retry mounting with ro when rw fails

[Andriy Gapon]

on 07/04/2011 20:19 Matthew Jacob said the following:
> 
>> The first part is in CAM/SCSI to make sure that ENODEV is consistently returned to
>> signal that an operation is not supported by a device (in accordance to intro(2))
>> and specifically to return ENODEV on write attempt to a read-only or
>> write-protected media.  Making this change in SCSI should cover real SCSI devices,
>> as well as ATAPI through ahci/siis/atapicam or similar, plus majority (all?) of
>> USB Mass Storage devices.
> 
> This may be the right thing to do (which is debatable) but I can tell you here and

I am always open to a debate.

> now this will break a lot of commercial usages of FreeBSD with SANS which depend
> (foolishly) on existing error codes.

Do you mean EACCES -> ENODEV specifically?
If yes, can you tell me how those application distinguish EACCES that means ro/wp
media from EACCES that means there is something wrong with some permissions?
Or EACCES that could be returned from cam_periph_mapmem?

P.S. do you have specific applications in mind or do you speak hypothetically?

P.P.S. the only reason for my EACCES -> ENODEV is that *I* wanted to be sure what
error condition I am seeing.  If compatibility is preferred over what I perceive
as correctness, then be it, I would not insist.

Thanks!
-- 
Andriy Gapon