I can only answer partially... I can change the id to allow to have both the hd connected only in 10/12 hours, so no fx until then.
The script I used is betxen's , and the prom settings are correct ( the old hd boots once reconnected ) SystemPartition=dksc(0,1,8) OSLoadPartition=dksc(0,1,0) OSLoader=sash OSLoadFilename=/unix -----Messaggio originale----- Da: cctalk [mailto:cctalk-boun...@classiccmp.org] Per conto di Pete Turnbull via cctalk Inviato: lunedì 29 maggio 2017 01:38 A: cctalk@classiccmp.org Oggetto: Re: Help cloning an SGI boot drive On 28/05/2017 23:45, Mazzini Alessandro via cctalk wrote: > I tried to clone a boot drive ( on a bigger one ), using one of the > scripts on nekochan. The disk was initialized , labelled then > formatted by the script, and cloned. The data is present if I mount it > > The issue starts when I change the id and try to boot from it... What is the script? The best online resource I can suggest is Ian Mapleson's page on cloning, at http://www.sgidepot.co.uk/disksfiles.html#CLONE (use the tar method if the disk is EFS rather than XFS). I do this quite often without any issue, mostly on Indys, though I use tar slightly differently (to exclude /proc and mounts). If you boot with the original working disk, with the clone connected, what does fx(1M) show for the clone? [start fx (you need to be root), select the controller, drive,and LUN 0, then do label, show, all] >> From Prom, ls gives : > > dksc(0,1,8)/: > sgilabel sash symmon > esf read error, bad count > no file > system found for "/". and what does the output of nvram(1M) show, especially for SystemPartition and OSLoadPartition? -- Pete Pete Turnbull
