I am looking at my image to try and clean it up a bit. Having
clearedCaches I can still find 78 instances of WASession, 702 instances of EscapeContinuation, 722 instances of ResponseContinuation, 905 instances of WACurrentSession, 865 instances of WARenderCanvas. and so on... any ideas where to look to get the image to let go and let garbage? Keith ___________________________________________________________ Now you can scan emails quickly with a reading pane. Get the new Yahoo! Mail. http://uk.docs.yahoo.com/nowyoucan.html _______________________________________________ Seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
Since I've missed you saying you tried 'chase pointers' on some of
these instances, I'll mention it. It's available in the Inspector yellow button menu. HTH, Steve On 9/25/06, Keith Hodges <[hidden email]> wrote: > I am looking at my image to try and clean it up a bit. Having > clearedCaches I can still find > > 78 instances of WASession, > 702 instances of EscapeContinuation, > 722 instances of ResponseContinuation, > 905 instances of WACurrentSession, > 865 instances of WARenderCanvas. > > and so on... > > any ideas where to look to get the image to let go and let garbage? > > Keith > > > ___________________________________________________________ > Now you can scan emails quickly with a reading pane. Get the new Yahoo! Mail. http://uk.docs.yahoo.com/nowyoucan.html > _______________________________________________ > Seaside mailing list > [hidden email] > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -- How wonderful it is that nobody need wait a single moment before starting to improve the world. -- Anne Frank Paradise is exactly where you are right now...only much, much better. -- Laurie Anderson _______________________________________________ Seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
Steven Elkins wrote:
> Since I've missed you saying you tried 'chase pointers' on some of > these instances, I'll mention it. It's available in the Inspector > yellow button menu. > > HTH, > Steve > > Indeed I had tried this many times. I discovered that Magma Proxies were confusing the PointerFinder, having attempted to fix this by making the PointerFinder ignore MagmaPRoxies I continued my search and ended up in WeakRegistry! One of my components is registered for finalisation, with its executor being a minimal copy of itself with an inst var pointing to the resource that needs to be tidied up on finalisation, which doesnt seem to be happening. I heard rumours that Squeak finalization was not all that good, is this likely to be the problem and is there a way to give it a kick? any ideas? ___________________________________________________________ Try the all-new Yahoo! Mail. "The New Version is radically easier to use" The Wall Street Journal http://uk.docs.yahoo.com/nowyoucan.html _______________________________________________ Seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
Keith Hodges wrote:
> I heard rumours that Squeak finalization was not all that good, is this > likely to be the problem and is there a way to give it a kick? > > any ideas? Did you try (note - I don't run Magma): http://lists.squeakfoundation.org/pipermail/magma/2006-September/000398.html I run the regular WARegistry>>clearHandlers (which does work to free up some sessions). I usually let my image grow by about 10-20MB, then I run the stuff mentioned in the message referenced above. I'm not sure I'd run that code in a production environment though. I still need to track down why the regular clearHandlers doesn't free up the sessions. _______________________________________________ Seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
Solved my memory problem, I had some stuff in the #finalizationRegistry,
whose executors may not have been as well crafted as they needed to be thus stopping the finalization process. Having fixed these, these instances still didn't go away. The, somewhat curious, missing piece of the puzzle was to find a way to kick off finalization. I created a TestGC class with a method #finalize that would write to the transcript in order to see whether finalization was indeed working at all. TestGC new finalizationRegistry add: TestGC new. Having run this a few times, I took a look at the finalizationRegistry and to my surprise everything had gone! My previous days looking at this problem were confused and frustrated due to MagmaMutatingProxies not being ignored by PointerFinder. many thanks to everyone for your help. Keith ___________________________________________________________ All new Yahoo! Mail "The new Interface is stunning in its simplicity and ease of use." - PC Magazine http://uk.docs.yahoo.com/nowyoucan.html _______________________________________________ Seaside mailing list [hidden email] http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside |
Free forum by Nabble | Edit this page |