Clone SBS 2008 OS drive
I have a SBS 2008 with the OS on a 160GB drive that is starting to fail and 4 x 2 TB drives in RAID 5 for storage. The C: drive is starting to say there is file corruption, and last week the server was rebooted and booted to a blinking cursor. After repairing the MBR and BCD it came back online. However while it was down we ran chkdsk /f and it returned ZERO errors. Very odd as event viewer and the failing to boot indicates there is corruption.
I am concerned if the server reboots again it will not come back online yet again. Can anyone please suggest a method to clone, while server is online, the 160GB C: drive to either a 500 GB drive I've purchased, or ideally I'd like to put it on an 80 GB solid state I purchased as the C: drive is only using 50/160GB. But really I just want to get the OS on another drive ASAP.
We have a storagecraft backup solution and the backups were shot due to the file corruption on the C: drive. Again an online cloning solution that will not be compromised by existing file corruption.
Thank you in advance.
I am concerned if the server reboots again it will not come back online yet again. Can anyone please suggest a method to clone, while server is online, the 160GB C: drive to either a 500 GB drive I've purchased, or ideally I'd like to put it on an 80 GB solid state I purchased as the C: drive is only using 50/160GB. But really I just want to get the OS on another drive ASAP.
We have a storagecraft backup solution and the backups were shot due to the file corruption on the C: drive. Again an online cloning solution that will not be compromised by existing file corruption.
Thank you in advance.
Start command line and type there CHKDSK c: then hit enter. It will show the test results.
Is there any bytes in bad sector listed?
Is there any bytes in bad sector listed?
ASKER
No it checked out clean. But our backups were useless and event logs show NTFS corruption.
ok. It could be temporarily problems with file system that should be fixed by CHKDSK c:/f
ASKER
CHKDSK c: /f took a while i thought it was finding and fixing errors but it says everything's clean after.
ASKER
This is also what it said while in recovery console when the server was down. CHKDSK never reports and errors but event logs say otherwise.
Can you give us exact message of EventLog report?
ASKER
Event 55 NTFS
The file system structure on the disk is corrupt and unusable. Please run the chkdsk utility on the volume .
The file system structure on the disk is corrupt and unusable. Please run the chkdsk utility on the volume .
Have a look on this thread: http://social.microsoft.com/forums/en-US/whshardware/thread/d59e62e7-c2b9-449a-aca2-e8a2304b03af
The problem is not necessarily in NTFS file system. It turns out that bad driver could also cause the problem. I suggest you updating SATA controller driver first.
The problem is not necessarily in NTFS file system. It turns out that bad driver could also cause the problem. I suggest you updating SATA controller driver first.
I am leaning toward this being a RAID or SATA controller issue. If you haven't changed the drivers recently I would suspect it is a hardware failure.
So let me get this straight,you have one hard disk for your OS and you use RAID 5 for you data dives?
Your OS partition needs to be in the array too!
You need to rethink your setup completely as you are just waiting for a disaster to happen.
Dude,you are probably running desktop drives in a RAID 5 environment,which is a very bad idea.
RAID drives ship with something called TLER turned on,desktop drives do not.
What can happen is a drive can take up to 30 seconds to try and recover a bad sector and if this was a desktop,that may be a good thing.
However if it is hardware RAID,after about 3 to 7 secs,the controller says bad disk and throws it out of the array.
If you need to do a rebuild,there is a 30% chance a 2nd drive will fail and guess what,you data is toast.
You need to setup a RAID 6 or RAID 10 with real RAID drives,not desktop drives.
If you want to clone your OS drive use the built in backup in SBS 2008.
It makes a WIM image to another drive (USB or otherwise).you then replace the failing drive,boot the SBS 2008 recovery DVD and run the restore function to the new drive.
It should restore the image intact without affecting the old disk.
Your OS partition needs to be in the array too!
You need to rethink your setup completely as you are just waiting for a disaster to happen.
Dude,you are probably running desktop drives in a RAID 5 environment,which is a very bad idea.
RAID drives ship with something called TLER turned on,desktop drives do not.
What can happen is a drive can take up to 30 seconds to try and recover a bad sector and if this was a desktop,that may be a good thing.
However if it is hardware RAID,after about 3 to 7 secs,the controller says bad disk and throws it out of the array.
If you need to do a rebuild,there is a 30% chance a 2nd drive will fail and guess what,you data is toast.
You need to setup a RAID 6 or RAID 10 with real RAID drives,not desktop drives.
If you want to clone your OS drive use the built in backup in SBS 2008.
It makes a WIM image to another drive (USB or otherwise).you then replace the failing drive,boot the SBS 2008 recovery DVD and run the restore function to the new drive.
It should restore the image intact without affecting the old disk.
I've bought from these guys in the past.
http://www.coastalmicrosupply.com/certance-cl1101-ss-400-800gb-cl800-ultrium-lto-3-internal-p1766.html
http://www.coastalmicrosupply.com/certance-cl1101-ss-400-800gb-cl800-ultrium-lto-3-internal-p1766.html
opps wrong thread
ASKER
We just put in another drive and mirrored the failing drive C: and windows backups have been running twice daily. Our plan is to break the mirror, and boot off the good side of the mirror.
Anyone have any advice on best practices for breaking the mirror and booting up? I've read that we may need to repair the MBR. So with our SBS 08 disc we can boot off that and either run Microsoft restore utility off the backups it's been taking or run fix mbr?
Anyone have any advice on best practices for breaking the mirror and booting up? I've read that we may need to repair the MBR. So with our SBS 08 disc we can boot off that and either run Microsoft restore utility off the backups it's been taking or run fix mbr?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Just using the OS. I have an LSI RAID controller with 4 2TB drives in RAID 5 and then a 160 (C drive with OS) and 250 on the onboard sata in a RAID 1. Yes I only have one drive as my C (boot drive) but it is now in a mirror with the 250GB through OS RAID 1.
I've just ran chkdsk on the C: drive showing zero errors I've been focused on setting up some reliable redundancy first. Do you have any utilities to suggest using? The drive has been sensitive lately so I didn't want to run much else on it. While taking a backup last week it blue screened. But if you have some in mind I'll try running it.
What can I do to avoid transferring the same problem to a different disk?
I've just ran chkdsk on the C: drive showing zero errors I've been focused on setting up some reliable redundancy first. Do you have any utilities to suggest using? The drive has been sensitive lately so I didn't want to run much else on it. While taking a backup last week it blue screened. But if you have some in mind I'll try running it.
What can I do to avoid transferring the same problem to a different disk?
Well I can't guarantee anything,but depending upon your drive vendor ,there are some very good HW diags.
WD and Seagate have a standalone diags that are quite reliable if there is an issue with the drive itself.
http://support.wdc.com/product/download.asp?lang=en
http://www.seagate.com/www/en-us/support/downloads/seatools
HD Tune is a limited use freebie that can be used to troubleshoot disk issues too.
http://www.hdtune.com/
You having the onboard controller and the OS doing the RAID is redundant and could cause issues .
Pick one or the other ,not both.
And I will riterate this one more time.having desktop 2 TB drives in a RAID 5 config is a very bad idea.
Seagate and WD specifically label which drives are which.
Make sure you are using ones meant for RAID.
Your chances of a false positive on one of the hard drives can mean lot's of issues with the controller having to rebuild a 2 TB disk taking many hours and if a 2nd drive gets a false positive ,the array is gone and so is your data.
WD and Seagate have a standalone diags that are quite reliable if there is an issue with the drive itself.
http://support.wdc.com/product/download.asp?lang=en
http://www.seagate.com/www/en-us/support/downloads/seatools
HD Tune is a limited use freebie that can be used to troubleshoot disk issues too.
http://www.hdtune.com/
You having the onboard controller and the OS doing the RAID is redundant and could cause issues .
Pick one or the other ,not both.
And I will riterate this one more time.having desktop 2 TB drives in a RAID 5 config is a very bad idea.
Seagate and WD specifically label which drives are which.
Make sure you are using ones meant for RAID.
Your chances of a false positive on one of the hard drives can mean lot's of issues with the controller having to rebuild a 2 TB disk taking many hours and if a 2nd drive gets a false positive ,the array is gone and so is your data.
To brake the mirror right click on volume C: on disk you want to boot from and select Brake Mirror. Same for other partitions.
After that turn off the machine - disconnect one drive and boot from another.
After that turn off the machine - disconnect one drive and boot from another.
ASKER
Thank you, we broke mirror, ran fixboot and it booted up on secondary drive.
http://www.drivesnapshot.de/en/