Subject: Re: [netatalk-admins] 1.4b2+OS8 patch & file assist sync/Trash Problem
From: Bob Smith, Hammett & Edison, Inc. (bsmith@h-e.com)
Date: Mon Jul 13 1998 - 17:07:57 EDT
On Mon, Jul 13, 1998, 03:19:08 Tan Chee Weei wrote:
>Most of the problems encountered have been attributed to the bug in afpd
>with regards to the Trash problem as described by bsmith@h-e.com in this list
>in March. Have been able to eliminate the disk -50 problem which it seems
>is a direct consequence of not being able to remove a file/folder to a shared
>Trash.
>
>File Assistant still doesn't behave correctly though in certain instances
when
>a file that was just synchronised is immediately examined again (although
>it should not require synchronising to begin with) This is repeated on every
>pass, without fail. It is only the case when synchronising the folder in
>which the file resides in. Direct synchronisation of the file only works
>correctly by just examining and reporting that it is synchronised. Do not
>know if the fact that the file begins with a numeral is a reason for its
>strange behaviour. It also seems as if file lists are ordered differently
>between those on the server and a local disk (not sure if that is how things
>work but it seems to suggest it if in fact a kind of list is utilised in
>determining files to sync).
>
>Is there a proper fix to netatalk to resolve the shared Trash Problem? It
>seems like a rather major problem that needs fixing.
I have followed the Trash issue closely since I developed the work-around, and
to my knowledge there is no fix available. This is a somewhat mysterious
problem, I don't know if anyone has even traced the cause yet, I tried but I
don't know enough about the AppleShare protocol to crack it. It might be
related to the lack of persistent directory IDs, but that's just a guess.
That also might be your problem with File Assistant; in my experience the DID
issue is at the root of most of the weird problems with afpd volumes.
Unfortunately there is no fix for that either, although that is definitely
being worked on, Adrian Sun has a plan for getting persistent DIDs into a
future version of his patches, and there are several others working on
different approaches, scan the list archives from a few months back for
details.
Sorry I can't be more helpful!
Bob Smith
Hammett & Edison, Inc.
bsmith@h-e.com
This archive was generated by hypermail 2b28 : Sat Dec 18 1999 - 16:32:55 EST