On 2/7/02 10:15 pm, "Chris McDonough" <chrism@zope.com> wrote:
I'm honestly not sure how this can happen. It looks impossible.
I do my best to make 'interesting' problems! :)
It might make sense to try to step through the code following line 279 of lib/python/OFS/Application.py ("if not hasattr(tf, 'session_data')") -- use pdb.set_trace() here when zope starts up) to see what is happening.
I'm not au fair with pdb stuff Chris, what I've done is this # Ensure that a temp folder exists LOG('Session', INFO, 'b4 temp folder') if not hasattr(app, 'temp_folder'): LOG('Session', INFO, 'no temp folder') ... # Ensure that there is a transient container in the temp folder tf = app.temp_folder LOG('Session', INFO, 'after temp folder') if not hasattr(tf, 'session_data'): LOG('Session', INFO, 'no session data') ... # Ensure that there is a transient container in the temp folder tf = app.temp_folder LOG('Session', INFO, 'after temp folder') if hasattr(tf, 'session_data'): LOG('Session', INFO, 'session data exists') if not hasattr(tf, 'session_data'): LOG('Session', INFO, 'no session data') And this is what's in the Zope.log 2002-07-02T21:53:19 INFO(0) Session b4 temp folder ------ 2002-07-02T21:53:20 INFO(0) ZODB Opening database for mounting: '10517752_1014201531.425206' ------ 2002-07-02T21:53:20 INFO(0) ZODB Mounted database '10517752_1014201531.425206' at /temp_folder ------ 2002-07-02T21:53:20 INFO(0) Session after temp folder ------ 2002-07-02T21:53:21 INFO(0) Session session data exists Seems to imply that temp_folder is there (it is), and my added lines imply that there is a session_data object in the temp_folder (there's not - drat!). I tried deleting the temp_folder as an admin user and that was no-go. Baffled... Tone -- Dr Tony McDonald, Assistant Director, FMCC, http://www.fmcc.org.uk/ The Medical School, Newcastle University Tel: +44 191 243 6140 A Zope list for UK HE/FE http://www.fmcc.org.uk/mailman/listinfo/zope