With lots of blood, sweat and tears. Well not that bad but it is a pain. Firstly rebooting the server regularly is the hacky horrible solution. Secondly the long term fix is more complex, in our case we were storing part of a sql result in a request, that caused a cyclical reference and the leakage. Go through any python you may have and double check any such references... Cheers. -- Andy McKay.
----- Original Message ----- From: "Ron Bickers" <rbickers-dated-995300033.6db3d5@logicetc.com> To: "Andy McKay" <andym@activestate.com> Sent: Monday, July 09, 2001 9:13 AM Subject: RE: [Zope] Refcount question
How did you resolve the problem? _______________________
Ron Bickers Logic Etc, Inc.
-----Original Message----- From: Andy McKay [mailto:andym@ActiveState.com] Sent: Monday, July 09, 2001 12:08 PM To: Ron Bickers; Zope List Subject: Re: [Zope] Refcount question
No, but we had a similar problem for a while. It would appear some objects are being leaked.
Cheers. -- Andy McKay.
----- Original Message ----- From: "Ron Bickers" <rbickers-dated-995298075.870779@logicetc.com> To: "Zope List" <zope@zope.org> Sent: Monday, July 09, 2001 8:41 AM Subject: [Zope] Refcount question
These two refcounts just keep going up. The server has been running Zope 2.3.3 for 18 days without any problems, but unlike many of the other refcounts I've watched, these never go down. I have a 14MB, 33276 object Data.fs and the load is fairly light.
Acquisition.ImplicitAcquirerWrapper: 4337 Shared.DC.ZRDB.Results.SQLAlias: 2687
Jun 24 Jul 9 Delta Acquisition.ImplicitAcquirerWrapper 765 4320 +3555 Shared.DC.ZRDB.Results.SQLAlias 357 2687 +2330
Is this normal?