Safety of Consistency Check

Any application like that will have an impact on the server. Try scheduling the task for A/Hours.

Message Edited by barhampa on 03-28-2006 11:31 AM

I can schedule after hours. Performance wouldn't be a consideration. However, server-based applications, particularly IIS and/or SQL might be running. It would be tolerable for them to slow down for awhile.

What I'd like to know is whether the consistency check itself might do damage to these applications (like damage a database, for example).

Thanks for your reply.

Linda

I have not run into any issues with running the consistancy check on these types of servers.  However, the consistancy check is really only neccessary if you have recently replaced a drive in a raid.  Otherwise it is basically verifying that the redundant data on the drives is correct.

Thank you.

Once when we had a problem and had to rebuild an array, the Dell tech suggest that consistence check be run once a month in order to catch problems early.

Linda

aafoglia wrote:

"I have not run into any issues with running the consistancy check on these types of servers.  However, the consistancy check is really only neccessary if you have recently replaced a drive in a raid.  Otherwise it is basically verifying that the redundant data on the drives is correct. :"

On the contrary Dell recommend doing at least one consistency check per month.
Please refer to the following extract from the Maintenance Best Practices for Direct-Attached SCSI Solutions document

Consistency checks for RAID Arrays
RAID arrays are an industry standard for the protection of important data through redundancy. This redundancy may take the form of parity calculations that are dispersed throughout the array. It could also be the simple mirroring of data to maintain a complete copy that does not require parity calculations to reconstruct the missing elements.

Hard drive media defects and other drive quality issues have steadily improved over time, even as drive sizes have grown substantially. However, hard drives are not expected to be totally free of flaws. In addition, normal wear on a drive may result in an increase in media defects, or “grown defects,” over time. The data block containing the defect becomes unusable and must be “remapped” to another location on the drive.

If a bad block is encountered during a normal write operation, the controller marks that block as bad and the block is added to the “grown defects list” in the drive’s NVRAM. That write operation is not complete until the data is properly written in a remapped location. When a bad block is encountered during a normal read operation, the controller will reconstruct the missing data from parity operations and remap the data to the new location.

A condition known as a double fault occurs when a RAID controller encounters a bad block on a drive in a RAID volume and then encounters an additional bad block on another hard drive in the same data stripe. This double fault scenario can also occur while rebuilding a degraded logical drive, leaving the controller with insufficient parity information to reconstruct the data stripe. The end result is a rebuild failure with the loss of data in that stripe.

Consistency checks will perform read operations on both the user data areas of the logical drive as well as the currently unused areas not containing any user data. This read operation causes the controller to record any bad block it encounters in the grown defects list so it is not used in the future (all RAID levels). The data in the damaged block is reconstructed from parity operations and written to a new location on the hard drive. The data stripe is completely recovered and now in an optimal state. The consistency check greatly reduces the risk of a rebuild not completing due to a defect on a drive in an area that has not been recently accessed or is outside the user data area. To avoid downtime and ensure data integrity, consistency checks should be part of the routine maintenance of any RAID system. 

A minimum recommendation is to perform consistency checks on each RAID logical drive once per month. This will increase the chance that any media defects that may be encountered are repaired and the data restored using parity data from the other members in the array (RAID 1, 5, 10 and 50). 

The amount of data in a given disk group has grown dramatically over the last few years as data storage needs have mushroomed in the business world. Additionally, Service Level Agreements continue to become more aggressive and data access is more mission critical in almost every business environment. These factors combine to create tension between business needs and technical realities. Backup and restore windows have shrunk almost inversely to the growth of data stores, creating a scenario whereby data recovery has become less timely due to data store sizes, while the need for the availability of that data has become more business-critical. Consideration should be given to the maximum volume size which can be useful in a given data storage situation This strategy will aid in avoiding the challenges of mounting, handling, and maintaining very large file systems coupled with very long restore windows in the event of data corruption or loss. Purposefully limiting data store sizes also reduces the numerical probability of encountering additional bad blocks during rebuilds. A similar benefit accrues to limiting data store sizes by reducing the amount of time and system load required to perform consistency checking on a given logical drive.

Additional consideration may be given to performing intermediary backups to disk, which are then shuttled to tape as a background operation. This will decrease the backup window while also making the most recent full backup available in a faster media than going back to tape to perform a restore.

To avoid downtime and data loss, follow the steps mentioned below to schedule consistency checks on a regular basis (once every month):

 

For more info please read through the complete document

 

Message Edited by barhampa on 05-20-200611:25 AM

Message Edited by barhampa on 05-20-200611:27 AM

After uninstalling Array Manager and installing Storage Management you will no longer be able to schedule consistency checks using the article found here

http://support.dell.com/support/topics/global.aspx/support/dsn/en/document?docid=DBFA11C643BEE26CE030A68F27284A07&c=us&l=en&s=biz

You can manually run a consistency check in Storage Management using these instructions http://support.dell.com/support/edocs/software/OMSS10UG/vdmgmt.html#323112

or if you save the CLI commands below to a batch file than use Windows Task Scheduler they can be run during off peak hours, which is more beneficial.

Instructions on how to run the CLI Commands using Task Scheduler

Go to a command line and run the following command to display the virtual disks and controllers. You do this because in the batch file you will have to specify the controller and virtual disk you would like to scan.

omreport storage controller

to display the controller IDs

omreport storage vdisk controller=ID

to display the IDs for the virtual disks attached to the controller.

Once you have the information above you will next open notepad and save the file to the root of C as cons_check.bat or a similar name. You will need to create one for each virtual disk you would like to check.

Inside the file you will add the necessary commands to check the consistency on the selected virtual disk and controller.

Example Syntax

For example, to run a check consistency on virtual disk 4 on controller 1, you would enter the text below into the cons_check.bat file and save it.

omconfig storage vdisk action=checkconsistency controller=1 vdisk=4

Once that is complete you will use the Windows Task Scheduler Wizard to run the batch file during off peak hours.

1. Click the Start button, point to Programs, point to Accessories, and then click Add Scheduled Task.

The Scheduled Task Wizard appears.

2. Click next and follow the onscreen directions. Click Browse and locate the batch file you created above

NOTE: The minimum recommendation for this task is to be run at least once a month

Thank You,

Message Edited by pcmeiners on 11-04-2006 04:52 PM