High Disk Usage

[darmoran]

Hey Guys,

First time posting on here. I'm still fairly new to the more advanced side of WHM/cPanel Administration (Have Been working with Linux/WHM for 2 years; part of that as a L1/2 Tech at a hosting company)

I recently have been experiencing issues with high load, which appears to be caused by disk utilisation (according to iostat), This has only happened after I attempted to enable quota's on my system last week via WHM itself.

According to Top, The CPU usage is fairly minimal, however at these times the load is peaking around 4-5 which is basically killing the sites which are running from the server.

Any Idea's on how I could troubleshoot/fix this issue?

OS is CentOS 5.4 Running cPanel 11.24.5-S38506 - WHM 11.24.2 - X 3.9

iostat readout:

Device:         rrqm/s   wrqm/s   r/s   w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await  svctm  %util
sda               0.00    39.50  0.50  4.50     0.00     0.02    10.40     0.35    0.80  68.80  34.40

Device:         rrqm/s   wrqm/s   r/s   w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await  svctm  %util
sda               0.00    31.50  0.00  2.50     0.00     0.28   230.40     1.00  436.80 400.00 100.00

Device:         rrqm/s   wrqm/s   r/s   w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await  svctm  %util
sda               0.00    16.50  0.00  6.50     0.00     0.09    28.31     0.52   42.69  80.00  52.00

Device:         rrqm/s   wrqm/s   r/s   w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await  svctm  %util
sda               0.00    38.31  1.00  5.47     0.00     0.06    19.08     1.84  348.31 153.92  99.55

Device:         rrqm/s   wrqm/s   r/s   w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await  svctm  %util
sda               0.00    13.50  0.00  7.00     0.00     0.17    50.86     1.71  215.14 142.86 100.00

Thanks for any help.

[cPanelDon]

I've relocated this thread into the cPanel and WHM Optimization discussions forum as it relates to improving performance and methods for optimization.

[darmoran]

Hi Don,

Thanks for that.

Just a little Addition. I have also noticed that the Memory is being hammered as well as the Disk.

Mem:   2065964k total,  1797340k used,   268624k free,   843564k buffers

[serversignature]

You can check the output of the following

root@desktop:~#hdparm -T /dev/sda
/dev/sda:
 Timing cached reads:   2254 MB in  1.99 seconds = 1131.64 MB/sec

This displays the speed of reading directly from the Linux buffer cache without disk access. This measurement is essentially an indication of the throughput of the processor, cache, and memory of the system under test.

and

root@desktop:~# hdparm -t /dev/sda

/dev/sda:
Timing buffered disk reads:  118 MB in  3.01 seconds =  39.16 MB/sec

This measurement is an indication of how fast the drive can sustain sequential data reads under Linux, without any filesystem overhead.

Kiran

[darmoran]

Thanks for your reply Kiran.

/dev/sda:
 Timing cached reads:   24880 MB in  2.00 seconds = 12467.74 MB/sec

/dev/sda:
 Timing buffered disk reads:    2 MB in  8.48 seconds = 241.38 kB/sec

Judging by that... I'm thinking that the Disk might be faulty/dying :|

[serversignature]

also, see what are the settings on the disks?

hdparm /dev/sda

hdparm -i /dev/sda

-
Kiran