I am trying to install a Toshiba MK1214 HDD on my Mark I player and I am stuck on the initialization process. The hardware works OK but when I try to "build" the drive I get an error during the "Selecting Pump" section. This happens no matter which version of the "upgrade" file I use.

The drive works perfectly OK when I install it in a laptop. I have tried deleting the DOS partition from the drive but I get the same response.

Attached is the Hyperterminal log from boot when the drive is connected. You can see the drive is correctly identified but then there is an "unexpected interrupt".

Before I give up in the drive, can anyone think of what I am doing wrong?

Thanks,
Enrique Rodriguez
[email protected]

Greetings!

I find it slightly unusual that your are finding your single drive as hdb, not hda. Is this a dual-drive unit? Do you have the drive set up as the slave drive? Was this drive built with the drive builder to partition the hard drive?

The unit is a single drive unit and I removed the original drive before installing the new drive. When I re-install the original drive everything is OK.

I had not noticed the HDA vs. HDB difference on the log. What is the significance? The drive does not have any jumpers so I do not know how to set it as "slave" or "master".

Thanks for any additional insight into this.

Enrique

Hmmmm... no jumper typically means drive is set to "Master"...

In Linux the Master drive on the first IDE channel is designated as /dev/hda and the Slave on the first IDE channel as /dev/hdb. Master and Slave on the second IDE channel are /dev/hdc and /dev/hdd respectively.

On a MkI you have two IDE cables (or at least 2 IDE headers, the MkI, unless it was a dual drive unit from start, only came with one cable), each with just one contact - the drives should both be set as Master and recognized as /dev/hda and /dev/hdc)

(Note: the MkII and MkIIa, that only have one IDE channel available, cheats a bit in order to be compatible with the MkI, and actually calls its Slave (on the first and only IDE) /dev/hdc)

/Michael

The hardware works OK but when I try to "build" the drive I get an error during the "Selecting Pump" section.

Are you following the drive upgrade guide and...

a) Using the correct builder image, and

b) Answering the prompt and pressing the Enter key at the right moment in Hyperterminal?

Well, thanks much for the insight on the HDA vs. HDB recognition. After a little digging I found out this particular drive model has a known problem on the master/slave selection.

I enclose the specific note for those interested. I made the mod referred to in the article and the process is working now.

Thanks again,
Enrique

FROM http://kharendaen.dyndns.org:8000/~henke/public/mk1214gap/

The Toshiba MK1214GAP appears to act as an IDE slave device when CSEL (pin 28 on the IDE bus) is driven high, even when the drive is jumpered as master. It is generally not a problem, as most non-cable-select systems pull CSEL low or allow it to float. Some, such as the Toshiba Libretto 50CT, drive CSEL high enough to cause the drive to misconfigure as slave.

One remedy for this is to short CSEL (pin 28) to a ground, such as the adjacent pin 30. This pulls CSEL low and forces the drive to correctly configure as master.

Similar issues (which respond to the above fix) have been reported for other drives (such as the Toshiba MK6409MAV and the Fujitsu MHN2300AT) in systems from various vendors.

After a little digging I found out this particular drive model has a known problem on the master/slave selection.

This is very good information. I'm going to link this thread from the drive upgrade guide. Thanks you!