what r the steps to check hdd in RHEL 5

[venugopalanramraj]

in my server i have 4 hdd are connected to it, now i need to check the status. because i thought some hdd is not working properly

[Darren Eckhoff]

You can check it manually with smartctl, or have the system check it automatically/periodically with smartd.

Look at the man page on those.

[wobbe]

Check with dd.(read)

Do something like;
dd if=/dev/hda of=/dev/null bs=4096.

[venugopalanramraj]

thanks for ur reply

darren can you explain more and give some example to use the command

[venugopalanramraj]

thanks for ur reply

can u tell what the dd command will do while we are passing it with example pls..

[Jayakrishnan G Naik]

HI Venu,

The dd command actually copies the contends of the disk and throw it to null, and the I/O happening is raw I/O. By doing this the disk is actually read & if any physical defects in the disk, the dd will throw the I/O errors to you, thus you can identify the faults in your disks. You can use dd command to copy disk to disk (raw copy, not the block based filesystem copy)

Thanks & Regards
Jayakrishnan G Naik

[Alzhy]

what server and what hdds?

if you have a plain PC server (off the shelf motherboards or commercial PC) and your hdds aer plain SATA or SAS drives -- then the utilities within RHEL or any Linux distro follow the SMART specifications. Your system automatically monitors proactively according to SMART continous analysis.

Try using hdparm or smartctl. Or if you have the GUI up -- there should be a disk tool which allows you to sum up all HDDs and their SMART status.

[venugopalanramraj]

hi Alzhy

it is hp server and sas hard disk

[Steven Schweda]

> what server and what hdds?

> it is hp server and sas hard disk

And HP only ever made one model of server?
And there's only one type of "sas hard disk"
in the world?

> [...] i thought some hdd is not working
> properly

What made you think that? Is there some
actual problem which you are trying to solve?