Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!caip!meccts!dicome!salmi
From: salmi@dicome.UUCP (john salmi)
Newsgroups: net.unix,net.unix-wizards
Subject: /etc/newfs failure
Message-ID: <958@dicome.UUCP>
Date: Wed, 9-Jul-86 11:16:06 EDT
Article-I.D.: dicome.958
Posted: Wed Jul  9 11:16:06 1986
Date-Received: Fri, 11-Jul-86 06:29:37 EDT
Organization: the /usr/local/maniacs society, suburbia, minnesota
Lines: 64
Keywords: ultrix, /etc/chpt, bummer
Xref: watmath net.unix:8560 net.unix-wizards:18773

hi again, net.  another problem.  the scenario...

vax 11/785 running ultrix 1.1 (please, no flames about running 1.2).
fuji eagle (2351a) hanging off of an emulex sc7000.  

the eagle worked magnificently on a vax 750 running 4.2.  after the
transition from the 750 to the 785, the eagle got sick.  here is a
``script'' session from a /etc/newfs ...


> Script started on Wed Jul  9 10:06:01 1986
> # 
> # /etc/chpt -q /dev/rhp0c
> /dev/rhp0c
> Current partition table:
> partition     bottom        top       size    overlap
>     a              0      15883      15884    c,f,g,h
>     b          16058      49497      33440    c,f,g
>     c              0     808319     808320    a,b,d,e,f,g,h
>     d          49910      65793      15884    c,f,g
>     e          65968     121903      55936    c,f,g
>     f              0     399999     400000    a,b,c,d,e,g,h
>     g              0     808319     808320    a,b,c,d,e,f,h
>     h              0          0          0    a,c,f,g
> 
> # /etc/newsfs -b 8192 -f 1024 /dev/rhp0c eagle
> 
> /dev/rhp0c:	808320 sectors in 842 cylinders of 20 tracks, 48 sectors
> 	413.9Mb in 53 cyl groups (16 c/g, 7.86Mb/g, 2048 i/g)
> 
> super-block backups (for fsck -b#) at:
>  32, 15440, 30848, 46256, 61664, 77072, 92480, 107888, 123296, 138704,
>  154112, 169520, 184928, 200336, 215744, 231152, 246560, 261968, 277376,
> 292784, 308192, 323600, 339008, 354416,write error 184928
> write error: 369840
> wtfs: I/O error
> # 
> script done on Wed Jul  9 10:06:50 1986
> 

this was an indication of a somehow corrupted disk.  so, i brought the vax
down and ran emulex's formatting diagnostics, which found 23 bad spots on
the disk.  now, he says, everything is ok.  hahahahahahahahahaha!  ;-)

the drive is still sick.  i have rerun the formatter (like 10 hours worth!)
and still i get a puking /etc/newfs.  as you can see from the script session
above, i have been changing partition sizes to try and "sidestep" the "bad 
spot" on the disk.  no luck.  

the controller is set up to emulate an eagle.  i have recieved mail from
many sources telling me to call the eagle an extended rm03.  why?

please mail responses, and as always, thanks in advance!


----

        -john salmi   
        -business - ...!ihnp4!dicome!root
        -pleasure - ...!ihnp4!dicome!salmi
        -software & systems administrator
        -dicomed corporation
        -minneapolis, mn
        -612/885-3092