root@aber-cs.UUCP (vax) (01/13/87)
WE are having problems using 'newfs' on some eagles that we have just had delivered attached to an sc41 Emulex controller. Any advice that can be offered will be gratefully received. We have source and run our ra81's with slightly different partioning to the standard uda.c driver layout. In particular we use the d,e,f partitions to split up the f partition rather than the useless definitions normally available. A Simple question is ' how many sectors should there be on our eagles' ?? I beleive the hardware is set at 48 sectors per track. UUCP : { ENGLAND or WALES }!ukc!aber-cs!dap JANET: dap@uk.ac.aber.cs PHONE: +44 970 3111 x 3267 Post: University College of Wales, Penglais, Aberystwyth, UK, SY23 3BZ.
chris@mimsy.UUCP (Chris Torek) (01/18/87)
In article <11@aber-cs.UUCP> root@aber-cs.UUCP (vax) writes: >WE are having problems using 'newfs' on some eagles >that we have just had delivered attached to an sc41 Emulex >controller. ... A Simple question is ' how many sectors should >there be on our eagles'?? I beleive the hardware is set >at 48 sectors per track. The SC41/MS is an MSCP-compatible controller. This means that it will reserve one or two sectors per track for bad block replacement. If the hardware is set to 48 sectors per track, the controller will report 47 sectors per track in its On Line end message, where it encodes the `geometry' of the drive. The Get Unit Status end message also includes the total usable size of the drive, in sectors. This can be found in `ra_dsize' or `radsize' or whatever name it has in your particular driver (see /sys/vaxuba/uda.c). The 4.2BSD UDA50 (MSCP) driver has several bugs that will make it act erratically on 780s, and may tie up a Unibus for long periods on 750s (I forget). The 4.3BSD driver is much improved, but still does not tell you how big your disk is. I have a replacement that is *still* not ready for distribution. Watch this space for future announcements. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) UUCP: seismo!mimsy!chris ARPA/CSNet: chris@mimsy.umd.edu