Ext2 or ext3 for large RAID array

I'm just in the process of creating a 10TB array of 5 x 2TB drives.
I've been burned too many times by EXT4 so it's out for the forseable future.
My concern is the crazy amount of time required to stabalize the file system when periodic checks are mandated. I'm using ext3 right now on a 7.5TB file system and have tuned the auto checking down to 2 years and 100 mounts. It's not the best situation but when the system goes down due to over heating (filter plugs every few months), I turn it on, and it goes into a 2 day recovery procedure during the boot process, it's outside the envelope of acceptibility.
Last edited by TomB17 (2010-06-26 02:32:55)

TomB17 wrote:
I appreciate the comments, gentlemen.
graysky wrote:Not what you asked but can you describe how you have been burnt by ext4?
I've been burned by the 0 byte file bug. The files were all there but some of them went to 0 bytes.
I did that on a backup array about 6~8 months ago. Thinking, "It's just a backup array", I tried EXT4 for the first time. It formatted up nicely, 36 hours of rsync, and I was good to go. I didn't realize I had the 0 byte file issue until my main array had some issues. When I went to the backup array, there were tons of 0 byte files, including fstab, and mdadm.conf. That made it more difficult to rebuild the main array. I did manage to rebuild the main array. Once done, I formatted the backup array EXT3 and I've been hessitante to experiment with filesystems.
The 0 byte file bug is well documented, and perhaps long solved, but I'm not ready to get back on that bandwagon.
For what it's worth, I was burned by EXT3 several times early in it's existance. That was a different issue. The whole filesystem would become corrupt after a while. It was disasterous but I didn't count on my PC then the way I do now. That was back in the days I could back up to CD-ROM. I kept at it and eventually EXT3 stabalized. These days, I trust EXT3 with my life.
I encountered very similar issues, which resulted in me switching this workstation to FreeBSD and using ZFS for my raid arrays.
The beauty of that file system far outweighs anything available on Linux at this current time.

Similar Messages

UnmarshalException for large byte array :: Weblogic 8.1 SP2

Hi,
1) We have a application running on Weblogic 8.1SP2.
2) Our scenario consist of Client code which reads a input file via Stream and converts it to a byte array before sending it to the Server.
When the server side code calls the Remote Bean (with byte array as an argument) it throws this exception :
"java.rmi.UnmarshalException: Software caused connection abort: socket write error"
Intermittently also got this error :: "weblogic.rjvm.PeerGoneException: ; nested exception is: java.io.EOFException"
3) This scenario works absolutely fine if the said Input File is of a smaller size.
4) I have also set the following params in config.xml <server .../> tag, but all in vain:
     CompleteMessageTimeout="480" also tried with "0"
     MaxMessageSize="2000000000"
     NativeIOEnabled="true"
Also tried with :
     MaxHTTPMessageSize="2000000000" & MaxT3MessageSize="2000000000"
Below is the exception trace:
java.rmi.UnmarshalException: Software caused connection abort: socket write error; nested exception is: java.net.SocketException: Software caused connection abort: socket write error
java.rmi.UnmarshalException: Software caused connection abort: socket write error; nested exception is: java.net.SocketException: Software caused connection abort: socket write error at weblogic.rmi.cluster.ReplicaAwareRemoteRef.invoke(ReplicaAwareRemoteRef.java:297)
     at weblogic.rmi.cluster.ReplicaAwareRemoteRef.invoke(ReplicaAwareRemoteRef.java:244)
Caused by: java.net.SocketException: Software caused connection abort: socket write error
     at weblogic.rjvm.BasicOutboundRequest.sendReceive(BasicOutboundRequest.java:108)
     at weblogic.rmi.cluster.ReplicaAwareRemoteRef.invoke(ReplicaAwareRemoteRef.java:284)
     ... 6 more
Caused by: java.net.SocketException: Software caused connection abort: socket write error
     at java.net.SocketOutputStream.socketWrite0(Native Method)
     at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:92)
     at java.net.SocketOutputStream.write(SocketOutputStream.java:136)
     at weblogic.socket.SocketMuxer.write(SocketMuxer.java:721)
     at weblogic.rjvm.t3.T3JVMConnection.sendMsg(T3JVMConnection.java:723)
     at weblogic.rjvm.MsgAbbrevJVMConnection.sendOutMsg(MsgAbbrevJVMConnection.java:276)
     at weblogic.rjvm.MsgAbbrevJVMConnection.sendMsg(MsgAbbrevJVMConnection.java:164)
     at weblogic.rjvm.ConnectionManager.sendMsg(ConnectionManager.java:549)
     at weblogic.rjvm.RJVMImpl.send(RJVMImpl.java:722)
     at weblogic.rjvm.MsgAbbrevOutputStream.flushAndSendRaw(MsgAbbrevOutputStream.java:292)
     at weblogic.rjvm.MsgAbbrevOutputStream.flushAndSend(MsgAbbrevOutputStream.java:300)
     at weblogic.rjvm.MsgAbbrevOutputStream.sendRecv(MsgAbbrevOutputStream.java:322)
     at weblogic.rjvm.BasicOutboundRequest.sendReceive(BasicOutboundRequest.java:103)
     ... 7 more
Any thoughts would be helpful?
Thanks,
Khyati

Hi,
1) We have a application running on Weblogic 8.1SP2.
2) Our scenario consist of Client code which reads a input file via Stream and converts it to a byte array before sending it to the Server.
When the server side code calls the Remote Bean (with byte array as an argument) it throws this exception :
"java.rmi.UnmarshalException: Software caused connection abort: socket write error"
Intermittently also got this error :: "weblogic.rjvm.PeerGoneException: ; nested exception is: java.io.EOFException"
3) This scenario works absolutely fine if the said Input File is of a smaller size.
4) I have also set the following params in config.xml <server .../> tag, but all in vain:
     CompleteMessageTimeout="480" also tried with "0"
     MaxMessageSize="2000000000"
     NativeIOEnabled="true"
Also tried with :
     MaxHTTPMessageSize="2000000000" & MaxT3MessageSize="2000000000"
Below is the exception trace:
java.rmi.UnmarshalException: Software caused connection abort: socket write error; nested exception is: java.net.SocketException: Software caused connection abort: socket write error
java.rmi.UnmarshalException: Software caused connection abort: socket write error; nested exception is: java.net.SocketException: Software caused connection abort: socket write error at weblogic.rmi.cluster.ReplicaAwareRemoteRef.invoke(ReplicaAwareRemoteRef.java:297)
     at weblogic.rmi.cluster.ReplicaAwareRemoteRef.invoke(ReplicaAwareRemoteRef.java:244)
Caused by: java.net.SocketException: Software caused connection abort: socket write error
     at weblogic.rjvm.BasicOutboundRequest.sendReceive(BasicOutboundRequest.java:108)
     at weblogic.rmi.cluster.ReplicaAwareRemoteRef.invoke(ReplicaAwareRemoteRef.java:284)
     ... 6 more
Caused by: java.net.SocketException: Software caused connection abort: socket write error
     at java.net.SocketOutputStream.socketWrite0(Native Method)
     at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:92)
     at java.net.SocketOutputStream.write(SocketOutputStream.java:136)
     at weblogic.socket.SocketMuxer.write(SocketMuxer.java:721)
     at weblogic.rjvm.t3.T3JVMConnection.sendMsg(T3JVMConnection.java:723)
     at weblogic.rjvm.MsgAbbrevJVMConnection.sendOutMsg(MsgAbbrevJVMConnection.java:276)
     at weblogic.rjvm.MsgAbbrevJVMConnection.sendMsg(MsgAbbrevJVMConnection.java:164)
     at weblogic.rjvm.ConnectionManager.sendMsg(ConnectionManager.java:549)
     at weblogic.rjvm.RJVMImpl.send(RJVMImpl.java:722)
     at weblogic.rjvm.MsgAbbrevOutputStream.flushAndSendRaw(MsgAbbrevOutputStream.java:292)
     at weblogic.rjvm.MsgAbbrevOutputStream.flushAndSend(MsgAbbrevOutputStream.java:300)
     at weblogic.rjvm.MsgAbbrevOutputStream.sendRecv(MsgAbbrevOutputStream.java:322)
     at weblogic.rjvm.BasicOutboundRequest.sendReceive(BasicOutboundRequest.java:103)
     ... 7 more
Any thoughts would be helpful?
Thanks,
Khyati

Fragment Size with large Raid Arrays

No, I am not a new user. I've done newfs probably hundreds
of times. But, I've never had to futz around with obscure
parameters because the device geometry was insane.
I want to get a file system made that
doesn't require 4096 byte fragments. Is that too much to ask?
I would think that there is some combination of cylinders,
heads, and sectors that should give that result, but I don't
have time to search for it.
tahoe# newfs -m 1 -o time /dev/rdsk/c2t8d0s2
newfs: construct a new file system /dev/rdsk/c2t8d0s2: (y/n)? y
With 16129 sectors per cylinder, minimum cylinders per group is 16
and the fragment size to be changed from 0 to 4096
With these parameters, it returns immediately, seemingly doing
nothing -f 4096 to specify the fragment size, like:
tahoe# newfs -f 4096 -m 1 -o time /dev/rdsk/c2t8d0s2
This worked, or seemed to.
The problem now is that it I am losing a lot of space when I have
directories with lots of little files.

The relocation you show is a relocation for a 32-bit value
relative to a linker section, in this case the .lbss section.
I would not expect such a relocation to be generated
unless the object file was produced by compiling for the
small memory model.
What type of symbol is maillage_curv.curv_rank_from_index_?
From the name, I would expect it to be a routine name, but
routine names should not be defined in the .lbss section.
Bob Corbett

Image Software on nVidia RAID arrays

Has anyone been able to get a disk imaging program to work with the nVidia RAID array? I have tried Acronis True Image, Norton Ghost 9.0 and the old Drive Image. I have been unable get the rescue/recovery disks to recognize the RAID Array. With Acronis, it appears that they don't have the Linux drivers for the RAID array. Ghost 9.0 is supposed to support RAID arrays, but when I try to load the drivers, I get a message that it can't find nvraid.sys, despite the fact that I supply that file on the floppy disk. I spent 1.5 hours on the phone with Symantec Technical Support and their only answer was that I would receive a phone call from a supervisor in 3 days.
I can get Norton Ghost to work from within windows, but in order to restore an image of my C: drive I need to gain access to the RAID array from the recovery disk.

I have sort of got it to work with ghost9 .
I have written a plug-in for nvraid/nvide for the bartWinPE cd made with PEBuilder 3032.
V2I Images can then be browsed with imagebrowser and files/catalogs be restored
to arrays partitions .
The thing that does not yet work properly is the V2I protector executable cause the bart plug-in for .Net gives problems .(not sorted out this yet)
The BartWinPE cd is also of course equipped with ghost9 plugin .
So basicly the PC is booted with bootable Windows PE cd (like the ghost9 CD) ,
windows runs from cd&ram . And all the nvraid arrays is visible .Restoration made
with V2I image browser .
The plug-in ( INF file ) :
Code: [Select]
; PE Builder v3 plug-in INF file
; NVIDIA RAID (Windows XP)
; Created by Syar2003
[Version]
Signature= "$Windows NT$"
[PEBuilder]
Name="DSK_NvRaid"
Enable=1
[SourceDisksFiles]
nvraid.sys=4,,1
nvatabus.sys=4,,1
[SetValue]
"txtsetup.sif","SourceDisksFiles","nvraid.sys", "1,,,,,,_x,4,1,0,0"
"txtsetup.sif","SourceDisksFiles","NvAtaBus.sys", "1,,,,,,_x,4,1,0,0"
"txtsetup.sif","SCSI.Load","nvatabus", "nvatabus.sys,4"
"txtsetup.sif","SCSI.Load","nvraid", "nvraid.sys,4"
"txtsetup.sif","SCSI","nvatabus", """NVIDIA NForce Storage Controller"""
"txtsetup.sif","SCSI","nvraid", """NVIDIA RAID CLASS DRIVER"""
"txtsetup.sif","HardwareIdsDatabase","PCI\VEN_10DE&DEV_008E", """nvatabus"""
"txtsetup.sif","HardwareIdsDatabase","PCI\VEN_10DE&DEV_0085", """nvatabus"""
"txtsetup.sif","HardwareIdsDatabase","PCI\VEN_10DE&DEV_00D5", """nvatabus"""
"txtsetup.sif","HardwareIdsDatabase","PCI\VEN_10DE&DEV_00EE", """nvatabus"""
"txtsetup.sif","HardwareIdsDatabase","PCI\VEN_10DE&DEV_00E3", """nvatabus"""
"txtsetup.sif","HardwareIdsDatabase","PCI\VEN_10DE&DEV_00E5", """nvatabus"""
"txtsetup.sif","HardwareIdsDatabase","*_NVRAIDBUS", """nvraid"""
"txtsetup.sif","HardwareIdsDatabase","GenNvRaidDisk", """nvraid"""
If the BartWinPE cd is equipped with ghost8 you should be able to restore
full disks/partitions with the *.gho extension .
But then the images must be buildt with ghost8 (from the BartWinPE cd first).
Though i have not had the time to test that yet .
Links:
Plug-in repository
http://www.reatogo.de/BartPE/BartPE_plugins/repository.htm
PeBuilder
http://www.nu2.nu/pebuilder/

Looking for Thunderbolt 2 RAID Array options that can match the new Mac Pro's quietness -- or at least come close!

Maybe it is too early to talk about anything beyond the Pegasus2 Series (http://www.promise.com/promotion_page/promotion_page.aspx?region=en-global&rsn=1 00) when it comes to Tunderbolt 2 storage options. Looking for something on the order of 8 to 12 usable TB of redundant storage or possibly a JBOD option of 4 bays. This is coming close but I do not think it is Thunderbolt 2 (http://www.g-technology.com/products/g-raid-thunderbolt).
Either way -- looking for users experience not only on usable Thunderbolt 2 solutions but more importantly noisiness. I want to upgrade my existing Mac Pro to one of the new ones and I have found the old one to be nearly unnoticable in a quiet desktop environment. Would love to find the same peace and quiet when migrating to the new form factor that is going to need external storage to match what I have been used to inside the old Mac Pro.

Not your quiet storage, but then I would put it in a cool closet or small room off to the side :-)
http://www.macgurus.com/store/Item/ARC8050 ( I was assuming but maybe not, that it would support RAID6 which is always preferable for larger arrays. Seems to have most features you want for 8 drives). MacGurus "Burly" are never quiet but they are rugged and well tested.
Migrating storage will be a concern. I don't like the noise from my older SATA OWC dual-drive metal enclosure though suppose to be 'acceptable' unless they have improved the fans they are out for me.
OWC for Thunderbolt products.
Helios enclosure (the new dual-slot Helios PCIe chassis)
… moving various hard drives into Thunderbolt cases
USB3 hard drive enclosure.
For those without eSATA enclosures, a simple eSATA to USB3 adapter might do the trick.
USB3 is plenty fast to house two hard drives.
OWC Mercury Elite Pro Dual.
Available now starting at about $430.
review of the OWC Mercury Elite Pro Dual.
Some upcoming new TB2 enclosures:
http://www.sonnettech.com/product/thunderbolt/index.html
http://gigaom.com/2013/12/21/the-technology-gap-between-the-old-and-new-mac-pro- could-cost-you-a-mini/

HT201250 Using Time Capsule for Backing Up a RAID Array

I am setting up an 8TB RAID 0 array at home. I am going to use 4 - 2TB FW drives. I already have an SSD in my 3.4 Ghz i7 iMac but am speed hungry and since I have the drives, I figure I would play a bit. I know it is vital to have solid back-up for this, but I am unsure of the best back-up solution. I am wondering if I can attach 3 - 3TB drives to my 1TB Time Capsule and use the combined 10TB in Time Machine to back it up? To date I have backed up each drive separately, but am realizing the folly of that and want to explore RAID 0 at the same time as explore a new back-up solution. Everything I read tells me I need a drive for backing up the RAID array that is at least as large as the combined total of the array. That doesn't make sense if the array is so large... I can't find anywhere if Time Machine supports spanning mulitple back-up drives.

According to the document linked below, you can, but it may not have the performance to back up that much data. Over Wi-Fi, forget it.
Uses for the USB port of Time Capsule, AirPort Extreme, AirPort Express

What Is the Best Raid Array Setup for Photoshop?

I'm considering going to a Raid Array both at the Scractch Disk Level 2X Solid State 90GB Drives, and also at the stroage level 3X Hitachi 2TB HDD.
Which Raid Array setups should I use for each?

Something to consider:
An advantage of RAID 0 is that you get all the space of both drives, while multiplying the throughput.
While RAID 0 also multiplies the possibility of failure, you can offset that by buying high-MTBF enterprise class drives. The Western Digital RE4 is an example (and is the model I use in my RAID). Such drives are not really that much more expensive than their consumer-grade counterparts, and the difference is well worth it, IMO.
Do good backups - e.g., to a large external USB drive (e.g., Western Digital MyBook), and you'll be covered against loss of the entire array.
-Noel

Alert for failed disc in RAID array

is there a third party software that will alert when a disc in a raid array is degraded or failed? right now, i have to open raid utility to see this. i would like to get a small alert so i can replace the disc quickly.
any ideas welcome.
thanks

SoftRAID http://www.softraid.com
I think there is something like <aa href=http://www.macupdate.com/search.php?arch=all&keywords=RAID&os=macosx> "RAID monitor"
( not rocket science )
Search http://www.macupdate.com
(as well as the usually news, tip, sites) on reports of how well it WORKS and for sending email alerts etc even.
There should be entries in system log of repeated I/O requests and failures, and look for attempts to reassign bad blocks as well.

Sudden read-only on RAID array.

I don't know if this was caused by an update but suddently my /raid is mounted read-only. It doesn't happen right after the mount but suddently after 1 day of uptime or a few hours and the /raid partition is locked again.
[dezza@dezza ~]$ cd /raid/
[dezza@dezza raid]$ touch me
touch: cannot touch `me': Read-only file system
[dezza@dezza raid]$ df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 34G 13G 22G 38% /
none 1013M 0 1013M 0% /dev/shm
/dev/sda3 46M 12M 32M 27% /boot
/dev/md0 466G 364G 103G 79% /raid
[dezza@dezza raid]$ cat /etc/fstab
# /etc/fstab: static file system information
# <file system> <dir> <type> <options
> <dump> <pass>
none /dev/pts devpts defaults
0 0
none /dev/shm tmpfs defaults
0 0
# Optical drive, CD/DVD
/dev/sr0 /media iso9660,udf ro,user,
noauto,unhide 0 0
# Root, swap, boot
/dev/disk/by-label/root / reiserfs defaults
0 1
/dev/disk/by-label/swap swap swap defaults
0 0
/dev/disk/by-label/boot /boot ext2 defaults
0 1
# RAiD BACKUP (2x 500GB Western Digital 7200RPM)
/dev/md0 /raid reiserfs defaults
,sync,user 0 0
# External 1TB
/dev/disk/by-label/external /mnt/1tb ntfs-3g u
ser,sync,exec,uid=1000,gid=1000,locale=en_DK.utf8,force,noauto 0 0
# VirtualBox
none /proc/bus/usb usbfs auto,busgid=108,busmode=0775,devgid=108,devmode=0664,lo
cale=en_DK.utf8 0 0
[dezza@dezza raid]$
Last edited by dezza (2009-08-24 14:42:13)

I had the same problem with a hardware RAID array... I was installing Arch in an previously Redhat server.
The error I got just just before the read-only thing was related to the filesystem... (Arch was in reiserfs and /home in Redhat is ext3)
That read-only thing only happened when I was performing heavy load on the raid by rsyncing almost 300GB from RH to the new arch (No error while installing and configure a very complete system).
I though maybe it was that reiserfs wasn't good enough and I tried with no luck using jfs (the same I succesfully use in LVM in my laptop).
At the end I decided to get rid of the filesystem difference and It worked using ext3.
It worked perfectly for almost a month until last week that we had a crash copying 180GB from an USB external drive. We tried again and it worked with no problems. So in conclusion:
I used ext3 successfully but I have one crashed (after the read-only thing the system becomes useless) documented.
I know this is not an answer but I hope this help you or maybe help someone to give us a point.
P.S: The raid arrays are in the same computer (PERC6 I think) using the aacraid driver. One is raid 5 (RH) and the other one is RAID 0 (Arch),

NAS Raid Array

I have a NAS Raid Array and when I try to set up time machine, it does not come up in the options list for storage devices. Where is it?
Thanks,
Scott

A disk image is treated by the OS as a local file. So look into how to create a growable sparse image (specify a size of 1 TB or something and if it's a sparse image the initial size will be small). This trick also works for Aperture Vaults. The caveat being that the file will be opaque to any non-Apple clients... on the NAS it will just look like a huge "block" called <name>.sparseimage. Fire up Disk Utility and search the help on creating images, or go to Apple's Knowledge Base:
http://search.info.apple.com/
And search for sparse image.
One example is here:
http://docs.info.apple.com/article.html?path=DiskUtility/10.5/en/duh1527.html
Note you don't have to use encryption on the image... and it's faster if you don't.
One of the main issues in using a NAS is the file system of the NAS. HFS+ supports all manner of characters in the file name. You can have spaces, and many characters which are "illegal" on Windows (NTSF || FAT32) or Linux/UNIX (UFS, EXT2, EXT3, ReiserFS) file systems. So if you try and run Time Machine over an SFP or AFP backup to a NAS that uses some other file system, you've got issues. Hence the reason not to support any NAS other than OS X Server (which is obviously running OS X and thus has HFS+).

Is It Possible to Clone RAID Array in a Safe and Easy Way?

Why do computer users need to clone RAID array, especially Windows Server users need to do RAID cloning? Generally speaking, they need to clone RAID array regularly in order to upgrade disk or migrate data from small disk to a larger one. A typical
example is that when the partitions on the hardware RAID runs out of space, you may prefer to rebuild the hardware RAID with larger hard disks. Then, you need to backup the data to another place, rebuild the hardware RAID and restore data again.
What a time-consuming task!
Is it possible to clone RAID array in a safe and easy way? The answer is yes and this article will introduce a RAID cloning software.

Hi jiangchunli,
Were you looking to post this question in some other forum? This looks like related to RAID and Disk upgrades.
This forum is meant for
http://www.windowsazure.com/en-us/services/recovery-manager/
Please let us know if this is indeed Hyper-V Recovery Manager related so we can help better.
Thanks
Praveen
Praveen/ www.PraveenKumar.in

[VIA] K7T266 PRO- RU motherboard: Windows XP on a RAID ARRAY (fasttrak100 LITE)

Hi All!
I have a
MSI K7T266 PRO- RU motherboard, BIOS is the latest, official from MSI website,
Problem short description: Windows XP (sp1 or sp2 english, I tried both)
will not install when I set 2 x 40 Gig = ~80 Gig Stripe Array.
detailed description:
motherboard K7T266 PRO- RU BIOS 1.9 AMI BIOS.
AMD ATHLON XP 1700 + "Palomino" microprocessor, at default speeds.
one piece of 512 PC 3200 RAM, can tell details later.
Power supply: 400 W CODEGEN model : 300X.
HDD-s : one IBM deskstar, fully functional, and a SEAGATE barracuda. seagate SEATOOLS, IBM DFT 32 no errors both.
I set in the FASTTRAK 100 LITE "PDC20276R" (written on the chip) bios a STRIPE array. Powerquest drive magic , or the Windows XP CD handles it fine , as a 80 gig disk.
I downloaded ALL available versions of the FASTTAK 100 LITE driver from the MSI website.
(only tried: MSI driver FASTTRAK 100 lite, the only available driver on the MSI website, and various drivers from www.promise.com) ( http://www.promise.com/support/download/download2_eng.asp?productId=15&category=driver&os=0&go=GO ) (but I guess they WONT support my raid controller on the motherboard)
Windows XP CD , booting, Press F6 for additional RAID /SCSI drivers, I put in the floppy, choose WinXP fasttrak lite driver, windows continues, even formatting the 80 GIG disk, copy files- access the FASTTRAK.SYS driver, EITHER it says: no fixed disk present, press F3 to exit windows setup, or windows setup continues, copies files, reboots, and then STOP ERROR
http://www.martin555.dyndns.org/stop.jpg
please, help me with the following:
how to install XP on a raid array.
is it worth, will it be little bit faster disk load times etc?
is it secure if I trust the HDD-s?
any clue, how to overclock the AMD 1700+ athlon XP palomino to its maximum, but still stability?
AND : I read a lot about some MOD-ded BIOS called " KUNIBERT" , which makes the LITE raid to a FULL RAID and some other extras? where to get it and what to do?
email me if you can martin5 "at" freemail.hu
THANKS !

pardon me, dump? my english is not very perfect.. you mean to trow away, replace to something better, more stable power supply, more powerful?
you must see... first I would like to resolve the "installing XP on a raid array" problem,( how is it possible, is it worth, will it speed up hard disk access and load times... etc... ) and after that maybe I will OC the machine...
it is now , with a 8 cm ventillator and a large aluminium cooler 39 Celsius IDLE and goes to 45 Celsius during EVEREST benchmarking test. (temperatures on 1700+ default setting and OC-'d to 1900+. just experiencing with it. ) (I saw the toms hardware guide movie , smoking processors..)
cheers, Marton.

Workaround to use drives with 4096b block for mirrored RAID?

I have 2 Hitachi 2TB Drives. One is a few months old and uses 512b blocks, the other is the same model, but newer and uses 4096b blocks.
When I try to set up the RAID Set I get the error: MediaKit reports block size error, usually caused by not being a multiple of 512.
From reading the message board, it looks like this a known bug when trying to encrypt drives with 4K block sizes or add them to a RAID Array.
Is there any way to set these up as a mirrored RAID array other than waiting for apple to fix the problem? Are there third party apps that can be used to fix this problem?
Thank you for your time.

I have the exact same problem, and until Apple solves this bug, I no longer have a hard drive large enoug for my Time Machine back-ups. I have submitted the bug, but I really need a solution. Please help!

RAID arrays and Windows

Not sure where to post this one. You have been very helpful before so I’ll try again.
The machine is the Media Centre in my signature.
Installed two new Maxtor SATA drives, set up as RAID 0, clean install of XP MCE. Boots on the RAID array. Not a single problem. Installed MSI drivers, XFS video card drivers and KCorp PCI network adaptor for wireless link. Download and install all Windows updates. All OK.
On a later start up it spends long time trying to detect RAID array, then RAID error. Automatically reboots and detects a healthy raid array. Tries to open Windows but screen goes almost black but faint Windows loading page with the short activity bar showing no activity. If left it eventually springs to life. Tried starting in safe mode, no problem. Restore to an earlier restore point, no better.
The RAID problem is intermittent. The Windows problem looks as if it stalls for a variable length of time before it gets going.
Don’t know where to go next. Any suggestions?

fafner: I wanted a large disk to hold recorded TV progs, videos, CDs etc and 160GB is the max for this mainboard. I went for 2x160GB Maxtor drives. As I was using two drives I reasoned I might as well use RAID 0 to get a single volume and a faster machine. In particular I read somewhere that RAID 0 halved, or nearly so, the start up time. Who wants to wait a couple of minutes for the TV to fire up?
Fredrik: The XFX video card drivers are in fact nVidia drivers which are the latest.
Doctor Stu: I have subsequently installed XP on both drives individually. On one it worked well, on the other same old problem: either the drive not found in the BIOS after the heading ‘Detect IDE drives’ or the screen fades as Windows starts up. Checked both drives with Maxtor’s Powermax disk checking utility. Provided the drive was connected to the SATA1 socket and the other drive not connected both passed the Full Test. I could not get Powermax to recognise SATA2 even though both were detected correctly in the BIOS. My conclusion is that the problem was a dodgy connection which I hope has been rectified with all the swapping around.
Is there disk testing utility that puts the disc under load similar to Mem86 for memory?
Will I try RAID again? I doubt it even though it did start up in far less time.
My thanks to all.

Updating a Hardware RAID array

I'm running a recent Mac Pro model with Hardware RAID and four 1TB SATA drives.
Seagate 1.5TB drives are now available, and they're cheaper than the 1TB drives were just a few months ago.
My Infrant/Netgear ReadyNAS NV+ implements a proprietary form of RAID that they called X-RAID, similar to RAID-5 but with the feature that you can swap larger drives into the array one at a time, and when you're done, the RAID will be re-synced to make use of the additional storage without requiring that you dump and restore the contents of the array. The final re-sync is an admittedly long process (it took three days for mine to update from 4x750GB to 4x1TB) but it's still a very convenient feature.
Does Hardware RAID on Mac OS X have anything like this, i.e. can I replace 1TB drives one at a time with 1.5TB drives and preserve my data? And when I'm done, will the system recognize and use the new storage?
If not, could someone please describe the procedure that I would need to go through in order to achieve this? I presume that I would need to recreate the RAID array entirely, re-install the OS, and then restore the system from my Time Machine backup. But I'm not sure of the details.
I'd appreciate any help with this. Thanks.
An aside: I was more than a little upset a few months ago when I called Apple Support for help with initial setup of my RAID array (I have Apple Care) and was told "Sorry, we don't support Hardware RAID." So Apple was quite happy to sell me an $800 piece of hardware for my lavish new system but didn't bother to tell me at time of purchase that my configuration was "unsupported." It seems a little outrageous to me.
Given Apple's refusal to help me I need to ask the community for help with problems like this.

Hi rrgomes;
Never having heard of X-RAID but going by your comment that is similar to RAID5 I would have to say that you will not be able to upgrade by replacing one disk at a time.
As to no support for the RAID card, I personally would not have taken his word for that. Instead I would have escalated it up through Customer Service just to be sure before I gave up on that point. To me the response you got there sounds bogus.
Allan

Ext2 or ext3 for large RAID array

Similar Messages

Maybe you are looking for