[rescue] playing with the new E250 ... disk performance vs. old PeeCee

[Magnus]

Joshua Boyd wrote:

> I don't recall what you said your drives were, but they are probably U2W
> (80MB/s bus speed) or U160 (160MB/s bus speed).  However, you also
> should look at the actual drive speed.  For instance, if you have two
> U320 drives that only do 50MB/s each, then you loose nothing by putting
> them on a U160 controller, or putting them on seperate channels of a U2W 
> controller. 

I am going to shut down the box and pull a drive when I get home.

Here is how Linux describes it in /proc/scsi/scsi:

Host: scsi0 Channel: 00 Id: 00 Lun: 00
   Vendor: IBM-ESXS Model: DTN073C3UCDY10FN Rev: S27P
   Type:   Direct-Access                    ANSI SCSI revision: 03

Here is the relevant section of dmesg:

[   15.162202] SCSI subsystem initialized
[   15.180035] sym0: <875> rev 0x14 at pci 0001:00:03.0 irq 5,7e0
[   15.266758] sym0: No NVRAM, ID 7, Fast-20, SE, parity checking
[   15.273964] sym0: SCSI BUS has been reset.
[   15.273994] scsi0 : sym-2.2.2
[    1.093291]   Vendor: IBM-ESXS  Model: DTN073C3UCDY10FN  Rev: S27P
[    1.093350]   Type:   Direct-Access                      ANSI SCSI 
revision: 03
[    1.093400]  target0:0:0: tagged command queuing enabled, command 
queue depth 16.
[    1.093460]  target0:0:0: Beginning Domain Validation
[    1.093665]  target0:0:0: asynchronous.
[    1.094890]  target0:0:0: FAST-20 SCSI 20.0 MB/s ST (50 ns, offset 16)
[    1.095635]  target0:0:0: Domain Validation skipping write tests
[    1.095654]  target0:0:0: Ending Domain Validation
[    2.556808]   Vendor: TOSHIBA   Model: DVD-ROM SD-M1401  Rev: 1007
[    2.556861]   Type:   CD-ROM                             ANSI SCSI 
revision: 02
[    2.556909]  target0:0:6: Beginning Domain Validation
[    2.557250]  target0:0:6: asynchronous.
[    2.567982]  target0:0:6: FAST-20 SCSI 20.0 MB/s ST (50 ns, offset 16)
[    2.575804]  target0:0:6: Domain Validation skipping write tests
[    2.575821]  target0:0:6: Ending Domain Validation
[    4.911977] sym1: <875> rev 0x14 at pci 0001:00:03.1 irq 5,7e6
[    4.999636] sym1: No NVRAM, ID 7, Fast-20, SE, parity checking
[    5.006940] sym1: SCSI BUS has been reset.
[    5.006974] scsi1 : sym-2.2.2
[   12.480822] Driver 'sd' needs updating - please use bus_type methods
[   12.482278] SCSI device sda: 143374000 512-byte hdwr sectors (73407 MB)
[   12.483398] SCSI device sda: drive cache: write through
[   12.484265] SCSI device sda: 143374000 512-byte hdwr sectors (73407 MB)
[   12.485410] SCSI device sda: drive cache: write through
[   12.485428]  sda: sda1 sda2 sda3
[   12.492904] sd 0:0:0:0: Attached scsi disk sda
[   13.018272] device-mapper: 4.4.0-ioctl (2005-01-12) initialised: 
dm-devel at redhat.com
[   13.327550] kjournald starting.  Commit interval 5 seconds
[   13.327603] EXT3-fs: mounted filesystem with ordered data mode.
[    2.192084] sr0: scsi3-mmc drive: 40x/40x cd/rw xa/form2 cdda tray
[    2.192120] Uniform CD-ROM driver Revision: 3.20
[    2.192307] sr 0:0:6:0: Attached scsi CD-ROM sr0
[    2.243674] sd 0:0:0:0: Attached scsi generic sg0 type 0
[    2.243767] sr 0:0:6:0: Attached scsi generic sg1 type 5


The really key thing I am seeing there is that Linux is seeing a 
20MB/sec SCSI device.  So if the controller (and certainly the disk) are 
good for 40MB/sec as you suggest, then there may be software issues.

The chipset is being detected as sym53c875.  From what I can tell from 
some googling, that chipset should indeed be capable of going 40MB/sec. 
  Something here is amiss.