JSG8A@ACADVM1.UOTTAWA.CA (Jean-Serge Gagnon) (05/20/91)
Thanks Valdis... I am very much aware that the solution will vary from one system to the next, especially in the environment that I currently work. The reason I was asking without giving any specifics, is that before having to go out and getting documents to do the thing, I was wondering if there wasn't any system on which it would be easy to do without having to get the doc. For example, because I havn't taken the time to look it up, is there any standard on an IDE, SCSI or an AT type controller card for PC's? Does a MAC have any signal that's standard on it? Etc... OR... do I really have to do it on the schematic level of a Hard Disk, and there's no way to do it from the controller level? P.S. Once again, is there a list I can subscribe to that would be a bit more suited to this type of question? Thanks again. _______________________________________________________________________ | | | Jean-Serge Gagnon <JSG8A@ACADVM1.UOTTAWA.CA> | |___________________________________ ___________________________________| | | | | Specialiste en Equipement | Computer Hardware | | Informatique | Maintenance Specialist | | Universite d'Ottawa | University of Ottawa | | (613) 564-7813 | (613) 564-7813 | |___________________________________|___________________________________|
lstowell@pyrnova.pyramid.com (Lon Stowell) (05/21/91)
Some older systems had hardware write protect to prevent accidental corruption of boot and load image files. These had address switches....the hard drive was write protected from 0,0,0 up through the address set in the switches. A separate switch enabled the protection...allowing updates when in the defeat position. You may need to prevent both write and erase current....the implementations I am familiar with actually performed the write protect in the disk controller...by inhibiting the Write Enable signal...which inhibited ALL write procedures for protected addresses. A failure in the disk write circuitry could destroy data, but the purpose of Write Protect was to protect the hardware from software...not hardware faults. If you were to implement it today, you could provide write inhibit for a range of addresses...using software to load these. You could then set an "expert" flag instruction in software to prevent ordinary mortals...such as root...from overwriting the data. '
lstowell@pyramid.UUCP (Lon Stowell) (05/21/91)
Some older systems had hardware write protect to prevent accidental corruption of boot and load image files. These had address switches....the hard drive was write protected from 0,0,0 up through the address set in the switches. A separate switch enabled the protection...allowing updates when in the defeat position. You may need to prevent both write and erase current....the implementations I am familiar with actually performed the write protect in the disk controller...by inhibiting the Write Enable signal...which inhibited ALL write procedures for protected addresses. A failure in the disk write circuitry could destroy data, but the purpose of Write Protect was to protect the hardware from software...not hardware faults. If you were to implement it today, you could provide write inhibit for a range of addresses...using software to load these. You could then set an "expert" flag instruction in software to prevent ordinary mortals...such as root...from overwriting the data. '
horne@sapphire.idbsu.edu (cs354- Broward Horne) (05/23/91)
In article <9105201340.AA10238@lilac.berkeley.edu> BITNIC IBM-NETS List <IBM-NETS%BITNIC.BITNET@lilac.berkeley.edu> writes: > >The reason I was asking without giving any specifics, is that before >having to go out and getting documents to do the thing, I was wondering >if there wasn't any system on which it would be easy to do without having >to get the doc. > >For example, because I havn't taken the time to look it up, is there any >standard on an IDE, SCSI or an AT type controller card for PC's? Does a >MAC have any signal that's standard on it? Etc... > >OR... do I really have to do it on the schematic level of a Hard Disk, >and there's no way to do it from the controller level? > >P.S. Once again, is there a list I can subscribe to that would be a bit >more suited to this type of question? A good place to try might be the Virus newsgroup ( comp.virus, I believe ). I couldn't help but notice your system looks like it's an Autocad lab or some such. I originally wanted to write-protect the hard disks for our Acad lab, but it finally proved to be unfeasible. For instance, how about temporary file saves? To do them to a temp directory, or to a user's floppy disk, makes a WORLD of difference to the users. Just takes too much time. There should be a READ/WRITE line to the hard disk, if you can get the schematic. Don't forget to tie the end still connected to the drive to either a high or low, depending on which means READ/WRITE. A lot of logic chips don't respect a hanging wire as being LOW. -- Broward Horne Distant lands are not so far away, horne@sapphire.idbsu.edu I don't know why *you* don't go. an030@cleveland.freenet.cwru Take my hands, I'll show you the way. Pack your bags, and.... sail away. :)
horne@sapphire.UUCP (cs354- Broward Horne) (05/23/91)
In article <9105201340.AA10238@lilac.berkeley.edu> BITNIC IBM-NETS List <IBM-NETS%BITNIC.BITNET@lilac.berkeley.edu> writes: > >The reason I was asking without giving any specifics, is that before >having to go out and getting documents to do the thing, I was wondering >if there wasn't any system on which it would be easy to do without having >to get the doc. > >For example, because I havn't taken the time to look it up, is there any >standard on an IDE, SCSI or an AT type controller card for PC's? Does a >MAC have any signal that's standard on it? Etc... > >OR... do I really have to do it on the schematic level of a Hard Disk, >and there's no way to do it from the controller level? > >P.S. Once again, is there a list I can subscribe to that would be a bit >more suited to this type of question? A good place to try might be the Virus newsgroup ( comp.virus, I believe ). I couldn't help but notice your system looks like it's an Autocad lab or some such. I originally wanted to write-protect the hard disks for our Acad lab, but it finally proved to be unfeasible. For instance, how about temporary file saves? To do them to a temp directory, or to a user's floppy disk, makes a WORLD of difference to the users. Just takes too much time. There should be a READ/WRITE line to the hard disk, if you can get the schematic. Don't forget to tie the end still connected to the drive to either a high or low, depending on which means READ/WRITE. A lot of logic chips don't respect a hanging wire as being LOW. -- Broward Horne Distant lands are not so far away, horne@sapphire.idbsu.edu I don't know why *you* don't go. an030@cleveland.freenet.cwru Take my hands, I'll show you the way. Pack your bags, and.... sail away. :)
CSSTEVEN@CPHKVX.BITNET (05/23/91)
I'm not advertising in any way but one of my friends working for Seagate
in Singapore told me that they've got password and write-protected facility in
the controller firmware.
So you may contact your local Seagate dealer to ask for suggestions.
Hope that helps.
Steven W.M.Wong,
RA, CS Dept,
City Polytechnic of HK.