[micro.ibm] Partial recovery of bad tracks with Spinrite

mvolo@ecsvax.UUCP (Michael R. Volow) (02/20/89)

Have just acquired Spinrite, and would be interested in your interpretation
and recommendations.  My current Seagate ST225 (swapped under-warranty for
one with 0-track seek problems) boots well and works well despite its age
(probably 1986).  As a swapped drive, I suspect it had had some damage
because it arrived with 112,640 byte of bad sectors (11 tracks). But what's
112k out of 21 meg, so I kept it because it worked well.


Hard error map:         HD    CYL         I suspect that the adjacency
                        ---------         of the bad tracks/cylinders
                        0     276         suggests that the drive may have
                        0     277         previous radial damage (from
                        0     341         a head crash before refur-
                        0     342         bishing?)
                        0     343
                        1     278
                        1     279
                        1     280
                        2     214
                        2     237
                        3     345

I then backed up and did a low level reformat with Spinrite, using the
deepest level of testing to evaluate the surface (took about 5 hours).  In
addition I changed the interleave from 5:1 (throughtput 104,448) to 3:1
(throughput 174,080).  Following is Spinrite's summary after reformatting.

***************************************** TRACK MAP********************
*  Low Level Format  **    0 ........................................ *
*  C:  3:1  Depth:4  **  160 ........................................ *
***********************  320 ........................................ *
*                     *  480 ........................................ *
***********************  640 ........................................ *
*    Track Status:   **  800 ..............1......................B.. *
*--------------------**  960 ...................................B1B11 *
*  |  Reading/Wrting ** 1120 B....................................... *
*  *  Reformatting   ** 1280 ....................BBB1.B.............. *
* .oO Patt Testing   ** 1440 ........................................ *
*  <  Relocating     ** 1600 ........................................ *
*  .  Reformat Okay  ** 1760 ........................................ *
* cC  Correctable    ** 1920 ........................................ *
* uU  UnCorrectable  ** 2080 ........................................ *
* 123 Defect Count   ** 2240 ........................................ *
*  B  Marked As Bad  ** 2400 ........... 2442                         *
***********************************************************************

Before reformatting, all tracks marked "1", were marked "B".  Spinrite told
me that of the previously 112,640 bad bytes, 102,000 bytes (50 clusters or
200 sectors) had been returned to active use (!) and that no additional
defective sectors had been found. This is obviously due to its ability to
recover partial tracks with damage to only 1 or two sectors. I was both
pleased at the increased storage and ambivalent about it's safety.

What do you think? Should I return the recovered sectors back to
inactivity? Or live with danger for a while? The drive stays fairly full,
running only 1.8-2.5 MEG free space.

Any suggestions would be appreciated.

Thanks in advance.

M Volow, VA Medical Center, Durham, NC 27705           mvolo@ecsvax.UUCP