[HN Gopher] Corrupted Blood incident
___________________________________________________________________
Corrupted Blood incident
Author : soheilpro
Score : 78 points
Date : 2023-08-24 17:50 UTC (5 hours ago)
(HTM) web link (en.wikipedia.org)
(TXT) w3m dump (en.wikipedia.org)
| elaus wrote:
| I think it's fascinating that this issue persisted for over a
| week with the developers seemingly unable to fix it - or not
| wanting to use "really hacky code" (quote from the article). Even
| with distributed systems, I would (naively) think you could just
| add a bit of code to the server that resets the database field
| containing the "infection"?
|
| In the end they seemingly saw no other way than to roll back the
| whole game to the state before the "outbreak". This must have
| been an immense annoyance for players, to lose one week of
| progress.
| aftbit wrote:
| Yeah I was also surprised that it took them this long to deal
| with it. Why not change the effect so that it was not
| contagious, or so that it did not do any damage? Or perhaps
| lean into the healers and modify a common healing ability to
| cleanse the effect and grant a buff that prevents reinfection
| for a period of time? Or make it so that a character could only
| be infected with the effect once they had entered the boss
| area?
|
| I clearly do not understand how these debuffs function in the
| codebase. I wonder why they felt the need to do a full reset
| instead of just nerfing the debuff.
| manquer wrote:
| It is more likely they didn't want to deal with player
| complaints and noise on lost progress or items because of the
| player deaths .
|
| It was probably easier to reset from a backup than any real
| difficulty to apply a fix .
|
| Same reasons we implement STONITH or just discard a game to
| restart from a save or use file from backup if it deviated a
| lot from where we want it to be, rather than fix the
| corruption so to speak even when possible .
| danjoredd wrote:
| I agree. Couldn't they have just disabled the boss and the
| effect? It seems that just disabling those two things, making
| it so that you can't exit the boss arena with corrupted blood,
| then enabling them again would be a simple solution
| input_sh wrote:
| I was there for this. Patches were fairly rare, as every patch
| required a rather long and annoying downtime. The player base
| was also split between the official server and a bunch of
| private ones (nobody I knew played on the official ones), so
| they couldn't really do anything server-side (I mean they
| _could_ , but that wouldn't affect all of the players).
|
| There wasn't much of a progress during that week, you needed to
| be really high up in levels to be able to survive for even a
| minute after infection. IIRC max level was 70 at the time, and
| you needed to be in the upper half of the upper half to even be
| able to react to the infection.
|
| In-game deaths was more of an annoyance than a deal-breaker.
| You'd spawn back as a ghost at the nearest safe town and then
| have to walk back to your corpse to revive, so basically for
| the majority of the players that week I'd just die, walk back
| to the corpse, respawn, and die again far away after
| interacting with an NPC, as my line of thinking was that the
| further I am from bigger cities, the odds of an NPC being
| infected decreased.
| bcoates wrote:
| This article is borderline incoherent and written as WoW roleplay
| for some reason.
|
| (this is from memory plus the less-incorrect looking sources I
| can find; I'm probably wrong about something but less wrong than
| the Wikipedia)
|
| The corrupted blood effect is a relatively weak, short lived
| negative effect ("debuff") that appeared on one of the players in
| a group trying to fight Hakkar. This effect did a modest amount
| damage over time, lasted a few seconds, and would (and this is
| very rare for the time in the game) spread to nearby players,
| each player getting their own countdown starting from the time of
| infection.
|
| While the debuff cured itself automatically after a few seconds,
| it didn't cause immunity, so if two or three players stood next
| to each other, they would re-infect each other over and over
| again.
|
| The strategy to defeat Hakkar was to do this _intentionally_ ,
| preserving the short-lived disease until the team could use it
| against him -- there was no "panic" and nobody was confused about
| how the debuff worked -- understanding it was necessary to
| attempt the fight.
|
| It was not possible to spread the disease by fast traveling.
|
| What actually happened was an "exploit" (or clever strategy or
| cheating, whatever you want to call it): A player with a
| summonable pet could get their pet debuffed, then put away the
| pet -- freezing time for the pet preserving the contagious
| debuff. That player could then bring out the pet anytime in the
| fight they needed to have the debuff to use against Hakkar. Much
| easier than intentionally juggling the effect.
|
| Nobody put away their pet in a panic to protect it not knowing
| how the debuff works, that's not a thing for several reasons.
|
| Someone performing this exploit, by accident or experimentation,
| found out that the debuff was not restricted to portion of the
| game world where Hakkar is, nor restricted to the team members
| fighting him.
|
| They then _intentionally_ released debuffed pets in crowds, and
| more significantly, released the information about how to
| preserve and transmit the short-lived effect.
|
| This is important: the only way the corrupted blood effect
| spreads (between in-game cities, for example) is by _intentional
| action_ , its a griefing weapon, not a emergent property like a
| disease. Once it's out in public and you don't need to qualify
| for a high level raid group to see it, however, it's easy to
| transport and deploy.
|
| The Wikipedia article makes it sound like "Azeroth" is a single
| virtual space that things can spread within; this is not the
| case, there are a hundred-odd (named) servers each with their own
| mostly identical world (only the players are different). Nothing
| can spread between these servers but information; they are
| totally independent.
|
| The way the disease spread and persisted over time was _memetic_
| : while the condition would self-persist in a few virtual cities
| on a few servers on its own [1], Blizzard could fairly easily fix
| this, what was less simple to deal with was the masses of players
| intentionally preserving or re-acquiring from scratch the debuff
| and gifting it back to any crowd of players they could come
| across, on every server in parallel.
|
| This was unusual because the game has few to no ways for players
| of the same side to grief/harm each other, so lots of players
| jumped on the opportunity to be the bad guys. Also, the non-
| participating players could not do anything to deter or stop the
| players intentionally repeatedly introducing the debuff -- this
| was fixed in the later, by-design, zombie plague which made the
| infected (temporarily) a third side hostile to and attackable by
| all other factions, even in otherwise enforced no-PvP zones.
|
| There was no singular "Hard Reset" (though Blizzard may have
| experimented with unscheduled server reboots). A server reboot is
| disruptive to players mid-dungeon or mid-fight so they are
| usually scheduled in advance.
|
| At no time was a player's, or the world's state or progress
| rolled back to before anything. Rolling world server reboots were
| used to apply patch 1.8 that altered the encounter to prevent the
| spread but that was the normal process for patching at the time.
|
| [1] It may have been possible for two NPCs standing next to each
| other to bounce the debuff back and forth. Maybe it was just
| transit terminals where the game was constantly dropping in
| unsuspecting players, I don't remember. The primary vector was
| dense clumps of players and intentionally stored pets.
| trafficante wrote:
| Was present for the Hakkar plague and can confirm the accuracy
| of this post. I was starting to doubt my own memory a bit
| because it's been so many years since the event and every
| article that pops up about it makes claims about stuff like
| "forced rollbacks" that I absolutely had no recollection about
| ever happening.
| shkkmo wrote:
| The lucky thing about Wikipedia is that you can help improve
| articles.
|
| Your comment is much more informative than the linked article,
| it would be great to see it updated with your information if
| you can find references for it.
| bcoates wrote:
| Feel free, I'm not so optimistic about the mechanisms of
| Wikipedia.
|
| Almost all the wrong information in the article has a high-
| quality source (presumably repeating either the Wikipedia
| article itself or some other random journo playing telephone
| about an event that happened in a then-relevant video game
| almost 20 years ago.) Or maybe the author pulled the ol'
| claim not found in source. I didn't check!
|
| Serious answer: This is a job for Internet Historian
| https://www.youtube.com/c/InternetHistorian who is... (double
| ironically? anti-hyperbolically? I'm not sure how to express
| surprise he actually lives up to the name) someone who does
| some quality research about dumb old internet stuff everyone
| misremembers and the news media got wrong the first time.
| ChrisArchitect wrote:
| (2005)
| whartung wrote:
| I've learned long ago to never second guess Blizzard on the
| complexity of a "fix" for anything that goes awry in this game.
| There are innumerable "how hard can it be" bugs that show up
| that, evidently, are hard to fix. The ripple affects seem to
| carry to all sorts of unintended consequences. And there have
| been uncountable bugs where you go "How the heck did they
| manifest that after this patch?"
|
| I remember the Blood Plague, I was there when it happened. I
| distinctly recall someone in front of the Ironforge Auction House
| calling out "everyone come to me, it's better when we get close".
| Which, of course, it's not. It's a bad idea. But it was his idea
| of a good time. The solution was to abandon the cities. But there
| were always folks who enjoyed spreading the plague. And,
| honestly, that never bothered me. It IS fun. It was a fun
| encounter, and certainly fun in terms of the whole dynamic going
| on, appreciating what a novel "World Event" it became.
|
| World Events is WoW have almost always led to disaster, so we
| don't get them anymore. The game simply doesn't handle huge
| volumes of players in a single area well. Their last big try was
| the Gate Opening that turned the Silithus zone into a slide show
| for most of the day. But as a player, it was still a great, epic
| experience, because you know it will never happen again. In a
| game of continual routine, outliers are always fun.
|
| The Blood Plague was a novel event, it made the World more
| "alive". I'm glad I was there to see it and be part of it. Just
| one of those things the makes WoW history even richer.
| blueridge wrote:
| Well said, this resonates. I was also there for the plague and
| for the gate opening on Mal'Ganis. I also remember a few events
| where someone had kited or teleported world bosses back to
| Orgrimmar and Stormwind. Absolute mayhem.
|
| Lots of fun, lots of memories, lots of people I met during
| those years who are still in my life today.
| dang wrote:
| Related:
|
| _Corrupted Blood Incident_ -
| https://news.ycombinator.com/item?id=9463545 - April 2015 (14
| comments)
| chunkyks wrote:
| FaxMonkey, a well known WoW player and video maker, made a video
| showing it, at the time:
| https://www.youtube.com/watch?v=JLZ_q4N55T0
| tedunangst wrote:
| > In 2008, Blizzard released an intentional plague in World of
| Warcraft to echo the success of the Corrupted Blood pandemic.
| [...] Blizzard put a stop to the Lich King plague on October 28
| after receiving complaints from players that the zombie infection
| was detracting from other parts of the game.
|
| Takes a few tries for the lesson to stick.
| whartung wrote:
| But there's the rub. Some of the population hates these kinds
| of events, and another portion love it.
|
| Is there "Griefing potential"? Yes, there is. But, man, the
| game is so mundane and routine the rest of the time, why not
| mix it up a little?
|
| They brought this back at the end of Battle For Azeroth, during
| the end of expansion event. Players could become ghouls and
| infect other players. You'd see them lingering and lying in
| wait when bosses would spawn in Ice Crown, then shenanigans
| ensued.
|
| They also did this back in Mists of Panderia, and not just for
| a short time, but it was end game content where you could use
| an object and become a hostile PVP player to EVERYONE, not just
| the other faction, even on PVE servers. Did the griefers have a
| field day with that? Yea, for a short time. More reason for the
| "citizenry" to posse up and beat the ne'er do wells back.
|
| But there were always those that complain. "This happened and
| inconvenienced me".
|
| I'm sorry. BFH. TDB. Boo hoo. Too bad. That game is playable
| that way the other 100 weeks of the expansion. There's lots of
| players, lots of playstyles, every decision Blizzard makes is a
| bad one in someones eyes. So, lighten up and let others have a
| turn.
| trafficante wrote:
| > They also did this back in Mists of Panderia, and not just
| for a short time, but it was end game content where you could
| use an object and become a hostile PVP player to EVERYONE,
| not just the other faction, even on PVE servers.
|
| This event was a callback to one of my favorite griefing
| exploits of vanilla WoW.
|
| There used to be a Horde-only quest called "Test of Honor" in
| one of the mid-level Kalimidor zones. Most players would just
| finish the quest and carry on with levelling - but, in the
| rare instance where the player FAILS the quest, they get hit
| with a several hour long debuff called "Mark of Shame" that
| flagged them as hostile-on-attack to all factions.
|
| This meant that players with the debuff could still safely
| move around inside Horde capital cities (though they couldn't
| interact with NPCs) but if any "friendly" player was unwise
| enough to try attacking them, they'd immediately be swarmed
| with lvl 60 elite Horde guard NPCs.
|
| By the time Blizzard removed the debuff, the griefing meta
| was to level a Tauren (largest hitboxes), get the debuff, and
| then either camp on top of the Org mailbox or clip through
| into the back of the Auction House and sit on top of the
| auctioneers. Any player attempting to check their mail or
| auctions would muscle-memory right click, which caused an
| auto-attack on the debuffed griefer, which shortly resulted
| in a trip to the graveyard as the elite lvl 60 NPCs would
| spawn to rock your world.
|
| We had gimmick guilds like <On Strike> where the characters
| were named AuctioneerX to prevent savvy players from using
| the /tar Auctioneer command to bypass the hitbox.
|
| Ah, those were the days _siiip_
| jowday wrote:
| It was a deeply divisive event - for a very large fraction of
| the playerbase, it was the most fun they've had playing WoW.
| The entire way the game functioned changed for a week - some
| people set out to grief, others worked to try to cure people
| infected. And it was done just before the release of a new
| expansion - any gear or progress that occurred in that week
| would be invalidated two weeks later when the new expansion
| raised the level cap by 10 - all of the best equipment you
| could earn prior to the new expansion would be replaced by
| normal leveling gear. The game was, so to speak, in a lame duck
| state at the time, so why not have some fun with it?
| thatguy0900 wrote:
| If you like this kind of stuff there's also the falador massacre,
| a runescape pvp bug that let people pvp in safe areas
|
| https://darkrunescape.fandom.com/wiki/Falador_Massacre
| mewse-hn wrote:
| Looking forward to this happening on the hardcore realms set to
| open today
| tsumnia wrote:
| They disabled the debuff outside ZG for Classic, as well as
| included leashing so people can't kite the world bosses to
| major cities. I'm still curious to see what the hardcore scene
| will turn into.
|
| The original Corrupted Blood incident is fun to look back on.
| Obviously it was a pain to die, but once you did, it became a
| game within itself of locating "safe" spots to rez and get out
| of the city asap. This was still during the MC/BWL days, so I
| remember being lucky enough to have my hearthstone set to
| Kargath. There were griefers there as well, but much fewer than
| Orgrimmar.
| mewse-hn wrote:
| I was kidding although a malignant debuff killing hundreds of
| hardcore characters would be really, really funny
| fsckboy wrote:
| [flagged]
| Mordisquitos wrote:
| A bit of a COBLUD-05 denialist are we?
| outsidetheparty wrote:
| I thought it was a great example of unintended effects in a
| large complex system... and the CDC asking Blizzard "if they
| could use data from what they perceived as a planned disease
| simulation to inform their disease modeling research" was an
| interesting twist.
| loa_in_ wrote:
| I find it immensely hard to believe that going through character
| saves and removing the disease or disabling the effect altogether
| was not an option. I do have a feeling that there was an
| incentive to not do that and milk the attention the outbreak
| generated and to profit from the collective equipment damage,
| leading to more gameplay and in-game currency purchase.
| kej wrote:
| There was no first-party currency purchase at the time, and
| they weren't hurting for attention back then, either, so I'm
| inclined to go with Hanlon's razor on this one.
| duskwuff wrote:
| > ... and to profit from the collective equipment damage,
| leading to more gameplay and in-game currency purchase.
|
| I don't think Blizzard profited from either of those. As far as
| I'm aware, their only paid WoW product at the time was monthly
| subscriptions to the game -- there was no officially supported
| way for players to buy in-game currency, and players paid
| Blizzard the same amount regardless of how much time they spent
| in game.
| shadowgovt wrote:
| World of Warcraft was one of the pioneer MMORPGs. My knowledge
| of how things go here is thin on the ground, but anecdotally
| from discussions of others who worked on other contemporary
| MMORPGS: it's easy for us in this era of distributed-everything
| to overestimate how easy doing anything in these systems was.
| I'm aware of one circumstance where a known buffer-overflow bug
| was delayed for weeks in fixing in another game because devs
| simply released changes to the codebase in exhaustively-tested
| integrated batches; in the meantime, they relied on "game
| masters" to monitor the bug and fix it by hand when it
| manifested.
| veave wrote:
| This was 2005, there was no attention to sell and they would
| gain nothing from equipment damage since there was no in-game
| shop.
___________________________________________________________________
(page generated 2023-08-24 23:01 UTC)