[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

RE: On very different journalling activity on 2 servers.



Hi,

On Tue, 2003-02-04 at 08:30, Nico Morrison wrote:

> Thank you, but unfortunately:
> 
> [root ns5 nico]# /sbin/hdparm -i /dev/hda
>  DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5 udma6 

That's not actually sufficient --- "hdparm -i" shows the ID block on the
disk which the kernel queried when the drive was first detected,
including the originally negotiated DMA mode, but it doesn't show
whether or not you are *currently* using DMA.  IO errors may have forced
the driver to back off to PIO, but that won't show up in "hdparm -i".

Do a simple "hdparm /dev/foo" and look for the "using_dma" line; that
will tell you if the driver is currently using DMA or PIO for transfers.

Cheers, 
 Stephen





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]