The_Game_Master

Bump? I would really like to know how to fix this crashes.

I was refering to the previsious crash, that you fixed in 10319. For this 2 crashes, here is the reproduction method. How to reproduce: 1. Create a level 80 NE Druid. 2. Create a level 80 NE DK. 3. Cast .learn all_myclass (with the DK) 4. Cast .max skill (with the DK) 5. /duel and accept between the 2 characters 6. add any weapon on the DK and cast Plague Strike (do not use .cast, cast it as if you were playing) 7. Blood Plague and Desecration will proc 8. Shapeshift into Cat Form (Druid) when desecration has 1 sec before expire 9. Make sure blood plague will expire AFTER desecration. Crashes 100% on x64. With x32 i tryed for hours and can't understand why it didn't crash. Preutty wird.

Thank you for the previsious crash fix. I managed to reproduce it, it was happening when 2 diffrent casters, casted the same buff on a 3rd targed. OS: Slackware 13 x64 Mangos Version: 10319 Custom Patches: * No custom patches, all removed for more accurate data. SD2 Version: 1758 Database Name and Version : YTDB 558 GDB Backtraces: * http://pastebin.com/hb01Hrj4 * http://pastebin.com/kMcHS2QQ * http://pastebin.com/Fc4kR9cE * http://pastebin.com/weSQdyjF There are 2 diffrent reasons for crashing. This is the second reason, more related then the 1st one. * http://pastebin.com/G22PLc3a * http://pastebin.com/jzjMEepP

Bump! Come on guys, i have this crashes for 3 weeks now. Even some workaround would help as a temp solution.

Mangos Version: 10313 Custom Patches: * BG Score 80+ crash * SD2 Achievement support * Extra instance functions * SD2 gameobject summon support * Gameobject loot rules * Thaddius encounter spells * Freya encounter spells * A few ICC spells * Spell disabled system None of the used patches may interfere with this crash dump or can be a lead cause. SD2 Version: 1756 Database Name and Version : YTDB 558 GDB Backtraces: * http://pastebin.com/v5GkzhCm * http://pastebin.com/HC87nnNx * http://pastebin.com/MAzxj2DH (not related to SpellAuraHolder) Revision: 10316 * http://pastebin.com/2S6qCQRb

Than wait for a reply from NoFantasy, he should know what's wrong with it.

I think it's related to this commit http://github.com/mangos/mangos/commit/6fa98d1226393aa84781fd0a9750e413bfca3643 and if i'm not mistaken you must update your database in order to work properly due to the recent script changes.

LordJZ, I think FeRkEl described the problem best of them all, I myself have tryed to reproduce the AuraHolder crash for over 3 days on a localhost with no luck. I don't know if you used a specific method, bot, etc. to be able to reproduce it but I can tell you one think. You had a ton of luck to find it. Good job! And thanks to Vladimir for fixing both problems. Toinan67, I don't think donating 2-3 euro gives you the right to be rude with diffrent people around the community. I myself donated anonymously a couple of times, and you don't see me beeing an asshole about it. Nor do i want or need a flashy rank to brag myself on how mutch i support this community. If you really want to be appreciated stop beeing an asshole and write some usefull patches. I can see you're not sutch an idiot as you look and you handle c++ preutty well. So try making yourself usefull for a change and keep the braging for your own forum.

Tip - you can add Wiki in the forum's header instead of the FAQ.

Nice catch, the method worked. Just cast Acid Spit http://www.wowhead.com/pet=42/worm#abilities when EoTB has 30 seconds left so it will expire a few ms later. Object.h line 115

What is the best solution, from a developer point of view to increase the current 3.3.5 branch stability? What feedback and help, exept from crash dumps (i know you got plenty of those) would it be required?

I was sure you were gonna suggest to use the browser's zoom. But that should be considered as an alternative.

Awesome! One sugestion, do you think you can increase the font size for codeboxes? It looks a little too small for 1600 resolutions. Allso, the linux section could include a way to get packages for other distros. A reference to eatch package website should be good enough.

Still no progress with it?

I just love when you all come with your air of superiority and have no idea how to solve the problem. No, i don't have a clue either. But atleast i try. tehmarto, thanks for the extra debug info. I found only 4 places where holders where deleted http://pastebin.com/VKR9UPSw did i missed something?

BUMP FOR THE LOVE OF GOD!!!!!111one I tryed on a x64 and x32 system, just in case was system dependent. No, it's not. Crash after crash after crash no matter the system. http://pastebin.com/6JESbHnh Just in case it's needed.

Becouse that one was made with custom patches. This one is on a clean core. Specially built to test and fix this crash problem. Thank you zergtmn.

Rev: 10272 Custom patches: gameobject summon, gameobject loot, dalaran sewer arenas, SD2. None of those 3 can influence this backtrace. http://pastebin.com/6JESbHnh I can post about 200 of these if you need more info! Help!

Rev: 10271 Custom patches: quite a few I got about 3 crashes every few minutes. Any ideeas? http://pastebin.com/5MSvbxqT http://pastebin.com/pkyvPk9A http://pastebin.com/Y1hSE7dt http://pastebin.com/ypbRJKsT

Try this patch, it's old, but should do the job. diff --git a/src/game/InstanceSaveMgr.cpp b/src/game/InstanceSaveMgr.cpp index 289ea07..b8ab0b1 100644 --- a/src/game/InstanceSaveMgr.cpp +++ b/src/game/InstanceSaveMgr.cpp @@ -505,6 +505,8 @@ void InstanceSaveManager::LoadResetTimes() if(t - tim[type-1] > now) break; + ScheduleReset(true, t - tim[type-1], InstResetEvent(type, mapid, difficulty, -1)); + for(ResetTimeMapDiffInstances::const_iterator in_itr = mapDiffResetInstances.lower_bound(map_diff_pair); in_itr != mapDiffResetInstances.upper_bound(map_diff_pair); ++in_itr) { @@ -610,6 +612,7 @@ void InstanceSaveManager::_ResetOrWarnAll(uint32 mapid, Difficulty difficulty, b return; time_t now = time(NULL); + time_t today = (now / DAY) * DAY; if (!warn) { @@ -638,10 +641,14 @@ void InstanceSaveManager::_ResetOrWarnAll(uint32 mapid, Difficulty difficulty, b // calculate the next reset time uint32 diff = sWorld.getConfig(CONFIG_UINT32_INSTANCE_RESET_TIME_HOUR) * HOUR; - uint32 period = mapDiff->resetTime * DAY; - time_t next_reset = ((now + timeLeft + MINUTE) / DAY * DAY) + period + diff; + uint32 period = (mapDiff->resetTime / DAY * sWorld.getConfig(CONFIG_FLOAT_RATE_INSTANCE_RESET_TIME)) * DAY; + time_t next_reset = today + period + diff; + // update it in the DB CharacterDatabase.PExecute("UPDATE instance_reset SET resettime = '"UI64FMTD"' WHERE mapid = '%d' AND difficulty = '%d'", (uint64)next_reset, mapid, difficulty); + + SetResetTimeFor(mapid,difficulty,(uint64)next_reset); + ScheduleReset(true, next_reset-3600, InstResetEvent(1, mapid, difficulty, -1)); } // note: this isn't fast but it's meant to be executed very rarely

Bump! Still needs a rewrite.

Bump! Still not working in 10252.

diff --git a/src/game/SpellAuras.cpp b/src/game/SpellAuras.cpp index e7447c9..9b1830c 100644 --- a/src/game/SpellAuras.cpp +++ b/src/game/SpellAuras.cpp @@ -8524,6 +8524,28 @@ void SpellAuraHolder::HandleSpellSpecificBoosts(bool apply) } case SPELLFAMILY_PALADIN: { + if (m_spellProto->Id == 19746) // Aura Mastery (on Concentration Aura remove and apply) + { + Unit *caster = GetCaster(); + if (!caster) + return; + + if (apply && caster->HasAura(31821)) + caster->CastSpell(caster, 64364, true, NULL); + else if (!apply) + caster->RemoveAurasDueToSpell(64364); + } + if (m_spellProto->Id == 31821) // Aura Mastery (on Aura Mastery original buff remove) + { + Unit *caster = GetCaster(); + if (!caster) + return; + + if (apply && caster->HasAura(19746)) + caster->CastSpell(caster, 64364, true, NULL); + else if (!apply) + caster->RemoveAurasDueToSpell(64364); + } if (m_spellProto->Id == 31884) // Avenging Wrath { if(!apply) Bump!

I used clean core WITHOUT SD2 and it did crashed. After i added SD2 and recompiled both core and SD2 it started working allright again. I don't understand what parts of SD2 remained after it was deleted the first time, I even deleted the release folder and used Clean Solution.

But i deleted SD2 folder and recompiled the core. The problem still persisted. Did i miss something?