On 3 November 2014 23:39, Alan Altmark <[email protected]> wrote:
> > I see installations that run ICKDSF CPVOLUME FORMAT (CPFMTXA) or ICKDSF > INIT against he whole volume, then turn around and give it to Linux to run > dasdfmt. (I think there's still some sort of residual institutional > memory of the days when you had to do ICKDSF INSTALL to unpark heads and > calibrate the drive.) > More than 10 years ago, a bug in dasdfmt would make it sometimes miss a track while formatting a volume. As a bypass, we used ICKDSF (or CMS FORMAT) to format the volume, and had dasdfmt just do the first few tracks. This got into fora and redbooks, and people continued the habit long after the bug was corrected. But some people forgot ICKDSF and just did the dasdfmt for the first tracks and ended up with I/O errors (just like people sometimes only format the first cylinder of a page pack). So our friends removed the option to have a partial dasdfmt, and people started to do full ICKDSF plus full dasdfmt and invented their own rituals around that. If you label the volume with ICKDSF and give Linux cylinders 1-END, then it will find a blank volume with no volser. Because the Linux installer had trouble with that, we populated that pseudo minidisk with something that looks like a volume already formatted by dasdfmt. Attractive for those who could flashcopy a ready-to-consume volume over it, because it skips the entire format process. When there's no volser on the (pseudo) minidisk, then Linux can't read one. If that's causing problems again to install Linux, then we seem to have an old bug back again. Rob ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For more information on Linux on System z, visit http://wiki.linuxvm.org/
