Frank Wennmohs said:
> On Sun, Mar 16, 2003 at 12:42:50PM +0000, Gavan Fantom wrote:
>> On Sat, 15 Mar 2003, Frank Wennmohs wrote:
>>
>> > Yes, I am also using quite a plain system:
>> > RPC700 board, internal IDE, SA110, 64MB
>>
>> Just to rule this out, whivh revision of the SA110?
>>
>
> The good one with the DEC Logo ;)
> Revision S

S or T is good 8)

>> > > Is the problem repeatable? What happens if you restart the
>> installation from scratch?
>> > >
>> >
>> > The whole installation was from scratch.
>> > I just did an installation of 1.6O (200302240000)
>> > and the problem does not occur.
>> >
>> > Any suggestions?
>>
>> It would seem that one of the more recent changes to -current has
>> eliminated this problem, then. As for what I'm not too sure as I
>> haven't been keeping up with -current recently, though.
>>
>
> I never had problems like this with -stable either.
> This morning I tried it agagin with different block- and filesizes. No
> way. The first hint that something is wrong is, after the
> entered root password, that the root directory is full.
> Well, this had happened before, so I didn't get nervous at this.
> But when I tried to mount them manually and got some messages
> of defect inodes etc. that got me nervous....
>
>> Are you in a position to try earlier versions of -current to give a
>> rough idea of when it started working?
>>
>
> Sure. The machine is not in production condition right now... ;)
>
> Any suggestion as for the date of -current?

Could you try 1.6, see if it was broken then, narrows it down to a change
that was pulled up...

>
> I tried to compile X with -current, but there seem to be
> some pthread problems. It compiled correctly, but
> when trying to startx it stops because of problems with pthread.
> But due to the fixed dynamic linker it's ways faster.

Hmmm, could you PR that one?  Just so we don't lose track of the PR's
(even if we let them rot, we at least track the rot 8)

Cheers,
Chris