Linux – use blockdev –setra on a formatted device or exisiting RAID Array

linuxxfs

I have an existing RAID Array and I want to increase the readahead setting. But the device has already been formatted and contains data that I cannot afford to lose. I'm unfamiliar with the blockdev command. Is it safe to run blockdev –setra=xxxxx on devices that have already been formatted? Will running the following command destroy (delete) the data on my device? Is there any risk of this happening?

$ blockdev --setra 65536 /dev/md0

Not sure this is important, but here are the devices details:

$ xfs_info /data/d1

meta-data=/dev/md0               isize=256    agcount=16, agsize=8191936 blks
         =                       sectsz=512   attr=2
data     =                       bsize=4096   blocks=131070976, imaxpct=25
         =                       sunit=64     swidth=256 blks
naming   =version 2              bsize=4096   ascii-ci=0
log      =internal               bsize=4096   blocks=64000, version=2
         =                       sectsz=512   sunit=64 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0

$ sudo mdadm --detail /dev/md0
/dev/md0:
        Version : 1.2
  Creation Time : Thu May  5 17:23:35 2011
     Raid Level : raid0
     Array Size : 524283904 (500.00 GiB 536.87 GB)
   Raid Devices : 4
  Total Devices : 4
    Persistence : Superblock is persistent

    Update Time : Thu May  5 17:23:35 2011
          State : clean
 Active Devices : 4
Working Devices : 4
 Failed Devices : 0
  Spare Devices : 0

     Chunk Size : 256K

           Name : d1:0
           UUID : 16d7fee4:a8903d2f:28c65d2f:161006b5
         Events : 0

    Number   Major   Minor   RaidDevice State
       0     202      241        0      active sync   /dev/sdp1
       1     202      242        1      active sync   /dev/sdp2
       2     202      243        2      active sync   /dev/sdp3
       3     202      244        3      active sync   /dev/sdp4

Best Answer

"blockdev --setra" is perfectly safe. We have it in our init scripts to run on every boot on a couple of servers. You probably want to do the same after you have benchmarked a few values and decided which is best for your workload.