OS : MS Win XP-SP2 Thunderbird : Trunk nightly - version 3.0a1pre (2007100303) Virtual memory size of Thunderbird increases 40MB after each "Shift+Delete of all 40,000 mails and Compact folder". I don't know whether the increased virtual memory will be re-used by other activities than "Shift+Delete" or never be re-used. This phenomenon is possibly involved in problem of Bug 369255(Junk move is being suspected) and phenomenon in some bugs which claim that large VM size was used when delete of many mails. However, as far as I remember, I didn't observe such large increase when Delete(not Shift+Delete) of many mails(near 30,000 mails) in the past. So, this bug's phenomenon may be "Shift+Delete" unique. I think this is core issue, but open as Product=Thunderbird initially. Change to appropriate Product/Component, please.

CC-ing to David & Mscott. What kind data is required to analyze this kind of mail db related problem? If required data is one that can be obtained by procedure described in next page, http://www.mozilla.org/performance/leak-tutorial.html what parameter is to be specified when testing?

m-wada, is your issue imap, and improved via Bug 519979 - move/deletes of large number of imap messages slow without caching? if not, might it be a dup of Bug 538378 - Mass delete of messages resulted in massive memory (not sure when imap deletes moved to offline. was it Bug 435153? - Better Faster IMAP: Pseudo-offline Delete and Move support)

Test was done with local mail folder, not IMAP folder. As I wrote in bug 583365 comment #1. it looks leak in UI - not-removed reference to resource till window close.

wada, is this better for you now that bug 538378 is fixed?

might one of you be able to test this using a current version? (In reply to Wayne Mery (:wsmwk) from comment #4) > wada, is this better for you now that bug 538378 is fixed [in TB14]? and also bug Bug 723248, fixed in TB15

(In reply to Wayne Mery (:wsmwk) from comment #5) > might one of you be able to test this using a current version? STR of this bug is very simple and clear. Does you question mean that no one including you can't reproduce problem I saw with the simple&clear STR?

I'm not currently in a position to test

(In reply to Wayne Mery (:wsmwk) from comment #4) > wada, is this better for you now that bug 538378 is fixed? Because patch for that bug touches code for imap folder only, and because my case is local mail folder, I don't think that bug is relevant.