Bit of an update... been poking around the database since my last post but am
still not able to find what is causing this issue.
I'd be curious to know where gs stores the group information as I cannot find
it in the db. I can see the group_id's across a few tables but not the actual
data for a group. I was hoping to find this and compare to a known working
group to see if different etc.
I've tried superimposing the new v16.04 release over the top of the upgraded
dir and re-run the buildout (in case something was missing). It completes but
still same errors.
Seems the issue relates to mostly (but not all) messages in the group which
have an attachment such as an image or document. I've found the actual
attachment files down under ~/var/instance and these are all present and
correct for all the broken entries. Posting a new message with attachment is
also broken so it's not just pre-upgrade entries that are affected.
If I init a fresh db, run a new buildout on that and then import a pg dump
there are no groups and no topics on the front page. Is it possible to find out
where this information is stored please as I would like to see how it is linked
and possibly re-link it all?
Running the instance in fg mode I get this when it breaks (Issue: KeyError:
'_all_columns')
2016-10-31 16:27:29 INFO Zope Ready to handle requests
2016-10-31 16:27:55 ERROR ZODB.Connection Couldn't load state for 0x07cd
Traceback (most recent call last):
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/Connection.py",
line 860, in setstate
self._setstate(obj)
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/Connection.py",
line 914, in _setstate
self._reader.setGhostState(obj, p)
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/serialize.py",
line 612, in setGhostState
state = self.getState(pickle)
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/serialize.py",
line 605, in getState
return unpickler.load()
File
"/app/groupserver-14.11/eggs/SQLAlchemy-0.9.10-py2.7-linux-x86_64.egg/sqlalchemy/sql/base.py",
line 584, in __setstate__
self.__dict__['_all_columns'] = state['_all_columns']
KeyError: '_all_columns'
2016-10-31 16:27:55 ERROR ZODB.Connection Couldn't load state for 0x0e09
Traceback (most recent call last):
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/Connection.py",
line 860, in setstate
self._setstate(obj)
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/Connection.py",
line 914, in _setstate
self._reader.setGhostState(obj, p)
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/serialize.py",
line 612, in setGhostState
state = self.getState(pickle)
File
"/app/groupserver-14.11/eggs/ZODB3-3.10.5-py2.7-linux-x86_64.egg/ZODB/serialize.py",
line 605, in getState
return unpickler.load()
File
"/app/groupserver-14.11/eggs/SQLAlchemy-0.9.10-py2.7-linux-x86_64.egg/sqlalchemy/sql/base.py",
line 584, in __setstate__
self.__dict__['_all_columns'] = state['_all_columns']
KeyError: '_all_columns'
My knowledge of python is non-existent so really not able to debug the above
error. Is there someone around with ZODB knowledge that can help please? Or
should I direct the errors above to the SQLAlchemy folk?
Many thanks
Peter