[ZODB-Dev] ZODB packing error

[Tim Peters]

[Marius Gedminas, on 0-day packing]
> ...
> IOW there is no corruption.  I must have misunderstood the discussions
> about 0-day packing I skimmed on zope and zodb-dev.

Oh, I'm not sure about that.  I'm not a ZODB expert (although I expect to
become one).  Things can certainly get complicated.  For example,

- Transaction T1 loads an object O with a reference to persistent object
  P, and modifies some other attribute of O (it does not load P itself).

- Other transactions remove the reference from O to P, and all other
  references to P, from the database.

- A pack to current time is done.  Now the only reference to P lives in
  memory, inside T1's view of O.

- T1 finally tries to commit.

At this point, I see nothing to prevent O from having a "dangling reference"
to P -- T1 itself never loaded P into memory, and the database no longer
contains any trace of P, but O still references it.  That probably isn't
good.

>> We did fix one major bug of this kind "recently":  if you packed to
>> a time earlier than a time previously packed to, it was possible to
>> lose data. That's fixed in the current releases of ZODB.

> Does this bug exist in 2.6.2 (the latest Zope packaged for Debian)?

The bug in question was fixed in ZODB 3.1.5, released on 12-Feb-2004.  This
corresponds to the Zope 2.6.4 release.  A brief account of ZODB 3.1 bugs and
release dates can be found here:

    http://zope.org/Products/ZODB3.1/NEWS