Print Page | Close Window

Phantom Siege Patch

Printed From: Illyriad
Category: News & Announcements
Forum Name: Announcements
Forum Description: Changes, patch release dates, server launch dates, downtime notifications etc.
URL: http://forum.illyriad.co.uk/forum_posts.asp?TID=5354
Printed Date: 18 Aug 2019 at 14:21
Software Version: Web Wiz Forums 11.10 - http://www.webwizforums.com


Topic: Phantom Siege Patch
Posted By: GM Stormcrow
Subject: Phantom Siege Patch
Date Posted: 13 Dec 2013 at 23:42
Hi all,

As many of you know, there have been a few incidences of phantom sieges.

All sieges should be from one of the 8 squares directly adjacent to a target city, and a phantom siege is defined as a siege that sets up and bombards a city from a non-adjacent square to a target city.

We know that this situation occurs when the target city has a siege inbound, and moves via Exodus - but we're really not sure precisely how it happens; as the code would appear to forbid this situation from occurring - and we're not sure why it doesn't work this way in all circumstances.

However, as a temporary patch, we've put some code in place that checks - every few minutes - to see whether a city is under siege where the [X|Y] co-ordinate difference between the sieging army and the target town is more than one square in any direction.  If it finds any sieging armies that meet this criteria, it repatriates them, with a message to the sieging army.

Please note in this phantom siege situation, a siege may still arrive and start to prepare for bombardment, but it should now be repatriated the moment the city is no longer located adjacent to the siege.  However, as it takes 12hrs to set up a siege before bombardment begins, we hope your city should not take any inbound bombardments from non-adjacent sieges.

Please do, as ever, open a petition if this issue reoccurs.  And if it does, as much detail as you can supply (timings of when the siege arrived, when you started Exodus, etc) would be most, most welcome; so we can hopefully make a more permanent fix rather than this automated "repatriation patch".

On another note, some people were experiencing an issue with the SSL certificate for the website in the last few days.  This affected people only using https (SSL) to view only the generic website (www.illyriad.co.uk) - rather than the game server (elgea.illyriad.co.uk) which had the secure certificate on it. 

Whilst we're not sure why anyone would want to view the generic illy website securely (!)... now you can if you wish, as we've installed the new wildcard *.illyriad.co.uk SSL certificate onto the www servers, not just the elgea gameservers.  

I suspect you can still pull up a certificate error if you try reading this forum (forum.illyriad.co.uk) using https... but I have yet to read anything on the forum (including my own posts!) that would require secure communications Wink

Best wishes,

SC



Replies:
Posted By: Anjire
Date Posted: 13 Dec 2013 at 23:50
Thank you!

But sorry to add to further headache with Exodus/Siege related bugs but petition #9508 exposes issues with peace of camp and armies camping out as both friendly reinforcements and enemy encampments inside a city they are at war with.


-------------
http://elgea.illyriad.co.uk/a/p/26125" rel="nofollow">


Posted By: Anjire
Date Posted: 14 Dec 2013 at 01:05
The_Dude asked me to post that in addition to phantom sieges - he had phantom blockades set up against his city San Jancinto.  




-------------
http://elgea.illyriad.co.uk/a/p/26125" rel="nofollow">


Posted By: Daefis
Date Posted: 14 Dec 2013 at 01:12
Thanks GM Stormcrow:)............... Whilst I'm not involved in the war issues like this make or break a game.

-------------
http://elgea.illyriad.co.uk/a/p/37796" rel="nofollow">


Posted By: Brandmeister
Date Posted: 14 Dec 2013 at 03:09
Stormcrow, it feels like some kind of multi thread issue, or maybe a variable caching issue. I encountered a similar problem once when I ported some code and things unexpectedly started getting passed call by value instead of call by reference. Variables would get caught as a snapshot, and then the main process would jump the rails and the other one would asplode (that's the technical term, for you non-techs).


Posted By: Rill
Date Posted: 14 Dec 2013 at 03:25
It's a dev ...

They DO exist!

Santa!


Posted By: Tatharion
Date Posted: 14 Dec 2013 at 04:09

It’s Beginning to Feel a Lot Like Christmas !!!   

http://www.youtube.com/watch?v=n543eKIdbUI



-------------
Be without fear in the face of your enemies. Be brave and upright, that God may love thee. Speak the truth always, even if it leads to your death. Safeguard the helpless and do no wrong.


Posted By: GM Stormcrow
Date Posted: 14 Dec 2013 at 04:55
Originally posted by Anjire Anjire wrote:

Thank you!

But sorry to add to further headache with Exodus/Siege related bugs but petition #9508 exposes issues with peace of camp and armies camping out as both friendly reinforcements and enemy encampments inside a city they are at war with.

Thanks, am looking into this too.

Originally posted by Anjire Anjire wrote:

The_Dude asked me to post that in addition to phantom sieges - he had phantom blockades set up against his city San Jancinto.  

And that's something I wasn't expecting either, but is definitely part of the same subroutine.

For both the above issues I'm going to patch (hopefully make sure it can't happen again) then repatriate armies in the "unlawful" positions.  This will happen Saturday 14 Dec.

Originally posted by Brandmeister Brandmeister wrote:

Stormcrow, it feels like some kind of multi thread issue, or maybe a variable caching issue. I encountered a similar problem once when I ported some code and things unexpectedly started getting passed call by value instead of call by reference. Variables would get caught as a snapshot, and then the main process would jump the rails and the other one would asplode (that's the technical term, for you non-techs).

Thanks Brand, and I completely understand what you're saying.  However, the rather uniquely peculiar architecture & programming language choice that run the game engine of illy preclude this as a possible answer to the issue.  

It's much more an issue of the complexity of the illy sandbox beast: the more depth we put in the game, the more unforeseen dependencies get unintentionally created between supposedly discrete functional areas of gameplay, and the harder these are to track down.

I know everyone says "Didn't you guys test this?"... but trust me... there is nothing like building a sandbox MMO for finding every single "1 in a billion" margin case that we:
  • didn't 
  • couldn't
  • wouldn't
  • shouldn't
  • would never... in a gazillion years... 
... have anticipated arising ingame, ever.

And this is partially a weaknesses - but also (imo) our greatest strengths, as it shows we have players committed enough to find the margins, and a loose enough ruleset that they can push the margins without entirely breaking the game. Hurrah for you and us!

Anyhoo.  Another patch tmw that should hopefully close down a few more of these cases.

Cheers,

SC


Posted By: Brandmeister
Date Posted: 21 Dec 2013 at 21:25
I'm pretty sure that the bug just kicked into action here:

http://elgea.illyriad.co.uk/#/World/Map/-373/-495

The siege is a few squares away to the NW.



Print Page | Close Window

Forum Software by Web Wiz Forums® version 11.10 - http://www.webwizforums.com
Copyright ©2001-2017 Web Wiz Ltd. - https://www.webwiz.net