[Info-vax] %MAIL-E-UAFOPNERR, error opening user authorization file

[Phillip Helbig undress to reply]

In article <nqn4d4$vrs$2 at dont-email.me>, Simon Clubley
<clubley at remove_me.eisner.decus.org-Earth.UFP> writes: 

> On 2016-09-04, Phillip Helbig (undress to reply) <helbig at asclothestro.multivax.de> wrote:
> > I saw this recently in the log of a batch job.  It occurred during a 
> > SET FORWARD command.  I've never seen it before.
> >
> > I was logged in at the time.  Nothing strange was going on.
> >
> > What can cause this?
> 
> Any I/O errors on the disk ?

Just the usual few.

> Is the disk either shadowed or part of a RAID set ?

Part of a shadow set.

> And yes, I _know_ that read errors are supposed to be handled
> transparently in the latter case :-), but given the age of some of
> the hardware Phillip is running...

Old but gold.  :-)

I do plan to move to x86 and, presumably, more modern disks if and when 
a hobbyist license (as promised) for x86 becomes available.  Also 
promised was a mixed-architecture cluster with 8.4 Alpha, so the 
migration can proceed in steps.  (I ran a mixed VAX/Alpha cluster for 14 
years).

My guess is that there is no SSD and/or flash supported on Alpha (except 
perhaps the old SBB 268-MB SSDs---so no MODERN SSD or Flash).  Or could 
I move to more modern disks sooner?  (Cost, of course, is an issue on a 
hobbyist system.  I now have enough spare SBB disks, in the sizes I 
need, to last several years.)

How does the MTBF of modern SSD and flash disks compare to that of 
traditional disks?  I recall hearing---several years ago---of SSDs which 
were shot after half an hour where conventional disks lasted months or 
years (admittedly on a very high I/O system).  Can one rewrite them 
often enough now?

> Any mount verification events ?

No.