I am using the GT version Client. I am going to check with all the people that play, but again I have not had this or heard of it from anybody playing, and if there is one thing for sure Mir players are quick to point out a bug that doesn't help them.
A friggin anooying bug
- Thread starter Demonic
- Start date
Demonic said:
Because I wrote the code myself, I know when SQL is/can be accessed, and its not when moving about.
Mrklawz said:
This is the cause of being stuck on mine.
The server and client become de-synced, the Server thinks there at one spot the client says there at another.
So when the client goes to move, it sends a packets with the coords it’s moving too, along with the direction. The server then works out where the client will be moving too, based on the direction sent. If they don’t match, the client isn’t allowed to move, and they are then stuck.
However yours sounds like a different bug, because when I get stuck I can’t move anywhere. I assumed the bug was the same, as I experienced it on Euro Mir 3, exactly the same as mine, wasn’t able to move anywhere.
But I also believe the problem is with the client, the client is either not sending the packet when moving, or it gets lost/corrupted.
I don’t think it could be getting lost/corrupted, as the only time you can get stuck is when running. Walking is absolutely fine.
- Dec 1, 2003
- 1,806
- 2
- 165
Sorry to keep posting Crap (LOL) but semi update + answer for Buzznet...
Update
Put Server files on Dual 933 + Sql Entperise + Windows 2000 Server...
From my testing so far, not had the bug (maybe couple stick but nothing serious). Rungate was switched to Orig version (I.e. - not hexed at all). Server is 30-06 plus a few more hex put in to close a couple more back doors and put a delay on Wizzy magic (they deserve it eh)...
Client I have tired is Visio One (3G if you like, old interface)...
Need to leave it running for a few days until im sure the problem is gone/less of and then i can format / try it on Athlon with windows 200 etc...
Buzznet
Well funny that - cos not a single post on LOM about this and i posted on a few forums and as soon as i mentioned it - the complaints came flooding in. People might think it is lag, they try to move, cannot, keep trying and then bug goes away and they carry on without much thought. As i say - its def more prominent if they have a pet and its in line with you...
I did a search on Euromir3 forums, lot of posts about it on there + on chinese forums they are speaking a lot about it also - so its def a bug, if yours works fine then i would be interested to know why, must be something diff about your setup to get rid of the bug???
Update
Put Server files on Dual 933 + Sql Entperise + Windows 2000 Server...
From my testing so far, not had the bug (maybe couple stick but nothing serious). Rungate was switched to Orig version (I.e. - not hexed at all). Server is 30-06 plus a few more hex put in to close a couple more back doors and put a delay on Wizzy magic (they deserve it eh)...
Client I have tired is Visio One (3G if you like, old interface)...
Need to leave it running for a few days until im sure the problem is gone/less of and then i can format / try it on Athlon with windows 200 etc...
Buzznet
Well funny that - cos not a single post on LOM about this and i posted on a few forums and as soon as i mentioned it - the complaints came flooding in. People might think it is lag, they try to move, cannot, keep trying and then bug goes away and they carry on without much thought. As i say - its def more prominent if they have a pet and its in line with you...
I did a search on Euromir3 forums, lot of posts about it on there + on chinese forums they are speaking a lot about it also - so its def a bug, if yours works fine then i would be interested to know why, must be something diff about your setup to get rid of the bug???
1 other thing i noticed these moving bug almost vanished after switching to JoB server and 1.45 Olymp client
these bug was multiply more often encountered with old 2.9b client
btw serverhardware not changed with switching to JoB
moving bug now almost non existent anymore (average usercount 50) but its still there..i hope its getting completly eliminated with new p4 ht hardware of the gameserver
will post infos after hardware upgrade tomorrow
these bug was multiply more often encountered with old 2.9b client
btw serverhardware not changed with switching to JoB
moving bug now almost non existent anymore (average usercount 50) but its still there..i hope its getting completly eliminated with new p4 ht hardware of the gameserver
will post infos after hardware upgrade tomorrow
- Dec 1, 2003
- 1,806
- 2
- 165
TritoN said:
A What Bug???? Maybe i never made myself clear here.....Lets recap just in case...
V2.9 - I had this running 100% bug-free - every single bug (hitting Bugs, Dead mob Bug etc) wiped out and tested to death.......
JOB Server, never noticed bugs at first to be honest - was not until lately when i had someone on (1 person lol) that a bug came about. As i mentioned - you get STUCK on the spot, after a while you can move again. For a GM this bug means nothing but for a player - death is around the cornor
What I am trying to esablish here if possible (in case you missed it it my Dribble)...
1. Do people who are running a JOB server get this bug, either a lot, not much or none at all.
2. Trying to work out if the problem is server side, client side, both or hardware related..
3. Maps - does the bug go away 100% if the new Maps are used for example..
Think thats about - basically I hate a Bug, any bug - no matter how small or how big, if i find a bug which is "Random" then that is DAMM hard to pin down, tested my server again (been runing for some time now) and still no stick bug.... Yet Euromir3 have it as people keep posting about it....
lol ok then call it stick bug
as i sad it almost vanished after switching to JoB and the new client
btw another point: the mir client/server systems seems to have difficulties while talking over a wlan..most of our users have bad lag while connected via an wlan but got no problems if connected to the router via cable..this may be related to this topic in some way...
as i sad it almost vanished after switching to JoB and the new client
btw another point: the mir client/server systems seems to have difficulties while talking over a wlan..most of our users have bad lag while connected via an wlan but got no problems if connected to the router via cable..this may be related to this topic in some way...
here another client to try http://mir3.wmwn.com/wg3g.exe
and the update for it http://mir3.wmwn.com/cq3g.exe
and the update for it http://mir3.wmwn.com/cq3g.exe
- Dec 1, 2003
- 1,806
- 2
- 165
Well first of all - its good were all discusiing it- chucking ideas in etc - hells bells we will all be chinese soon (no disrespect to chinese people btw)...
Since i formatted the pc (atlhon machine) with windows 2003 server the bug has simply gone away - maybe something to do with Nforce 2 drives and the memory controller - cos remember i said if i left the machine on doing nothing then ran the server the bug was out in force where as a fresh boot it was fine. At least now when im reading the chinese forums i can understand what they are talking about ref Landing bug as i thought they meant not connecting (like using Euromir3 client on JOB server type thing)...
Btw Nightwolf - tried 5-6 clients and its def not that, least not that I saw anyhow...
Since i formatted the pc (atlhon machine) with windows 2003 server the bug has simply gone away - maybe something to do with Nforce 2 drives and the memory controller - cos remember i said if i left the machine on doing nothing then ran the server the bug was out in force where as a fresh boot it was fine. At least now when im reading the chinese forums i can understand what they are talking about ref Landing bug as i thought they meant not connecting (like using Euromir3 client on JOB server type thing)...
Btw Nightwolf - tried 5-6 clients and its def not that, least not that I saw anyhow...
- Dec 1, 2003
- 1,806
- 2
- 165
Update for you all - I now can recreate this bug almost all the time.......
I started the Server (all 6 programs) left it running for 4 days or something and not once did i get the Bug...I thought it was a pc releated matter and never thought any more about it. Now the other night, i closed down RunGate, Selchr gate and the server.exe - I had chnaged a few merchants and needed to restart the server you see to pick these up. Notice though I never closed and restarted DBSRV, LOGINSRV etc - this was mainly to save a little bit of time restarting them...
Soon as i logged in, within 5 mins the bug came back. I rebooted all the programs and the bug went away again. So basically from what I can tell so far, if you dont restart all the programs, there is a timimg issue of sorts that causing the bug. I tested this on my Dual 933 machine to see If the same happened and surely enough it did...
This would probably explain why some people never had the bug as they probably re-started all of the programs maybe...
I started the Server (all 6 programs) left it running for 4 days or something and not once did i get the Bug...I thought it was a pc releated matter and never thought any more about it. Now the other night, i closed down RunGate, Selchr gate and the server.exe - I had chnaged a few merchants and needed to restart the server you see to pick these up. Notice though I never closed and restarted DBSRV, LOGINSRV etc - this was mainly to save a little bit of time restarting them...
Soon as i logged in, within 5 mins the bug came back. I rebooted all the programs and the bug went away again. So basically from what I can tell so far, if you dont restart all the programs, there is a timimg issue of sorts that causing the bug. I tested this on my Dual 933 machine to see If the same happened and surely enough it did...
This would probably explain why some people never had the bug as they probably re-started all of the programs maybe...
Here is something I just noticed there is a huge difference in the processor usage between EISerevr-hexed and EIServer-30-6. I am talking Hexed runs at about 8 or 9% for some reason and 30-6 runs about 85 to 98%(old 2.9 days).
I know the old 2.9 was a processor hog and Job was allot more conservative there, that was one of the big differences I noticed and like about the JOB server.
Now this difference was discovered on my working/testing server not my actual running server, I work on the files and test new things on this server then when I am happy I transfer the items to the running server. The running server again like I said before is a quad compaq and it really doesn't matter with that system as no one processor ever goes above 30% before distributing the load to the others. by the way I am running Windows Advanced Server (4 Processor edition) on that server and it does distribute and arrange the processor usage as needed. I use "coolmon' to constantly monitor it.
I started using 30-6 on the main server about 6 days ago and last night someone complained about freezing up and dieing. They were playing computers side by side with somebody else and the other person did not experience it so it could have been her computer.
Just thought you might be interested in that.
I know the old 2.9 was a processor hog and Job was allot more conservative there, that was one of the big differences I noticed and like about the JOB server.
Now this difference was discovered on my working/testing server not my actual running server, I work on the files and test new things on this server then when I am happy I transfer the items to the running server. The running server again like I said before is a quad compaq and it really doesn't matter with that system as no one processor ever goes above 30% before distributing the load to the others. by the way I am running Windows Advanced Server (4 Processor edition) on that server and it does distribute and arrange the processor usage as needed. I use "coolmon' to constantly monitor it.
I started using 30-6 on the main server about 6 days ago and last night someone complained about freezing up and dieing. They were playing computers side by side with somebody else and the other person did not experience it so it could have been her computer.
Just thought you might be interested in that.
- Dec 1, 2003
- 1,806
- 2
- 165
Thanks for the Input Buzznet - Ive seen posts (lots) on Chinese forums about 100% CPU Usage - now mine runs at 4-6% Oddly enough though again I dont have anybody on the server. I have a few versions of the 6 Magic EI exe as it goes, some use a DLL, some dont at all (yet still the magic works -- odd)...
Wanna hear a really odd thing---- Run Server on Windows XP SP2 + SQL Desktop edtion, no pause on magic (Tbolt etc)- same server but using Windows 2003 and SQl Enterprise instead and the delay is as it should be on spells, no slow down on anything noticed, no "STUCK" on the spot bug etc....
One thing I did do though was roll back the drivers for Memory controller to default (something I looked into ref NFORCE 2 boards). Looking on the chinese forums seems to show something:
* Some people reporting 100% cpu Usage
* Some people reporting "they cannot move" reffered to as Landing Bug
* Other people have no problems what-so-ever...
As proven by myself - depending on What OS/Driver set etc I run it on - depends on what Bug I get as it goes....wierd eh..
Wanna hear a really odd thing---- Run Server on Windows XP SP2 + SQL Desktop edtion, no pause on magic (Tbolt etc)- same server but using Windows 2003 and SQl Enterprise instead and the delay is as it should be on spells, no slow down on anything noticed, no "STUCK" on the spot bug etc....
One thing I did do though was roll back the drivers for Memory controller to default (something I looked into ref NFORCE 2 boards). Looking on the chinese forums seems to show something:
* Some people reporting 100% cpu Usage
* Some people reporting "they cannot move" reffered to as Landing Bug
* Other people have no problems what-so-ever...
As proven by myself - depending on What OS/Driver set etc I run it on - depends on what Bug I get as it goes....wierd eh..