Just browsing around and stopped on a page to read, app had plenty of free ram (over 1mb), encountered a crash sitting still, reloading an image (i think). Calling chain using A6/R1 links Back chain ISA Caller 00000000 PPC 14D11FFC 15D37100 PPC 14CFC938 main+00130 15D370A0 PPC 14CFBE4C main1(int, char**, nsISupports*)+0083C 15D36DF0 PPC 16D40E48 nsAppShellService::Run()+00018 15D36DB0 PPC 14862F44 nsAppShell::Run()+00048 15D36D60 PPC 14863680 nsMacMessagePump::DoMessagePump()+0003C 15D36D10 PPC 14863C70 nsMacMessagePump::DispatchEvent(int, EventRecord*)+0014C 15D36CC0 PPC 1487B860 Repeater::DoIdlers(const EventRecord&)+00030 15D36C80 PPC 1487BCA0 nsTimerPeriodical::IdleAction(const EventRecord&)+00014 15D36C40 PPC 1487C048 nsTimerPeriodical::FireNextReadyTimer()+0003C 15D36C00 PPC 1487BE64 nsTimerPeriodical::FireAndReprimeTimer(nsTimerImpl*)+00094 15D36BB0 PPC 1487CB5C nsTimerImpl::Fire()+00024 15D36B70 PPC 14801480 timer_callback(nsITimer*, void*)+00020 15D36B30 PPC 161F5094 gif_delay_time_callback(void*)+000AC 15D36AF0 PPC 161F4F88 process_buffered_gif_input_data(gif_struct*)+00098 15D36AA0 PPC 16CE457C ImgDCallbk::ImgDCBHaveImageAll()+00018 15D36A60 PPC 16CE6C04 il_image_complete(il_container_struct*)+002D8 15D36A10 PPC 14808164 ImageNetContextImpl::GetURL(ilIURL*, ImgCachePolicy, ilINetReade r*, int)+006C4 15D36810 PPC 161B26C8 nsURILoader::OpenURI(nsIChannel*, int, const char*, nsISupports* )+00018 15D367D0 PPC 161B2F14 nsURILoader::OpenURIVia(nsIChannel*, int, const char*, nsISuppor ts*, unsigned int)+00260 15D366F0 PPC 161B09EC nsDocumentOpenInfo::Open(nsIChannel*, int, const char*, nsISuppo rts*)+0007C 15D366A0 PPC 149D038C nsHTTPChannel::AsyncRead(nsIStreamListener*, nsISupports*)+000C8 15D36650 PPC 149D3814 nsHTTPChannel::Open(int)+0008C 15D36490 PPC 149D2164 nsHTTPChannel::CheckCache()+002D0 15D36210 PPC 14996678 nsCacheManager::GetCachedNetData(const char*, const char*, unsig ned int, unsigned int, nsICachedNetData**)+000F4 15D36140 PPC 14998EDC nsReplacementPolicy::GetCachedNetData(const char*, unsigned int, nsINetDataCache*, nsCachedNetData**)+00084 15D360C0 PPC 14998AEC nsReplacementPolicy::CheckForTooManyCacheEntries()+000F0 15D36060 PPC 14999094 nsReplacementPolicy::DeleteAtleastOneEntry(nsINetDataCache*, uns igned int, unsigned int*)+000E8 Closing log

keyword luvin', cc'ing pam and gagan.

Fix summary. Shame, I liked "crache in cache".

And again. This is not the image cache, but the necko cache.

This looks like a dup of 52818.

pink/et al: could you guys confirm if this still happens since the checkins to the possible dup are in now?

what is an easy way to confirm this? what cases would this happen? just browse porn for 3 straight days? well, if you insist....;)

When the number of entries in the cache hit 512, we start deleting entries. If your cache was corrupted at this stage, we were crashing.

Marking "needinfo". This seems like a bad thing, but we need a patch, review and super review to clear for checkin.

adding space between 'need' and 'info'

*** This bug has been marked as a duplicate of 52818 ***

Verified dupe.