are all powered, and are spinning, and plugged in.
http://www.amcdoors.com/lorenzo/leftside3b.JPG
when you click "Monkey Drive", nothing shows up in it. i don't remember if anything ever used to. if you click 'mount', nothing happens.
http://www.amcdoors.com/lorenzo/leftside4.JPG
http://www.amcdoors.com/lorenzo/leftside5.JPG
this shows the exact same problem as the first RAID set.
http://www.amcdoors.com/lorenzo/verifyRAIDset1.JPG
If you try to Verify the RAID set, it shows that no repairs are necessary
http://www.amcdoors.com/lorenzo/verifymonkey.JPG
If you try to verify Monkey, you get a "Volume Needs Repair" error... but when you try to repair it, you get this:
http://www.amcdoors.com/lorenzo/repairmonkey.JPG
Really need help; this is a pretty critical problem. Someone that's familar with RAID solutions on MACs please help me out. We can't risk losing any data if at all possible.
Thanks
~Lorenzo

have you gotten this to work yet ?? if not let me know
-d

Software RAID failing constantly

Dear all,
I need help as I cannot work out what is going wrong the the Mirrored RAID i set up.
I bought 2 x 1TB drives and setup Mirrored RAID (software) through the GUI Disk Utility Application. I assumed it would all "just work".
After about a month the RAID failied. It happened after I ran TechTool Pro 5. Could this have caused permanent damage which leads me to where I am today (see below)? I unplugged the drives and plugged into my laptop (I don't know why I didn't just unplug and plug them into the Mac Mini they were in). Anyway it came up saying "Rebuilding" and it took about 9 hours. All was good again.
Until early October 2009 when Disk Utility reported that the RAID was "degraded" and one of the drives had "failed".
I was pretty shocked as it was a new drive. I thought the best thing to do would be to get a new 1TB drive to rebuild the RAID and then deal with the failed drive.
It took over 2 weeks to get the new drive and enclosure (long story). I setup the new drive last night, plugged it in and the "Rebuilding" message came up with the same "it will take 9 hours message".
I left it over night and when I got up Disk Utility told me that the older drive of the two has "failed". There has to be something weird going on here and this is where I need your help.
Configuration
1. The drives are 1TB Hitachi 3200rpm SATA I/II, 32 MB Cache drives
2. The 'old' enclosure is an Icecube G2 SuperS 800/400/USB2/eSata 3.5" Enclosure
3. The 'new' enclosure is an OWC Mercury Elite-AL, Quad Interface eSata/FW 400/FW 800/USB 2/USB 1
4. The Mac Mini only has one FW800 port and I connect the 2 drives up in serial.
<hr />
Here is the behaviour I'm seeing now.
I'll start with both drives off and turn both on.
1. Both drives mount in Disk Util
2. 20 seconds later a message comes up about rebuilding slices for "disk1 ... ?" and "disk (null)" and disk util crashed 2 seconds later (thus why didn't catch all of the message)
3. I turn to the terminal:
bsmith@kanga ~ $ diskutil list
/dev/disk0
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *298.1 Gi disk0
1: EFI 200.0 Mi disk0s1
2: Apple_HFS Mini 297.8 Gi disk0s2
/dev/disk1
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *931.5 Gi disk1
1: EFI 200.0 Mi disk1s1
2: Apple_RAID 931.2 Gi disk1s2
3: Apple_Boot Boot OSX 128.0 Mi disk1s3
/dev/disk2
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *931.5 Gi disk2
1: EFI 200.0 Mi disk2s1
2: Apple_RAID 931.2 Gi disk2s2
3: Apple_Boot Boot OSX 128.0 Mi disk2s3
/dev/disk3
#: TYPE NAME SIZE IDENTIFIER
0: Apple_HFS terra *931.2 Gi disk3
bsmith@kanga ~ $
bsmith@kanga ~ $ sudo diskutil listRAID
Password:
RAID SETS
===============================================================================
Name: terra
Unique ID: 2AC46C49-5F0B-4BD5-B1FE-C322A6F51DD9
Type: Mirror
Status: Degraded
Size: 999860895744 B
Rebuild: automatic
Device Node: disk3
Apple RAID Version: 2
# Device Node UUID Status
0 disk1s2 A12D748F-0779-4217-8C75-B31D9AE45B4E 0% (Rebuilding)
1 disk2s2 CF03D24E-AB5D-4F86-A7B1-26A1F4589CB7 Online
===============================================================================
bsmith@kanga ~ $
Run it again 10 seconds later:
bsmith@kanga ~ $ diskutil list
It just hangs.
bsmith@kanga ~ $ sudo diskutil listRAID
It just hangs.
4. Now I turn off the 'old' drive and look at the terminal (that was hanging). When I turn it off I'm told (through the Hardware Growler Growl extension to show the status of hardware) that the old drive ("Macpower"), the new drive ("OWC") and the RAID disk ("terra") have all unmounted. The diskutil commands follow.
bsmith@kanga ~ $ diskutil list
/dev/disk0
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *298.1 Gi disk0
1: EFI 200.0 Mi disk0s1
2: Apple_HFS Mini 297.8 Gi disk0s2
/dev/disk1
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *0.0 B disk1
1: EFI 200.0 Mi disk1s1
2: Apple_RAID 931.2 Gi disk1s2
3: Apple_Boot Boot OSX 128.0 Mi disk1s3
/dev/disk2
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *931.5 Gi disk2
/dev/disk3
#: TYPE NAME SIZE IDENTIFIER
0: Apple_HFS terra *931.2 Gi disk3
bsmith@kanga ~ $
bsmith@kanga ~ $ sudo diskutil listRAID
RAID SETS
===============================================================================
Name: terra
Unique ID: 2AC46C49-5F0B-4BD5-B1FE-C322A6F51DD9
Type: Mirror
Status: Degraded
Size: 319728959488 B
Rebuild: automatic
Device Node: disk3
Apple RAID Version: 2
# Device Node UUID Status
0 disk1s2 A12D748F-0779-4217-8C75-B31D9AE45B4E 0% (Rebuilding)
1 -none- CF03D24E-AB5D-4F86-A7B1-26A1F4589CB7 Online
===============================================================================
bsmith@kanga ~ $
I don't know alot about this output, though its weird that disk2 is still in the list but only shows the GUIDpartitionscheme. And its interesting that disk1 is also in the list but shows 0.0B against the GUIDpartitionscheme for it.
5. Just for kicks I turn off the 'new' drive aswell and run the commands again. When i do it Hardware Growler doesn't say anything has unmounted (I guess since they all did so before).
bsmith@kanga ~ $ diskutil list
/dev/disk0
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *298.1 Gi disk0
1: EFI 200.0 Mi disk0s1
2: Apple_HFS Mini 297.8 Gi disk0s2
bsmith@kanga ~ $ sudo diskutil listRAID
Password:
No RAID sets found
As expected I guess.
6. Now I'll turn both drives back on and confirm I get back to the same state I was in before after both drives were turned on.
This time I get the Hardware Growler message that both drives have mounted followed by the RAID drive about 20 seconds later.
bsmith@kanga ~ $ diskutil list
/dev/disk0
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *298.1 Gi disk0
1: EFI 200.0 Mi disk0s1
2: Apple_HFS Mini 297.8 Gi disk0s2
/dev/disk1
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *931.5 Gi disk1
1: EFI 200.0 Mi disk1s1
2: Apple_RAID 931.2 Gi disk1s2
3: Apple_Boot Boot OSX 128.0 Mi disk1s3
/dev/disk2
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *931.5 Gi disk2
1: EFI 200.0 Mi disk2s1
2: Apple_RAID 931.2 Gi disk2s2
3: Apple_Boot Boot OSX 128.0 Mi disk2s3
/dev/disk3
#: TYPE NAME SIZE IDENTIFIER
0: Apple_HFS terra *931.2 Gi disk3
bsmith@kanga ~ $
bsmith@kanga ~ $ sudo diskutil listRAID
RAID SETS
===============================================================================
Name: terra
Unique ID: 2AC46C49-5F0B-4BD5-B1FE-C322A6F51DD9
Type: Mirror
Status: Degraded
Size: 999860895744 B
Rebuild: automatic
Device Node: disk3
Apple RAID Version: 2
# Device Node UUID Status
0 disk1s2 A12D748F-0779-4217-8C75-B31D9AE45B4E 0% (Rebuilding)
1 disk2s2 CF03D24E-AB5D-4F86-A7B1-26A1F4589CB7 Online
===============================================================================
bsmith@kanga ~ $
Now let me run the commands again:
MMM, the exact same result is returned - it didn't hang this time.
It's 21:44, I'll leave this a while and see if the rebuilding continues.
Its 13 mins later and this time the commands are just hanging. Before I went away at 21:44 the lights on the front of the drives where flashing showing something was happening. They had stopped by the time I returned.
7. Now I turn off the 'new' drive and run the terminal commands again. When I turn it off I'm told by Hardware Growler that the new drive ("OWC") and the RAID disk ("terra") have both unmounted. Last time (when the 'old' drive was turned off) Growler told me all 3 drives unmounted. The diskutil commands now show.
bsmith@kanga ~ $ diskutil list
/dev/disk0
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *298.1 Gi disk0
1: EFI 200.0 Mi disk0s1
2: Apple_HFS Mini 297.8 Gi disk0s2
/dev/disk1
#: TYPE NAME SIZE IDENTIFIER
0: GUIDpartitionscheme *931.5 Gi disk1
It hangs at this point and hasn't completed yet.
bsmith@kanga ~ $ sudo diskutil listRAID
Password:
RAID SETS
===============================================================================
Name: terra
Unique ID: 2AC46C49-5F0B-4BD5-B1FE-C322A6F51DD9
Type: Mirror
Status: Offline
Size: 999860895744 B
Rebuild: automatic
Apple RAID Version: 2
# Device Node UUID Status
0 disk1s2 A12D748F-0779-4217-8C75-B31D9AE45B4E Failed
0 -none- CF03D24E-AB5D-4F86-A7B1-26A1F4589CB7 Missing/Damaged
===============================================================================
bsmith@kanga ~ $
So HELP PLEASE!. Does the above provide enough information to suggest to someone with more knowledge than me about this, what is going on and what can I do to fix this? Let me know if there are any more tests I can run.
I have a backup of the data so can erase and recreate the RAID drive. But what would stop this happening again? Could it have been TechTools Pro 5 which was running when the problem first happened that caused some permanent damage?
Any and all help appreciated.
Thank you,
Brooke

Thanks ridogi and Antonio for your replies,
For clarification, both drives have dual FW 800 to allow them to be connected. No FW 400 involved (I'm not that silly .
I can't say i've done extensive research into setting up RAID drives but I've never come across advice to not use multiple external drives. Here is an extract from a Nov 2007 MacTech article on using RAID for backup:
{quote}
The Desktop Mac User:
The second user I want to look at is the professional who relies on his desktop Mac for his business. He is a lawyer, developer, architect or other professional whose business relies totally on the contents of his computer. If he loses the files on his computer, his business will really suffer. In addition, the time lost when dealing with a failed hard disk is money down the drain. This user relies on a three disk mirror volume. His Mac has two internal SATA disks which are used as the primary and secondary disks of a mirror volume. If one of the disks fails at any given point, the other one will take over and become the new primary disk, so his files are always protected from a hard disk failure. The third disk is an external FireWire disk which is also a secondary disk. It is stored offsite (either at home in a fireproof safe or at another safe location). Every Friday, he brings this third disk into his office and connects it to his desktop Mac. Once this disk is connected, a mirror rebuild starts automatically. He can perform this rebuild as he continues to use his Mac during the day. At the end of the day, when he shuts his Mac down, he disconnects his external secondary disk and returns it to its offsite location. This offsite secondary disk becomes his insurance against his building getting broken into or his building burning down. Like the laptop user, if his Mac gets destroyed, he can just purchase a replacement Mac and then boot up from his external secondary disk. At most, he will use one week's worth of work.
{quote}
So I just assumed it is ok to use external drives. This is discussing a software product called SoftRAID and perhaps that is able to handle the situation. Or perhaps one external drive is ok???
My plan, developed when deciding to buy the Mac Mini April 2009 was to amortize costs by buying external drives first for doing backups and then buying a NAS within 12 months. I guess I should have just gone the NAS.
However ridogi I see you've recommended WiebeTech which don't seem to have NAS devices but DAS (direct attached storage) as the devices are connected directly to a network machine via Firewire etc. (Whereas NAS connects via the Network - I'd connect such a device to the Airport wireless router). Do you prefer DAS over NAS? I guess the downside of NAS is that the complexity is somewhat greater.
And reading http://www.wiebetech.com/just-say-no-to-nas/ they say:
{quote}
Let's face it - Networked or "shared" storage is a great feature but most NAS devices are just too slow for many applications. 3MB/s for a NAS device is common and 10MB/s is considered fast. Direct Attached Storage (DAS) performs much faster while still having the ability to be "shared". The RTX400-QR can transfer data to your computer at speeds exceeding 100MB/s.
{quote}
How can this be? If connected via Gigabit ethernet (128MB/s) then isn't that faster?
Though it is true that if you connect via Wifi then the best speeds you are going to achieve is 6.75 MB/s (802.11g) or 17.5 MB/s (802.11n).
Of course the speed advantages for DAS are only going to be on the machine its connected to as the other machines will connect to it (via the machine its on) over the network anyway!
Why I like the idea of NAS is to move the drives away from the Mini (noise and powerpoint reasons), make it easier to share with other Macs/PCs in the house (including via SMB) and I have a vague hope that this might stop the delay caused by the connected external drives spinning up when an open/save dialog box is activated (or is this just me?!).
Your thoughts appreciated! And thanks for the help so far.
Cheers,
Brooke

Raid configuration -- the elusive best practice/best value framework

After much research on RAID, both on and off the Apple site, I am still looking for answers. I searched RAID threads for thoughts on some of the top users in the Forum so I apologize, I am sure some of this seems old hat to you old Pros.
I just bought a Mac Pro. Haven't even fired it up yet because I want to get the storage issue (RAID 0 or 1, or 0+1) settled before I transfer my jpg and video files. I have pretty much decided against the expense of the Hardware RAID card when its performance seems somewhat less than rock solid (perhaps a myth). Anyway, with a 4-core PRO I assume (correct if I am wrong) that the CPU hit to do software RAID is reasonable.
Other Questions:
1) The single 640 Gb hard drive as retailed. Can I separate "Users and their jpg/video files" to a drive separate from the boot, OSX, and Applications? Is it advisable?
2) If #1 is recommended, should I then mirror the boot drive or simply Time Machine the backup of the single drive?
3) Somewhere someone suggested better performance by putting together drive 1 and 3, and drive 2 and 4? I don't recall if that approach was RAID 1 for drive 1&3 (mirroring the boot), and striped (RAID 0) for drive 2 and 4. Your thoughts on this -- assuming its even relevant based on your recommendation for the boot drive on question #1 and #2?
4) So at this point we're down to how we should use either the remaining two or three drive bays depending the choice taken with the boot drive. What is your recommendation for optimal value -- e.g. maximizing storage, data protection, and costs. Note: I am willing to purchase external drive(s) for Time Machine backup.
5) Name your top two or three internal and external drive picks for this arrangement.
Thanks much for your help on this. Anything else you'd like to suggest or question for clarification.
cougar90

Well, if you're going to be using the system for dual purposes, right off the bat maybe software raid is not for you (especially a 0+1). That's a lot of overhead to be dealing with: video editing plus users accessing files on the same comp.
If you have users needing to access files, I would keep the "video editing" system and "server" system separate.
Hatter's idea of a PC/Mac compatible NAS sounds good; very easy and affordable to implement. I only wonder about the speed, if you will be transferring large video files, and have multiple users connected at once (although if hatter recommended it, I'm sure it's fine). If you do go the NAS route though, make sure you have a gigabit network running. If you have any old computer system laying around (pc or mac), you can also configure that very easily as a server. Add hard drives or external enclosures for space. If it's a spare mac, and you have tiger 10.4, the app sharepoints works very well.
The PVR can be done on the Mac pro; keep it with the "video editing" system.
In regards to Raid card stability, I believe you were looking at the "Apple Raid Card" for the mac pro. Yes, there have been many problems with it regarding the battery. However this is pertaining to only the Apple raid card, NOT hardware raid in general.
There are other companies that put out very solid raid cards. Check the before link to www.amug.org, they are a great resource of raid info. On my setup, I use an ATTO Card connected to a D800RAID from Sonnet (mine is the previous model).
http://www.sonnettech.com/product/fusiondx800raid.html
http://eshop.macsales.com/item/ATTO/ESASR380000/
Also, before even attempting raid, get a good grasp on it.
http://www.acnc.com/040100.html
http://en.wikipedia.org/wiki/RAID
Note RAID 5 needs Hardware Raid. Also if hardware raid is too expensive, you can also go with esata enclosures. This sonnet enclosure with esata card for example:
http://www.sonnettech.com/product/fusiond500p.html
http://www.sonnettech.com/product/temposatae4p.html
More affordable, with great performance. Use for Raid 0 scratch, temp files.

Warning lights on XRAID but RAID Admin won't mount it.

I have 3 Final Cut edit suites sharing 2 x 7TB XserveRAIDs.
I had an audible alarm on one of my RAIDs an amber flashing light by the System Identifier Button.
Edit 2 started up fine and my editor was happily working away. I then went Edit 2 to mount the RAID storage - made up of 2 x 7TB XserveRAIDs each split into 2 partitions providing a total of 4 volumes creating my volume based SAN. As I tried to mount the RAIDs using SANmp I had real issues. Eventually got the volumes mounted but then had the greyed out screen of doom when I tried loading up a Final Cut Pro project and had to do a forced shut down.
Edit 2 was still working but I then had similar problems in Edit 1 - not able to mount partitions and unable to load projects.
I tried looking up the RAIDs on the RAID Admin Tool but I'm only seeing 1 RAID when I should be seeing 2. The AID that is mounted is not giving me any errors which makes me think I've lost communication with the RAID with the problem.
Should I just shut the whole system down and hope it starts talking to me again when I fire it up again?
I just want to get on an edit my show.
Thanks.

William: Thanks for your reply, but you seem to be wide of the target here.
RAID Admin can connect to the RAID,
but doesn't see drives?
No. I can run RAID Admin on the Xserve or on a separate laptop, and either way RAID Admin can't find any RAIDs to show in the list of available RAIDs. So it doesn't even get to the stage where it can see any individual drives, since it doesn't find the RAID unit in the first place.
The red light you're seeing, as well as the
fact that Disk Utility can see all the drives
as individuals,
Clarification: Disk Utility is showing symbols for drives, but of course these are really partitions of the RAID, not the individual physical hard drives in the RAID.
indicates that either stuff is set in JBOD mode
... well I'd love to report what mode it's in, but RAID Admin won't show me . However, judging by the fact that there's no part of the file system that's obviously missing, (and also by our original intent when this was set up) I'm 99.9% sure it's in some redundant mode, and 99.8% sure that is simply in default RAID-5 mode.
or something is seriously wrong...
... depending on the meaning of "seriously"!
I'd guess the RAID set (and any data that
was on it) is gone.
... though as I noted in original post, it's in everyday use, including by me, and I've not detected any missing data, aside from a handful of files (out of total approx 2 million files) that return read errors.
Graham
Many Mac OS X (10.3.9) Many

Raid setup with g-tech raid

hello there.
bit confused by this raid business. Just got a 4tb g-tech raid system which has 2 drives of 2tb each.
It shows as one disk in finder. DO i only have to set 'RAID Type' to 'Mirrored RAID set' in disk utility? Do i need to make 2 partitions first?!
its just that it still shows 4tb available and i was presuming it would say 2tb?
any clarification welcome..! Dan

I don't know much about these, but since nobody who does has happened along . . .
There are different kinds of RAID. The two most common are:
A +Concatenated RAID Set+ (or RAID-0) combines multiple disks into one volume; the size is the total of all the disks in the set. It sounds like that's how yours is set up.
A +Mirrored RAID Set+ (often called RAID-1) keeps identical copies on each disk, so if one fails, you don't lose any data, and can keep running. The size is the size of the smallest disk in the set. It sounds like that's what you want.
If all the drives are in one enclosure, they're usually configured by hardware switches on the controller, or possibly software options you can set via an app that comes with the set.
Check with the maker (or instructions that came with the system) for setup information.

Xserve Raid Card battery conditioning

Good afternoon, I have observed that the raid controller on my Xserve run conditioning cycle the batery every 3 months... is that normal?
I appreciate a clarification on this.
Thanks.
Hipolito Lopes

Hi Hipolito,
3 months is normal. It's a shame you can't choose the time of day it does this but it's generally around the 6hour mark for the conditioning to take place.
Hope that helps
Beatle

Re-create a mirrored RAID after restoring backup

As a follow-up on my thread http://discussions.apple.com/thread.jspa?threadID=833779&tstart=0 :
On a XServe G5 with OS X Server 10.4.8 and a two-drive mirrored RAID I do backups by cloning the RAID onto an other disk (in bay three) using Intego's Personal Backup X4.
Now it may happen that there is some kind of data loss or an other emergency case that urges me to boot from the cloned HD. I would do this by simply removing the HD in bay one and replacing it with the clone, then restart the server (second HD of the RAID1 remains in bay two).
The information I got until yet says that the mirrored RAID is then broken and needs to be completely re-build. Is this correct so far ? I originally assumed that the clone "remembers" that it was part of a RAID system before and rebuilds automatically with the second HD still in place. Very likely wrong ?
So if I need to rebuild the clone from scratch, which steps do I follow exactly ? As far as I understand I have two choices:
(1)
Via command line (as root): diskutil createRAID mirror <set name> Journaled HFS+ <device identifier disk1> <device identifier disk2>
<set name>: Would be volume name of the RAID to create, right ?
<device identifier disk1/2>: What is this and how to determine ?
(2)
Via disk utility: Select one disk (first line on the left disk/volume list, e.g. "76,2 GB Hitachi 123ABC"), click in "RAID" tab, do settings.
Then drag & drop each harddisk from the left to the right list (again, the first line of each disk entry). Finally creating the RAID by clicking at the button "Generate".
I'm not sure, although I would likely handle this with the disk utility, it appears easier to me using the diskutil command in terminal. But maybe I simply lack of explaining the GUI... let me know how to post images and I will add some screenshots here.
Well, if my proceedings are too complex and there is something easier, please let me know. I'm open for everything.

>This part of my question originally referred to the fact that when booting from the cloned backup in bay one while still having the second disk of the old RAID in bay two, it could occur that the OS rebuilds the RAID by using the data on the remaining mirror disk in bay two, couldn't ? This would be not desired, because the data on that disk would be no longer usable. So just for clarification, not that we have a misunderstanding in this point.
I see what you're getting at, and I don't believe it's an issue.
Given three disks, disk0, disk1 and disk2, where disk0 and disk1 are mirrored and that mirror fails, the OS will not rebuild the mirror using disk1 and any other disk. It won't arbitrarily choose any disk in the system to rebuild on - you have to specifically mark as disk as usable, either by enabling auto-rebuild and adding a spare disk to the array, or by using 'repairMirror' and specifying the disk to use.
In any case, if you reboot the system from disk2, there is no way it will rebuild the array using that disk since that disk is the active boot drive and can not be unmounted (which is necessary before being added to an array).
So, no, there is no chance that the system will auto-rebuild the working half of the mirror with your backup disk.
>Okay, let's summarize... I would follow this procedure in order to boot from the cloned backup and build a new RAID:
1. Remove both disks from the XServe.
2. Insert the cloned backup disk and boot.
3. Open terminal, su root, execute:
diskutil enableRAID disk0
diskutil repairMirror disk0 disk1
This is not correct.
When you 'enableRAID disk0' the OS will create a new, virtual disk (likely 'disk2'). This is the mirror array (currently consisting of only one drive), and it is this virtual disk that you need to specify as the mirror to be repaired.
So, given disk0 and disk1 as being the physical disks in the server, the steps are more likely to be:
<pre class=command>diskutil enableRAID disk0</pre>
(this creates a new, virtual disk2)
<pre class=command>diskutil repairMirror disk2 disk1</pre>
(this adds physical disk1 to the degraded mirror disk2)
When complete disk2 will be a mirror RAID consisting of disk0 and disk1.
>When using "diskutil enableRAID", I cannot determine the label of the RAID as I would do with "diskutil createRAID" (in command line manual this is referred to "setname"). Which setname does my "new" RAID system get, a default one ?
You know, I have no idea
I've never noticed this before. If I think about it, I'd guess that the setname is based on the source disk name.
>> If you add a drive to a RAID it gets reformatted, so it doesn't matter what's
on it.
Even if it was part of the RAID system before (the original disk in bay two) ? It could still have the RAID information on it and try to rebuild RAID when getting mounted, couldn't ?
No. See my comments above. The OS won't auto-rebuild unless a) the RAID is set to auto-rebuild and b) there is a spare disk in the array. If either of those two conditions are not true then the disk acts as a degraded mirror and can be reformatted/reallocated as you like.

ASM - Concept - Clarification Request

Hello All,
I'm about to go ahead and install ASM for one of my clients. After going through the book ASM - Under the hood, I have a few clarifications, which I hope can be answered by the experts here.
1- Since ASM uses its our algorithm for mirroring - Can I have an in-pair number of disks in +DATA diskgroup? say 11 disks ?
2- In regards to Failure Groups, what is concept? Say I have 1 diskgroup +DATA - 4 disks - does failure groups mean that id Disk 1 goes, then move the primary extents to another disk, say disk 3.
- Can failure groups be in different diskgroups, lets say failure group for DATA disks, would be disk in RECOVERY ?
- Or are failure groups additional disks which just sit there and are activated if case of a disk failure
3- On installation, ASM 10gR2, are there any things a firs timer should watch out for.
4- Should I have a hot spare disk on a 15 disk array Dell MD1000 - is this really necessary - why? if one disk goes bad, then we can simpy change it. Does this make sense if I have 4 hour gold-support on site with a new disk?
Thank in advance for any assistance.
Jan

1. Yes, ASM will determine the most suitable block mirroring strategy regardless the number of disks in the diskgroup.
2. Failure groups affect how ASM mirrors blocks across them. By default, each disk is in its own failure group - it is assumed that each disk can fail independently of others. If you assign two different disks to the same failure group, you indicate that they are likely to fail together (for example, if they share the access path and controller for that access path fails,) so ASM will only create single mirror on them and will try to create another mirror in another failure group. For example, you assign disk1 and disk2 to the same failure group: ASM will never create a mirror of a block from disk1 on disk2, it will only mirror to a different failure group. Note that if your storage is already RAIDed, EXTERNAL redundancy diskgroups are pretty safe: hardware RAIDs are usually more efficient than NORMAL redundancy ASM groups while maintaining the same level of protection, thanks to hardware acceleration and large caches they sport these days.
3. Not really, as long as you follow the documented procedures and have Oracle patched to the current patchset level. However, if you employ ASMLIB, there might be issues that differ by the storage vendor.
4. If you are sure that no other disk will fail within those 4 hours, hot spare is probably not that necessary. If availability is of concern, always plan for the worst case though. Having hot spare will protect you from such second failure while the replacement is en route.
Regards,
Vladimir M. Zakharychev

RAID & ASM

hello we have the below scenario for HA, your help is appreciated
Hardware:
- 2 identical servers in primary & secondary mode (Node1 and Node2)
- Each node having 2 CPUs
- One 2TB SAN storage with 10*200GB HDs + 2 redundant FC switches
Sofware:
- Oracle 10gR2 sofware and instance to be installed and created locally on
native HD of each node with 1:1 RAID mirroring with another identical local
HD
- Nodes in a RAC configuration.
- Storage disk arrays managed using ASM.
HA to be implemented as follows:
- 1:1 Storage redundancy (Disks 1-5 as G1 primary ASM group; Disks 6-10 as G2 secondary ASM group) -- Allowing for storage redundancy
- Both are active at any given point in time. Each node takes over the complete load in case of failover -- Allowing for server redundancy.
Clarifications needed are:
- Is this scenario possible? Does this require any additional configuration or software? Is there anything missing?
- Does the SAN need to be confgured with RAID5? Without RAID, 1TB is effectively usable at any point in time with the other 1TB used up as an ASM
mirror. Is this correct?
is RAID needed since ASM does the SAME ( STRIPE AND MIRROR EVERYTHING)
thanks

Hi!
Hardware Checklist:
2 servers with local storage, 2cpu each
MUST HAVE 2 NICs
Software:
Need 4 Public Static IPs with default gateway
Need 2 Private Static IPs for Interconnect
So far as storage is concerned:
Will require 5 RAW partitions - Mirrored
(For OCR and Voting Disk)
- Approx size 1gb/2gb each
Mirroring is RECOMMENDED for redundancy
Here there are 2 possibilities:
1. Mirror at Hardware Level(RAID)
in ASM configure External Redundancy
2. Mirror using ASM redundancy levels Normal/High
Normal Redundancy - 2 way mirror
High - 3 way
Oracle 10gR2 Clusterware and database software is required to do the setup
- Atul Joshi

RAID Clarifications

Similar Messages

Maybe you are looking for