hdparm output

cli_man · 07-11-2003, 11:05 AM

I am very confused by the output of testing my drive on one of my linux servers, when I run 'hdparm -Tt /dev/hda it will give me 3.25 MB/sec 2 or 3 times in a row then when I come back and try it I get 25.10 MB/sec, it will go back and forth - there is no middle ground on it, it is eather really low numbers or higher.

This server is not being used for anything so there is no proccess's taking up the time, it is not even plugged into the network.

Does anyone know why it would be doing this? DMA is enabled on the drive.

fancypiper · 07-11-2003, 11:18 AM

What are all your settings?

hdparm /dev/hda

Code:

root@uilleann # hdparm /dev/hda        

/dev/hda:
 multcount    = 16 (on)
 I/O support  =  0 (default 16-bit)
 unmaskirq    =  0 (off)
 using_dma    =  1 (on)
 keepsettings =  0 (off)
 nowerr       =  0 (off)
 readonly     =  0 (off)
 readahead    =  8 (on)
 geometry     = 4865/255/63, sectors = 78165360, start = 0
 busstate     =  1 (on)
root@uilleann # hdparm -Tt /dev/hd{a,a,a,a}

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.91 seconds =140.66 MB/sec
 Timing buffered disk reads:  64 MB in  1.37 seconds = 46.72 MB/sec

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.92 seconds =139.13 MB/sec
 Timing buffered disk reads:  64 MB in  1.36 seconds = 47.06 MB/sec

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.99 seconds =129.29 MB/sec
 Timing buffered disk reads:  64 MB in  1.92 seconds = 33.33 MB/sec

/dev/hda:
 Timing buffer-cache reads:   128 MB in  1.01 seconds =126.73 MB/sec
 Timing buffered disk reads:  64 MB in  2.08 seconds = 30.77 MB/sec

cli_man · 07-11-2003, 11:42 AM

[root@cache root]# hdparm -Tt /dev/hd{a,a,a,a}

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.75 seconds =170.67 MB/sec
Timing buffered disk reads: 64 MB in 2.58 seconds = 24.81 MB/sec

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.84 seconds =152.38 MB/sec
Timing buffered disk reads: 64 MB in 2.59 seconds = 24.71 MB/sec

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.76 seconds =168.42 MB/sec
Timing buffered disk reads: 64 MB in 2.58 seconds = 24.81 MB/sec

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.86 seconds =148.84 MB/sec
Timing buffered disk reads: 64 MB in 2.58 seconds = 24.81 MB/sec
[root@cache root]# hdparm /dev/hda

/dev/hda:
multcount = 16 (on)
I/O support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 5005/255/63, sectors = 80418240, start = 0
[root@cache root]#

fancypiper · 07-11-2003, 11:53 AM

The disk reads are a little low. My drive is a 7200 rpm with the 8mb cache.

Here is my 5200 rpm 2 mb cache drive

Code:

 root@uilleann ~ # hdparm -Tt /dev/hd{c,c,c,c}

/dev/hdc:
 Timing buffer-cache reads:   128 MB in  1.01 seconds =126.73 MB/sec
 Timing buffered disk reads:  64 MB in  2.30 seconds = 27.83 MB/sec

/dev/hdc:
 Timing buffer-cache reads:   128 MB in  1.17 seconds =109.40 MB/sec
 Timing buffered disk reads:  64 MB in  2.28 seconds = 28.07 MB/sec

/dev/hdc:
 Timing buffer-cache reads:   128 MB in  0.96 seconds =133.33 MB/sec
 Timing buffered disk reads:  64 MB in  2.29 seconds = 27.95 MB/sec

/dev/hdc:
 Timing buffer-cache reads:   128 MB in  0.97 seconds =131.96 MB/sec
 Timing buffered disk reads:  64 MB in  2.28 seconds = 28.07 MB/sec

cli_man · 07-11-2003, 11:58 AM

Let me put a bit of history on this:

This machine is sopposed to be a squid cache server, we had it in production for about a week and it worked great, it was using about 6% proccessor, however every once in awhile I would see it jump to 100% use for about 25 to 45 seconds and then squid would start giving errors. As much as I could tell the hard drive was getting overrun with info, when I would run the hdparm -Tt /dev/hda it would come back with a throughput of only 3 to 6 MB/Sec then 15 minutes later it would report 25 MB/Sec

What I am wondering is if the MB is bad and switches off the DMA sometimes then a few seconds later resets the drive or something and the DMA starts again, I really don't know but it is starting to bug me :-)

cli_man · 07-11-2003, 12:01 PM

Oh and some server hardware stats:

667 MHz PIII Slot 1
384 Megs of pc-133 ram
40 Gig Western Digital 7200 rpm IDE Drive
3Com Network card

Thats pretty much the Hardware info the software that is running is:

Redhat 7.2 with updates on it
Squid
Webmin
SSH

There is not load on it now as I have pulled it and replaced it with a 1.6 GHz P4 but I now need this machine for a smaller site where its speed would work just fine if I could get the hard drive to be more stable with its throughput.

fancypiper · 07-11-2003, 12:25 PM

What's the condition of your /var partition? Would it be getting full? Have the old logs been checked and deleted when no longer needed?

# Show disk usage by current directory and all subdirectories
du | less
# How much free drive space
df -h

Check /var/log/messages and other appropriate logs for the times things are funky and see if anything shows up there.

cli_man · 07-11-2003, 12:32 PM

Actually I am only using about 3 gigs out of the 40 Gig Drive, I tried looking at the logs to see but they are huge from the times that it was running as a cache server, I will look again and see if I can find anything in them. It has been running really steady today but I did see it drop performance once today. I will check the logs to see if I can see anything from that.