1 |
On Sun, May 13, 2012 at 4:28 PM, walt <w41ter@×××××.com> wrote: |
2 |
> I have a usb3 docking station which is showing some behavior I don't |
3 |
> understand: |
4 |
> |
5 |
> #hdparm -t /dev/sdc |
6 |
> |
7 |
> /dev/sdc: |
8 |
> Timing buffered disk reads: 266 MB in 3.00 seconds = 88.59 MB/sec |
9 |
> Timing buffered disk reads: 268 MB in 3.01 seconds = 89.05 MB/sec |
10 |
> Timing buffered disk reads: 266 MB in 3.01 seconds = 88.43 MB/sec |
11 |
> Timing buffered disk reads: 266 MB in 3.02 seconds = 88.10 MB/sec |
12 |
> Timing buffered disk reads: 306 MB in 3.01 seconds = 101.72 MB/sec |
13 |
> Timing buffered disk reads: 266 MB in 3.00 seconds = 88.59 MB/sec |
14 |
> Timing buffered disk reads: 306 MB in 3.00 seconds = 101.84 MB/sec |
15 |
> Timing buffered disk reads: 306 MB in 3.00 seconds = 101.86 MB/sec |
16 |
> |
17 |
> That's all the same disk, repeating hdparm as fast as I could. The |
18 |
> disk was not even mounted at the time, and no other disks were active. |
19 |
> |
20 |
> Two very different but reproducible numbers, changing values at random |
21 |
> times. The only thing I can think of is that the disk may be doing |
22 |
> a SMART self-test, but for some reason the USB connection prevents |
23 |
> me from accessing the data so I can't test my theory. |
24 |
> |
25 |
> Any other ideas? |
26 |
|
27 |
bonnie++? |
28 |
|
29 |
My first guess is that something on the same USB bus might be |
30 |
periodically active, changing how the kernel manages talking to USB |
31 |
devices. Try ensuring no other USB devices are connected (or active), |
32 |
and running something like "hdparm -t /dev/sdc; hdparm -t /dev/sdc; |
33 |
hdparm -t /dev/sdc; hdparm -t /dev/sdc; hdparm -t /dev/sdc; hdparm -t |
34 |
/dev/sdc" ...letting the shell automate it for you. |
35 |
|
36 |
My second guess would be something relating to the kernel's page |
37 |
cache...but I'm unsure if a filesystem needs to be mounted first. |
38 |
|
39 |
My third guess would be some internal buffering behavior inside the |
40 |
USB disk, and hdparm's behavior happens to straddle an edge condition. |
41 |
|
42 |
|
43 |
|
44 |
|
45 |
-- |
46 |
:wq |