Topic: 'SG5 Server Status, up or down...

[KBFLordKrueg]

'This thread dedicated to maintain that SG5 is always up and running. Please post whenever the server goes down. TY

[Dizzy]

I dont think the process sentinel is working... thats what auto reboots it.

I'll see if someone is logged in to reboot it...

[Corbomite]

That mixed with the AI generation on the map I think is causing too much of a load. Remember the AI calls up to 3 missions every time it moves too.

[Dizzy]

I'll tell bonk to drop it down to two mission choices... unless I find the gf setting myself. That'll help Im sure.

I logged in and closed the server kit... and it started back up. Mb the process sentinel is working again. Will login to see...

[Corbomite]

Its in the Mission Matching .gf under Move =. Set it back to 2. I'd rather turn the AI off though as it is usless and clutters the map.

[Dizzy]

Ok, well it came up by itself, hehe. So the last two times it got whacked was by me.   I think the process sentinel is working again, doing auto reboots, but it may have been another admin in there. I guess we wil find out soon.

Should be ok now.

Who all has access to rebooting this baby? It's best to post on the forums announcing you are the one that is gonna log in to reboot it... otherwise we can mess each other up.

[Dizzy]

Its in the Mission Matching .gf under Move =. Set it back to 2. I'd rather turn the AI off though as it is usless and clutters the map.

Ok, ill see if I can change the mission choices back to two the next time it crashes and I catch it in time. AI cannot be turned off, for some reason the kit needs it there or the db starts spewing out errors. Weirdest thing ever. The monster population needs to be cut back... The less AI the better just as long as there are some, the kit is happy.

Edit: post here again if it crashes even if it auto boots... I'll be afk a bit.

[Corbomite]

Well that's weird. We had it shut off for years and had no problems. What changed?

[Dizzy]

Thats a very good question. Although u may have had it shut off, doesnt mean the db wasnt accumulating errors. If you recall some of the GW servers, also known as the crash a lot servers, the shiplist was primarily responsible... Each turn break the processor spike was quite noticeable. When errors accumulate in addition to other things that are causing processor usage inthe kit, it takes only a second or so at 100% processor usage to kill it, but DH's machine was so fast, it tolerated the errors much better than XC's servers and thus appeared to be more stable.

Anyway, we should really delve into AI and see whats up there.

[Corbomite]

I'm talking back during the CW servers. BBJ had them running very smoothly.

[Dizzy]

Well, everything worked stock... with bugs galore. Funny how we fixed most the bugs, but now it doesnt work.

The server just went down, but I dont want to change the gf setting until I know how to suspend the process sentinel long enough to make the change to the mission choices offered upon move...

So anyone know how to pause the process sentinel?

[Corbomite]

You could ask Frey. He's on D.net TS right now.

[Dizzy]

Abrupt gf server setting edit... hang on...

[Dizzy]

Server is back up on process sentinel with a 2 mission per move. Sorry bout all the problems till now... should be better.

[Bonk]

Is not the gfs, its no the AI, if others are editing them I'm done. The forge is using similar settings but is on OP+4 and is stable.

[Corbomite]

Is not the gfs, its no the AI, if others are editing them I'm done. The forge is using similar settings but is on OP+4 and is stable.

Ok. If you can give a reason why the shiplist would cause the server to crash I'm all ears. I hate modded shiplists anyway so it would be no loss to me if we went "stock".

[Hexx]

I think the shiplist can do it as it effects the  requirement for missions BPV draw.
Don't think it'sa huge problem though, of course I very much like the expanded shiplists.
(Carriers for Lyrans are simply cool, too bad a buncha whiney gits got rid of the CVP's/CVD's)

[Dizzy]

Tracey just confirmed what I have suspected all along about having more than 2 mission choices per move on flatfiile, that it will crash.

Server should not crash anymore... But I wonder why chat isnt working?

[Bonk]

Tracey just confirmed what I have suspected all along about having more than 2 mission choices per move on flatfiile, that it will crash.

Server should not crash anymore... But I wonder why chat isnt working?

Survey says... wrong!

We need to remove donors form the serverside shiplist
and pull the EEK missions, this will stabilise the server.
(I suspect software firewalls are why it does not
work on the live server but works fine in testing)

But it looks like a lot has been changed since I was away
at that funeral. (webmap and chat both gone)

If I can figure out what has been done while I was away,
I'll edit the serverside list and remove the eek missions tomorrow.

If you do not agree with this assesment then we're out of luck.

If I'm to try and rescue it, then I need to know everything
that was done while I was away, So I can put it back the way it was
then stabilise it as described above.

I am taking the webmap offline to reduce errors on the webserver
until a course of action is decided on, then I will look to see what
happened to the autosave folder share if I try to rescue this.

[Dizzy]

Its not crashing like it was since I changed the gf mission on move setting from 3 to 2. Thats the only gf setting change. And the chat which was down is working again. I think it was related to the forums crashing and all the rest.

According to you, Bonk, that gf setting has nothing to do with crashes, yet TRacey and DH say differently. They say a flatfile server offering more than 2 missions has never worked and wont.

So I dont know the whys or whos or whats here... Mb you guys need to talk and figure this out, because if the admins have different ideas on how a server works we are all f*cked.

I dont want to remove eek missions if we dont have too. They are the only ones that AI strip. We need to talk about this i think... Till then Im logging on and killing someone...

[Dizzy]

There's an horde of AI on the map... We need a DB clean. Gonna reboot.

[KBFLordKrueg]

Probably should do one of those everyday anyway...maybe even twice a day. 

[Dizzy]

I cant find Bonk. The DB is huge atr 2.6mb. It's crashing once every hour now. There are AI everywhere... This must be a great processor and stable server settings cuz Ive never seen a db this lage not crash every 15 mins.

As soon as I can get ahold of bonk, Ill see if I can get him to do a db clean... Till then, it will crash every hour... or more. Nothing I can do til he comes along. Mb Bonk will show me how to do a db clean. 

[Hexx]

I don't know if it means anything, but if I right click on a hex to determine ownership the
"are a longe term ally/are mortal enemy" line repeats 2-3 times rather than just once.
ie
(24,17) United Federation of Planets
The United federation of Planets is a competent minion of our Empire
The United federation of Planets is a competent minion of our Empire
The United federation of Planets is a competent minion of our Empire

[Corbomite]

I don't know if it means anything, but if I right click on a hex to determine ownership the
"are a longe term ally/are mortal enemy" line repeats 2-3 times rather than just once.
ie
(24,17) United Federation of Planets
The United federation of Planets is a competent minion of our Empire
The United federation of Planets is a competent minion of our Empire
The United federation of Planets is a competent minion of our Empire

They're just being polite. They are repeating it because they know you have to jump up to see the screen.

[KBF-Soth]

 

[Dizzy]

rebooting server... I'll be gone for an hour anda half... so hope it stays up and Bonk cleans the DB.

[762_XC]

Down again 00:17 EDT.

[Commander Maxillius]

I knew something was fishy when my PC stayed up


Damn Windows, if SFC could be ported to the Mac I'd play for life!

[Riskyllama]

Down again 00:17 EDT.

Maybe its up, but they're jamming us.
Still down as of 00:03 CDT

[Dizzy]

Back up. 

[KBF-Soth]

dwn

[Dizzy]

Server is down for a DB clean by TraceyG. Bonk is busy atm.

[KBFLordKrueg]

Strange drops/burps tonight. Multiple times 99% of the players would suddenly drop, TS would also drop most of the players.
But, always, there would always be 2-5 players (out of about a dozen) who didn't lose connection.

[Dizzy]

Thats not the server, those are connection issues. The server has been remarkable stable since the DB clean. There are 22 peeps on aand have been fpor a while. Encouraging.

[Dizzy]

yup no crashes in a long time. Very stable. But what machine is SG5 running on? I logged into thor and its not there.

[FPF-DieHard]

yup no crashes in a long time. Very stable. But what machine is SG5 running on? I logged into thor and its not there.

Are you guys riunning this under the same login?    You should be, if this is run under Tracey's loging, you wlll not see this in your session.

[Dizzy]

she was running under her login. Wer fixed that.

Bonk, we had 25 peeps on last night after tracey did a DB clean and all thru the night no server crash. We took it down this morning and she had me learn to do a DB clean. Had no idea how easy it was.

Everything is running smooth. Cant pull the EEK, we'd not have AI strip. Server really isnt crashing at all... so I say leave it alone.

Hey, we all missed u this weekend! glad ur back.

[KBF-Soth]

mmmm...guess what.......dwn

[FA Frey XC]

They probably just restarted it under the correct service account.

Regards,

[Khemaraa]

My server list shows it up, but it says I have a bad CRC on the ship list.   

(sigh!)

[LordSaxon]

My server list shows it up, but it says I have a bad CRC on the ship list.   

(sigh!)

Same thing here, downloaded the shiplist Zip file, put it in my assets/specs file, same problem. What gives?

[KBFLordKrueg]

7:40 pm EDT server down...5 min, no restart yet...

[KBFLordKrueg]

My server list shows it up, but it says I have a bad CRC on the ship list.   

(sigh!)

Same thing here, downloaded the shiplist Zip file, put it in my assets/specs file, same problem. What gives?

Khemaraa made it on...what did you find was the problem K?

[KBFLordKrueg]

7:40 pm EDT server down...5 min, no restart yet...

8:00 pm...still down...been 20 min...

[Dizzy]

Its up, been up. Be nice when it gets rebooted peeps would say so. 

[Commander Maxillius]

up, take a mission, win, lost connection.

log back in

up, take a mission, win, lost connection.

log back in

up, change ships to comething faster, take a mission, win, lost connection.

quit.



also note others had my same idea.  1st log in there were 15 people logged in, 2nd: 8, 3rd: 3

it does get old.

[762_XC]

Diito on d.net TS. p00py network problems!

[Dizzy]

up, take a mission, win, lost connection.

log back in

up, take a mission, win, lost connection.

log back in

up, change ships to comething faster, take a mission, win, lost connection.

quit.



also note others had my same idea.  1st log in there were 15 people logged in, 2nd: 8, 3rd: 3

it does get old.

Then stfu. It's getting old hearing you peeps complain about it all the time. If you are gonna quit, quit and stop posting about it. If you dont have something good or contructive to say, dont.

FYI the server isnt crashing, there were network problems. It probably doesnt matter to you why it went down only that it isnt up. But if you dont know where to place blame or how to be contructive, then just report its down plz. Thats the purpose of this thread.

I'm really disgusted with the few naive players who think threatening everyone is gonna quit because of a server burp will produce something contructive. Well it wont so STFU.

FYI, SG5 is VERY stable. VERY few crashes. It can hold a player load 30+ peeps. It can go for hours on end without crashing. Occasionally it will crash. This is due to players who Alt in mission and forfeit missions (so mb STOP alting and forfeiting). Other times it is due to netwrok issues. For the most part, its up most of the time. If you have ideas on how this volunteer server can be run more effectively and efficiently, please by all means share with us your wisdom. Till then, a simple post with server ups or downs will suffice. TY.

[KBFLordKrueg]

 
Damn, Dizzy...having bad day...? 

[Dizzy]

My recent sessiopn with thor was terminated. Mb a network issue. XC TS went down too. Not SGO5 related. But we do run off that machine. Dunno if its on the network side or with freys machine...

I'l see if I can get a session to hold and clean the DB which kinda needs it anyway. DB is over 1 MB.

May be down for a bit.

EDIT: Posted here: D.Net Directory Services Status and Outages Thread

Most likely a network issue. It may clear up on its own... If so, I'll wait to do a DB clean till later unless someone beats me to it. If not, we wait for word from Frey.

afk for a while...

[Commander Maxillius]

up, take a mission, win, lost connection.

log back in

up, take a mission, win, lost connection.

log back in

up, change ships to comething faster, take a mission, win, lost connection.

quit.



also note others had my same idea.  1st log in there were 15 people logged in, 2nd: 8, 3rd: 3

it does get old.

Then stfu. It's getting old hearing you peeps complain about it all the time. If you are gonna quit, quit and stop posting about it. If you dont have something good or contructive to say, dont.

FYI the server isnt crashing, there were network problems. It probably doesnt matter to you why it went down only that it isnt up. But if you dont know where to place blame or how to be contructive, then just report its down plz. Thats the purpose of this thread.

I'm really disgusted with the few naive players who think threatening everyone is gonna quit because of a server burp will produce something contructive. Well it wont so STFU.

FYI, SG5 is VERY stable. VERY few crashes. It can hold a player load 30+ peeps. It can go for hours on end without crashing. Occasionally it will crash. This is due to players who Alt in mission and forfeit missions (so mb STOP alting and forfeiting). Other times it is due to netwrok issues. For the most part, its up most of the time. If you have ideas on how this volunteer server can be run more effectively and efficiently, please by all means share with us your wisdom. Till then, a simple post with server ups or downs will suffice. TY.

I meant quit for the night.  I just thought I'd give a creative account of what happened.  I played three missions last night where no points were recorded and I got frustrated, which I see you are as well.  I came back on after about an hour and there were no more problems and my apologies for not posting here about that too.  I also played today for about an hour and that was flawless as well.  Your server *is* the most stable and the fastest I've ever played on, it's just sad that the network is such a problem.

Please remember to read the post before making an asshat out of yourself and assuming I'm just an alting n00b.

[Dizzy]

Please remember to read the post before making an asshat out of yourself and assuming I'm just an alting n00b.

Perception is everything. Like you said, most peeps dont post when the server is up and running flawlessly. They only see these down posts and what do you think they think?

FYI: Frey says the network provider was responsible tonight for all the ups and downs... XC's server, TS, and forums all experienced issues. Unfortunately, after I logged off to get some sleep, no one posted the problem was resolved.

Apparently its been working for quite a long time and the network issue was brief. Still have to do a DB clean, but the server isnt dropping so I'll leave it alone till there are fewer peeps on the server. Still about 18 on.

[Dizzy]

Server DB cleaned.

[Dfly]

TY Dizzy

[Commander Maxillius]

Please remember to read the post before making an asshat out of yourself and assuming I'm just an alting n00b.

Perception is everything. Like you said, most peeps dont post when the server is up and running flawlessly. They only see these down posts and what do you think they think?

FYI: Frey says the network provider was responsible tonight for all the ups and downs... XC's server, TS, and forums all experienced issues. Unfortunately, after I logged off to get some sleep, no one posted the problem was resolved.

Apparently its been working for quite a long time and the network issue was brief. Still have to do a DB clean, but the server isnt dropping so I'll leave it alone till there are fewer peeps on the server. Still about 18 on.

If it makes you feel better, it took me about 20 minutes to post a reply once I read your post because my first reaction was to start a big-ass flamewar, but then I realized that you're under serious stress and you really *REALLY* care about this game and everyone's enjoyment of it and it just takes one guy too many to post "server's down" to set the bomb off. 

So I apologize for any percieved hard feelings, none were intended.  I'll be on the server tonight in my KE giving some Plasma R lovin' to the Feds

[Dizzy]

No need to apologize, + karma to u, it was me who needs to apologize cuz i snapped at ya.

Yes, perception is a BIG deal for me. No where is there a thread discussing the coolness of the server. I'll say its incredibly balanced, fun, great pvp and all sorts of other goodness, but its a little light on players. I believe its because of the 1st week false start we had and the impression to new and casual players that the server is down a lot which it isnt... Well everythings fixed and running smoorthly.

Ironic tho, atm the server is down, hehe. Newtwork issue, server DB did not crash. MCI prolly, the forums were down a min too. Did a DB clean this morning so we are real good till tonight or tomorrow.

[Commander Maxillius]

Well, I'm having connection problems too.

I have to turn off the Windows firewall to get on the server, but while the firewall's off I'm vulnerable to anything and everything that finds me.

But I'll be on, if not tonight then tomorrow.

[Dizzy]

The last netwrok disruption lasted 5 mins or so. Been fine since.

[el-Karnak]

Que Pasa?

This server is bouncing like racquet ball on steroids. 

[Hexx]

Is it the AI markers?
I noticed (in Lyran & Fed space anyway) there are dozens of AI markers- and I was only in a few hexes, there must be hundreds over the server-mostly pirates/monsters
Almost one in every hex and every third or fourth hex has 3 AI ships.

There's no way this many were created by players, maybe pirate production is in Overdrive?

(or maybe it has no effect whatsoever..)

And if it matters- these were hexes far away from the front lines, so they weren't being created by players running missions.

[762_XC]

It's a network issue, not a server issue. TS and Dnet are bouncing along with the server.

[Soreyes]

yep it was kind of bad tonight. The Network was real choppy.

[Dizzy]

DB Clean, server will be back up in a few after this post.

[Commander Maxillius]

I've noticed that it's rather pointless to try to get on between 10:30 and midnight eastern.  I figure your ISP is doing maintenance or something at that time.  Before that everything's fine, and after it's clear sailing too.  If you could pinpoint what time exactly the network gets fubar, that wuld be a perfect time for DB cleans and such.

It sucks much that it looks to be in primetime though.

[Dizzy]

There have been several nbights where there were disconnects. I dont know why. Ask Frey, doubt he knows. O know he's complained about it tho.

DB cleans take 5 mins. I do them when no one is on. There is no predictability to network issues.

[Hexx]

.

DB cleans take 5 mins. I do them when no one is on. There is no predictability to network issues.

Sure there is.
Whenever I have a 3hr+ block of time free to play.
Look for more network problems tonight sometime after 8pm Eastern.

[Commander Maxillius]

.

DB cleans take 5 mins. I do them when no one is on. There is no predictability to network issues.

Sure there is.
Whenever I have a 3hr+ block of time free to play.
Look for more network problems tonight sometime after 8pm Eastern.

Definately.  I'm not even going to try logging back on until after midnight.  However, it's working great now, but since I have my firewall turned off so I can play, all sorts of crap I don't have to deal with on my mac are afflicting my machine, such as that lsass bug and various other things that freeze the whole show up.  If anyone wants to know why I switched to Mac for the majority of my computing, just turn your firewall off and wait 3 minutes.  You don't even need to visit any sites.

[Dizzy]

Server DB cleaning... Done

[Dizzy]

Having server issues right now. Bonk has been notified.

Dynanet has been up and down and sluggish for the past hour on and off. I had a few tries logging into XC TS fail.

Players report not being able to successfully draft other players in PvP or co-op. Endless countdowns and disconnects. Players get hung at the map screen and have to logout and reboot OP even. Some take forever to get the map to load.

Did a DB clean shortly after experiencing some of these issues. Thought a DB clean would help, but apparently that isnt the issue...

To be honest, I do not know what is wrong, Doesnt make sense to leave the server up in this condition so the VC's for tonight and the server itself have been suspended till I get the server problem worked out with Frey and Bonk and talk to the RM's on the VCs.

Diz

[Bonk]

The Dynaverse.net website was down for about 10 minutes while I updated Apache. It was fine before and after from my connection. This is unrelated to any problems on SGO5. I'll look into the issues with SGO5.

[KBFLordKrueg]

 

[Dizzy]

ty bonk!

[Bonk]

Just started it, logged in, was drafted by Julin and then drafted Julin, logged out and restarted it on the process sentinel - all fine. (Neither Julin or I were running software firewalls.) I changed nothing.

Server is up.

edit: but I did empty the rather full (of old dbs) recycle bin and remove the copy of the cleaner in the server root folder.

edit#2: I backed up the db at 7:27 pm. A daily db clean should not be necessary, this db should hover around 1MB. A clean every 2 or three days should be all that is needed. Server AI will not accumulate, only dead mission AI will accumulate. Please turn off your software firewalls to minimise bad connections and thus dead mission AI and reduce the need for db cleaning.

[FPF-DieHard]

Just started it, logged in, was drafted by Julin and then drafted Julin, logged out and restarted it on the process sentinel - all fine. (Neither Julin or I were running software firewalls.) I changed nothing.

Server is up.

edit: but I did empty the rather full (of old dbs) recycle bin and remove the copy of the cleaner in the server root folder.

edit#2: I backed up the db at 7:27 pm. A daily db clean should not be necessary, this db should hover around 1MB. A clean every 2 or three days should be all that is needed. Server AI will not accumulate, only dead mission AI will accumulate. Please turn off your software firewalls to minimise bad connections and thus dead mission AI and reduce the need for db cleaning.

It's gotta be something on the internet.   I can sign in now, but the map takes about 30 seconds to load.   This is what it was doing before.

Sofware firewall is offf, getting 5 MB down and 1 MB up on cable connection.

I cannot even draft AI right now.   Count down to the map in empty hexes.

[Bonk]

It's gotta be something on the internet.   I can sign in now, but the map takes about 30 seconds to load.   This is what it was doing before.

Sofware firewall is offf, getting 5 MB down and 1 MB up on cable connection.

Map will always take longer to load then you are used to. We are using xenocorp's IRC server. It does a hostname lookup and does an identd - neither of which the gamespy IRC server does as I recall. Once this is done and IRC login is complete, then the map will load. THe server is not slower, only IRC login is slower, it responds at full speed in all other chat tasks. Theis way, chat does nto go down regularly, we always have chat, as the gamespy IRC server periodically dumps Dyna servers for creating more channels than their server allows. I discussed this in detail with gamespy, they acknoldeged the problem but refused to do anything about it. Firesoul and I thn identified UnrealIRCd as an apporporiate IRC server for dynaverse servers in testing about 4- 5 years ago...

I cannot even draft AI right now. Count down to the map in empty hexes.

Server AI are not "drafted", mission AI is created. I'll login again to test some more...

edit: I just got a timeout on the webserver... something may be up with the network...

[stoneyface]

heya bonk.

just logged on and got a report that some people can't draft others or AI. seems to be hit or miss as to who is affected. just reporting in!

[Julin Eurthyr]

This is what I've seen tonight:

Bonk and I did the counter-draft thing, both tries were excellent.

Got drafted after Bonk's restart, no problem.

Got 7-8 people online, started getting "unable to draft AI issues".  Personally, my first solo try counted 60 secs then "connection to the server has been lost".  Re-signed on, had to play that particular mission I had started earlier.  That worked, with full DV shift, hex flipping neutral, etc.  Second try was till 40 left on clock before being dropped (might have had something to do with the fact that I right-clicked a player name while waiting for the draft), however the mission was not waiting for me upon my return...

Came off server to post observations to this moment...

[Bonk]

Getting weirdness now also - Catothelder disappeard from my map and list yet was still logged in...

[Bonk]

Logging debug output to look for a cause...

[Bonk]

So far no errors in the debug output... All I noticed out of the ordinary tonight was "K" on the chat channel instead of K'Tool2853408246... please avoid using apostrophes in your charactername.

I'll shut it down once I have 1MB of debug output. Then I'll restart it under the process sentinel with debug output off and go over the log file.

I'm seeing a lot of player drops from the IRC channel / server.

[Bonk]

Just got two timeouts on the webserver here and my remote session on thor just died, pointing to network... I checked heimdall and the virus scanner is tying up a constant 25% cpu. heimdall is the teamspeak and IRC server. Frey, is there a nightly scheduled virus scan on heimdall?  (Did I mention I think Norton is evil?)  I suppose this would have shown before if it was the problem...

edit: I'm informed teamspeak just burped also...

I'll check Loki's web logs for DoS attacks... can ya check Heimdall's? Is SFC2.Net still OK?

[Dizzy]

Is there any way, Bonk we can figure out if its the DB or a network issue? Not sure where SGO5 is going right now...

[Bonk]

I went over the debug output from last night. I did not find any errors. Which would have me leaning toward network issues.

On the other hand, we know the flatfile db is corrupted somehow ever since the webmap has been unable to read it. But did this current problem coincide with that event? I don't know. Can you say when this started? I have not taken the time to go over the db with a fine toothed comb. I have only taken a quick look at it.

I will download the current db and have a look at it.

[FPF-DieHard]

Is there any way, Bonk we can figure out if its the DB or a network issue? Not sure where SGO5 is going right now...

To test the DB:   Reset it (backup the old DB) and see if the bad behavior continues.

To Test the Network:  Host someplace else to see if bad behavior continues.

[Bonk]

Good thinking DieHard, that's worth a try.

I just found a "Database Clean in progress.scr" in the root server folder, not sure if this will confuse the server...(very well might) I zipped it into the db backup folder.

[Soreyes]

Reads the above stuff...... Hmmmmmmm we all know that it's a plot to keep me and Chuut from over running the klingons and Hydrans this morning

Server went down at 0520 pst.   Was working fine for the last 4 hours though

[KAT Chuut-Ritt]

Server seems to be up.....but seems to be started over beginning map and 2265 is the year 

[Bonk]

Yes, as per DH's suggestion above, I am testing to see if the database is corrupted. We are back on the starting db to test for today.

[Bonk]

I suspect it is related to the very heavy bandwidth demands of the Teamspeak server and the bandwidth limiter that is in place. Since the pronblem only seems to occur when the coalition is on in numbers... and they are using XC TS...

[KAT Chuut-Ritt]

Soreyes told me it had been going fine for 4 hours after the masses logged off last night.

[Soreyes]

suspect it is related to the very heavy bandwidth demands of the Teamspeak server and the bandwidth limiter that is in place. Since the pronblem only seems to occur when the coalition is on in numbers... and they are using XC TS...

Well thats easy to fix......... Take away there TS

[KAT Chuut-Ritt]

Although it stayed up fairly decently friday late night when about 20 coalition pilots were on after midnite. Of course don't know how many of them were on TS 

[FPF-DieHard]

I suspect it is related to the very heavy bandwidth demands of the Teamspeak server and the bandwidth limiter that is in place. Since the pronblem only seems to occur when the coalition is on in numbers... and they are using XC TS...

So, we get to blaim the Klingons?

All jokes asisde, there were very few people on when this was happening last night.

[Bonk]

Although it stayed up fairly decently friday late night when about 20 coalition pilots were on after midnite. Of course don't know how many of them were on TS 

So this problem has started since then?  When was the "K'Tool" account created?

[KAT Chuut-Ritt]

So this problem has started since then?  When was the "K'Tool" account created?

No idea but I believe he was happily droning away in his D5D on Thursday with a fair amount of pp in the bank.  He might have had the account beforehand, or he might have just waited for the arrival of the D5Ds.

[Bonk]

I suspect it is related to the very heavy bandwidth demands of the Teamspeak server and the bandwidth limiter that is in place. Since the pronblem only seems to occur when the coalition is on in numbers... and they are using XC TS...

So, we get to blaim the Klingons?

All jokes asisde, there were very few people on when this was happening last night.

It only takes two or more users on a Teamspeak server to create a heavy bandwidth demand. (why I ran an RW base when on cable instead of a TS server)

But your observation would suggest database corruption instead... We'll see what happens tonight on the starting db...

[Bonk]

So this problem has started since then?  When was the "K'Tool" account created?

No idea but I believe he was happily droning away in his D5D on Thursday with a fair amount of pp in the bank.

OK, good to know, it not likely that then...

[Hexx]

So this problem has started since then?  When was the "K'Tool" account created?

No idea but I believe he was happily droning away in his D5D on Thursday with a fair amount of pp in the bank.

OK, good to know, it not likely that then...

Hey!

Let's not jump to conclusions here...

[Bonk]

What we also noticed, at least a few of us was a hex number called
-1,-1 with horribly disfigured values for defense and such that a few
got when logging in. Dont know if thats related, but once logged in
that hex was no where to be found. And only several saw it, including
me.

Got this report... where exactly was this -1,-1 hex seen? On the map, in the news?

(IIRC the server sets a players location to -1,-1 when they are moving... lost packets during a player movement may result in this being observed inthe client...?)

[PGC Graakna]

12:10 EDT, tried to login to the SGOV server and it asked me to generate a new account. Has the DB been reset/whiped?

[Bonk]

12:10 EDT, tried to login to the SGOV server and it asked me to generate a new account. Has the DB been reset/whiped?

Read the above posts in this thread. (Yes, we are testing for database corruption vs network issues.) And don't worry... the db has been saved before the testing reset (though it may be corrupted).

[762_XC]

K'Tool was created several days before all this started. Maybe the third day of the server or thereabouts.

The drafting problems only started last night. Previous to that it was just network/burp issues.

[Bonk]

What we also noticed, at least a few of us was a hex number called
-1,-1 with horribly disfigured values for defense and such that a few
got when logging in. Dont know if thats related, but once logged in
that hex was no where to be found. And only several saw it, including
me.

Got this report... where exactly was this -1,-1 hex seen? On the map, in the news?

(IIRC the server sets a players location to -1,-1 when they are moving... lost packets during a player movement may result in this being observed inthe client...?)

Actually I have seen this before on slow map loads on other servers... I think  know what you mean.. only one hex is showing and it is -1,-1? I think this is how the client responds when the conection is interrupted on map initialisation...

[Bonk]

K'Tool was created several days before all this started. Maybe the third day of the server or thereabouts.

The drafting problems only started last night. Previous to that it was just network/burp issues.

Thanks for the info... The server was not burping, only suffering network interruptions, perhaps the db has been corrupted by these interruptions, resulting in the drafting problems...while the network interruptions continue..

[el-Karnak]

*Reads Thread*  

I'll check back in another couple of days to see if y'all have the dyna server stabilized.

[Bonk]

*Reads Thread* 

I'll check back in another couple of days to see if y'all have the dyna server stabilized.

Sigh... wish I could say that...   

[Capt_Bearslayer_XC]

I've noticed that AI seem to be on and moving, could that be an issue?

Also, is it not correct that some bases are destroyable and others are not?

If so could that be an issue also, IIRC, we were not able to do that before.

[Dizzy]

We just removed base station assault scripts, bear.

Bonk, if XC TS is using considerable system resources, why are we running it?

[Bonk]

I've noticed that AI seem to be on and moving, could that be an issue?

Also, is it not correct that some bases are destroyable and others are not?

If so could that be an issue also, IIRC, we were not able to do that before.

The server AI are a good thing no matter what others will tell you. They do not accumulate. (the server is creating one and destroying 10 per turn carrying 100, they cannot take affect DV.) Only AI from bugged missions will accumulate. (so turn off your firewalls)

Yes some bases are destroyable, others are not. (only because there are no Base tation assault missions though - simple idea, but a good one...). THis should not cause any problems, as it it has been the case since the server start.

[Bonk]

We just removed base station assault scripts, bear.

Bonk, if XC TS is using considerable system resources, why are we running it?

I cannot say for sure its use exceeds the network's capacity, I am just suspicious of it. Frey has a bandwidth limiter in place for xenocorp servers hoated at BearTech, but I'm not sure how it works and if it would result in lost packets or not.

[Hexx]

Sooo can't we fire it up again and ban the coalition from using TS?

[Bonk]

Lets see what happens tonight after 10:30 pm EST to see if it is network or db issues at this point. (still on the starting db)

And, we can't do that, but we can ask them to useanother channel on the SSCF teamspeak server tomorrow if we still see problems toinight.

[FPF-SCM_TraceyG_XC]

Good thinking DieHard, that's worth a try.

I just found a "Database Clean in progress.scr" in the root server folder, not sure if this will confuse the server...(very well might) I zipped it into the db backup folder.

That file is used for database cleaning. It is placed into the ValidatedClientsFile directory while the server is being run to clear the shipyard to prevent people from logging in. They will receive a CRC error message that says "Database Clean in Progress.scr" is missing. After the shipyard has been cleared, then this file is removed from that directory.

Did we ever find out if the database cleaner was removing ships up for bid still? If it doesn't, then we no longer need to clear the shipyard.

[Bonk]

Good thinking DieHard, that's worth a try.

I just found a "Database Clean in progress.scr" in the root server folder, not sure if this will confuse the server...(very well might) I zipped it into the db backup folder.

That file is used for database cleaning. It is placed into the ValidatedClientsFile directory while the server is being run to clear the shipyard to prevent people from logging in. They will receive a CRC error message that says "Database Clean in Progress.scr" is missing. After the shipyard has been cleared, then this file is removed from that directory.

Did we ever find out if the database cleaner was removing ships up for bid still? If it doesn't, then we no longer need to clear the shipyard.

I figured that was what it was for (I have one called "server_maintnenance.scr"), but I zipped and moved it anyway to be prudent, considering the way the server traverses the server folder for gf files, it would not be inconsistent for it to look for .scr files the same way...

Have not tested the shipyard on db cleans yet... just testing for db corruption vs network issues at the moment... but we do need to test that at some point.  (use at least 2 second turns on thor on my gfs to run up the clock or clear the yards, as the server is trying to delete 10 AI per second per turn change so turns need to be >1 second)

[el-Karnak]

*Reads Thread* 

I'll check back in another couple of days to see if y'all have the dyna server stabilized.

Sigh... wish I could say that...   

Mesa very happy not to be Admin.  I get enough of that stress at work, thank you very much.  Even when I delegate the job, it's not very fun.

No me gusta dyna Admin job.

I give +karma to you for trying to fix it. 

[Dizzy]

Bonk, lets make available to DH the entire current server settings and DB and all that. He will host it on his machine tonight. We will start the DB where we last left off when it was having problems. If its truly network issues then drafting and all that weirdness will be gone. If it is a DB problem, it will be apparent right off. Then we can decide how to fix it. Whether you can debug the DB, or how far we go back to a backup thats not corrupted or if we ditch the whoile thing, we can figure out after this test.

[762_XC]

Bonk, lets make available to DH the entire current server settings and DB and all that. He will host it on his machine tonight. We will start the DB where we last left off when it was having problems. If its truly network issues then drafting and all that weirdness will be gone. If it is a DB problem, it will be apparent right off. Then we can decide how to fix it. Whether you can debug the DB, or how far we go back to a backup thats not corrupted or if we ditch the whoile thing, we can figure out after this test.

 

[KBFLordKrueg]

And...so when are we getting back to the map as it was last night?
Like as in Time...? I assume it's tonight...?

[Capt_Bearslayer_XC]

From an MSN converstation I had with Frey a bit ago...

Bearslayer says:
have you been able to help Bonk tract down any of the server issues we have been having?
FA_Frey_XC says:
It's nothing to do with the server
FA_Frey_XC says:
It's been our connection
FA_Frey_XC says:
I keep telling people that
Bearslayer says:
'our' connection?
FA_Frey_XC says:
Work connection
FA_Frey_XC says:
Up here at the office
Bearslayer says:
ah, ok... so basicly, your work is having connectivity issues and when a bunch of d2ers get on it amplifies the problem
FA_Frey_XC says:
No
FA_Frey_XC says:
It doesn't matter who gets on
FA_Frey_XC says:
It had absolutely nothing to do with anyone connecting
FA_Frey_XC says:
It was a BGP routing issue that resulted from MCI's "flapping"
FA_Frey_XC says:
Which was causing routing tables to be re-written
FA_Frey_XC says:
thus, some could connect
FA_Frey_XC says:
while others couldn't
FA_Frey_XC says:
then they could
Bearslayer says:
ah... ok.  Have you posted to that effect on the thread they got going?
FA_Frey_XC says:
but mission matching wouldn't work because once the serverkit matches peeps inna mission
FA_Frey_XC says:
No, I haven't had time
FA_Frey_XC says:
I'll try to tonight
FA_Frey_XC says:
But my home connect has been flakey for the last day and a half 2
FA_Frey_XC says:
When it rains, it pours eh ?
Bearslayer says:
Can I cut and paste what you just told me?
FA_Frey_XC says:
Sure, no one will understand it anywho
FA_Frey_XC says:
I'll make a post there in about an hour
Bearslayer says:
Well, it might help....
FA_Frey_XC says:
Help what ?
Bearslayer says:
b/c they will have a sense of what is going on and stop pulling their hair out over something they have no control over
FA_Frey_XC says:
Who's pulling their hair out ?
Bearslayer says:
Bonk and Dizzy
FA_Frey_XC says:
If anyone had half a clue, then they'd know it's got nothing to do with the serverkit
FA_Frey_XC says:
The serverkit isn't crashing
FA_Frey_XC says:
has it been crashing ?
Bearslayer says:
Well, that is the primary suspect right now
FA_Frey_XC says:
Because if so then that's a different issue
Bearslayer says:
That and using D.net TS at the same time as the server
FA_Frey_XC says:
??
FA_Frey_XC says:
Those are on two seperate servers
FA_Frey_XC says:
LOL
Bearslayer says:
There is a 'test' server up right now to see what is going on
FA_Frey_XC says:
Did someone actually think that could cause a problem ?
Bearslayer says:

FA_Frey_XC says:
"Test" what, did they move it off THOR ?
Bearslayer says:
IE You really need to chime in...

[Dizzy]

So the problem this whole time has been 'MCI FLapping'?

 

And I thought the DB was f*cked! ahahahahaha

If THAT is the problem, then what we need to do is move the machine to a different host or shelve SGO5 till the connectivity issues are fixed.

[Capt_Bearslayer_XC]

Well, that means you need to get in the F-CF and let DH have the F-DNL for a bit.

B/c he can't draft when he hosts....

Unless he has fixed that issue....

[Hexx]

Well, that means you need to get in the F-CF and let DH have the F-DNL for a bit.

B/c he can't draft when he hosts....

Unless he has fixed that issue....

So we won't let him play,
he'll be watching movies anyway, just keep him there to reboot
as needed.

[el-Karnak]

Well, that means you need to get in the F-CF and let DH have the F-DNL for a bit.

B/c he can't draft when he hosts....

Unless he has fixed that issue....

Hey, that's my job.  I'm supposed to draft DH and then we go beat up  Dizzy and Agave, and then they go boom.

Ooops!!  Wrong dyna.

[FA Frey XC]

You know, I REALLY WISH people would USE the numbers I provide them before trying to hypothesize.

Let me clear a couple things up:

1. IN NO WAY, CAN YOU DIG IT, DOES THE BANDWIDTH LIMITATIONS I HAVE effect any of the services. Currently, the Firewall will cap @ 3 Mb...that's two T1 connections. If you really think we've hit that much bandwidth, well, heh, no.

2. TEAMSPEAK is running on a DIFFERENT BOX than THOR. THOR is a DEDICATED CAMPAIGN SERVER. The two don't affect each other in any way.

3. The issue was with MCI , YES THAT'S RIGHT, MCI, not THOR, not the Serverkit, not Teamspeak....

Before you people go off hypothesizing please go to the source. I've published my phone number, but didn't receive a single call this weekend. If anyone wants a technical detail of what happened, then just PM me.

From the looks of it, MCI has their shizit straightened out. We are 100%. If you still want to move the serverkit, that's fine.


P.S. Yes, I am frustrated, but more so with MCI than anyone here. I apologize for the difficulties.

[Capt_Bearslayer_XC]

So when is the server starting back up?

[Hexx]

Thanks for the heads up Frey!

and uhmm.. sorry I didn't call you over the weekend...

[Dizzy]

P.S. Yes, I am frustrated, but more so with MCI than anyone here. I apologize for the difficulties.

Apology accepted, we know its not your fault bout MCI, as such no apology needed... Lesson learned, pick up the phone and call Frey.   We could have nipped in the butt all the speculation about what was going on had we called you. Apparently you wernt aware of the problems we were having and we obviously cant expect you to keep up with all the threads on the forums... So I will surely call next time we start going into two+ days of shizit.

Game on! Ummm... Bonk, what say we go back to the DB we had yesterday when you got there and backed it up?

[FA Frey XC]

Okay, but if MCI decides to shizit all over themselves again....just don't blame me or my servers.

Yea, I had internet issues from Comcast from about Sat on up til present ... they got a tech scheduled for tomorrow...which is why I gave you guys and gals my cell and home phone.

I'm kinda peculiar about that....

[Bonk]

You know, I REALLY WISH people would USE the numbers I provide them before trying to hypothesize.

Let me clear a couple things up:

1. IN NO WAY, CAN YOU DIG IT, DOES THE BANDWIDTH LIMITATIONS I HAVE effect any of the services. Currently, the Firewall will cap @ 3 Mb...that's two T1 connections. If you really think we've hit that much bandwidth, well, heh, no.

2. TEAMSPEAK is running on a DIFFERENT BOX than THOR. THOR is a DEDICATED CAMPAIGN SERVER. The two don't affect each other in any way.

3. The issue was with MCI , YES THAT'S RIGHT, MCI, not THOR, not the Serverkit, not Teamspeak....

Before you people go off hypothesizing please go to the source. I've published my phone number, but didn't receive a single call this weekend. If anyone wants a technical detail of what happened, then just PM me.

From the looks of it, MCI has their shizit straightened out. We are 100%. If you still want to move the serverkit, that's fine.


P.S. Yes, I am frustrated, but more so with MCI than anyone here. I apologize for the difficulties.

<looks sheepish> Sorry dude. I thought the bandwidth limitations applied to the xenocorp servers altogether, not individually. I'll call before I theorise publicly. Again, my apologies.

[KBFLordKrueg]

You know, I REALLY WISH people would USE the numbers I provide them before trying to hypothesize.

Let me clear a couple things up:

1. IN NO WAY, CAN YOU DIG IT, DOES THE BANDWIDTH LIMITATIONS I HAVE effect any of the services. Currently, the Firewall will cap @ 3 Mb...that's two T1 connections. If you really think we've hit that much bandwidth, well, heh, no.

2. TEAMSPEAK is running on a DIFFERENT BOX than THOR. THOR is a DEDICATED CAMPAIGN SERVER. The two don't affect each other in any way.

3. The issue was with MCI , YES THAT'S RIGHT, MCI, not THOR, not the Serverkit, not Teamspeak....

Before you people go off hypothesizing please go to the source. I've published my phone number, but didn't receive a single call this weekend. If anyone wants a technical detail of what happened, then just PM me.

From the looks of it, MCI has their shizit straightened out. We are 100%. If you still want to move the serverkit, that's fine.


P.S. Yes, I am frustrated, but more so with MCI than anyone here. I apologize for the difficulties.

We luv ya, Frey!
It's all good man...No one blamed you.
I think it was all just a bunch of "shooting in the dark" kinda thing because no one could figure it out. 

[Bonk]

So the possibilites are:

1) To move the server to DH's box tonight to eliminate connection as the issue and we'll know pretty quickly if its the db thats bejiggered.

2) Put current db back in and let er rip on thor.

3) Leave it as it to determine if connection issues with MCI have stopped.

What's yer call Dizzy?

[Capt_Bearslayer_XC]

Ummm....  Frey said that they are 100% right now....

[KBFLordKrueg]

So the possibilites are:

1) To move the server to DH's box tonight to eliminate connection as the issue and we'll know pretty quickly if its the db thats bejiggered.

2) Put current db back in and let er rip on thor.

3) Leave it as it to determine if connection issues with MCI have stopped.

What's yer call Dizzy?

I would vote to give Frey's connection another chance... 

[Bonk]

From an MSN converstation I had with Frey a bit ago...

Bearslayer says:
have you been able to help Bonk tract down any of the server issues we have been having?
FA_Frey_XC says:
It's nothing to do with the server
FA_Frey_XC says:
It's been our connection
FA_Frey_XC says:
I keep telling people that
Bearslayer says:
'our' connection?
FA_Frey_XC says:
Work connection
FA_Frey_XC says:
Up here at the office
Bearslayer says:
ah, ok... so basicly, your work is having connectivity issues and when a bunch of d2ers get on it amplifies the problem
FA_Frey_XC says:
No
FA_Frey_XC says:
It doesn't matter who gets on
FA_Frey_XC says:
It had absolutely nothing to do with anyone connecting
FA_Frey_XC says:
It was a BGP routing issue that resulted from MCI's "flapping"
FA_Frey_XC says:
Which was causing routing tables to be re-written
FA_Frey_XC says:
thus, some could connect
FA_Frey_XC says:
while others couldn't
FA_Frey_XC says:
then they could
Bearslayer says:
ah... ok.  Have you posted to that effect on the thread they got going?
FA_Frey_XC says:
but mission matching wouldn't work because once the serverkit matches peeps inna mission
FA_Frey_XC says:
No, I haven't had time
FA_Frey_XC says:
I'll try to tonight
FA_Frey_XC says:
But my home connect has been flakey for the last day and a half 2
FA_Frey_XC says:
When it rains, it pours eh ?
Bearslayer says:
Can I cut and paste what you just told me?
FA_Frey_XC says:
Sure, no one will understand it anywho
FA_Frey_XC says:
I'll make a post there in about an hour
Bearslayer says:
Well, it might help....
FA_Frey_XC says:
Help what ?
Bearslayer says:
b/c they will have a sense of what is going on and stop pulling their hair out over something they have no control over
FA_Frey_XC says:
Who's pulling their hair out ?
Bearslayer says:
Bonk and Dizzy
FA_Frey_XC says:
If anyone had half a clue, then they'd know it's got nothing to do with the serverkit
FA_Frey_XC says:
The serverkit isn't crashing
FA_Frey_XC says:
has it been crashing ?
Bearslayer says:
Well, that is the primary suspect right now
FA_Frey_XC says:
Because if so then that's a different issue
Bearslayer says:
That and using D.net TS at the same time as the server
FA_Frey_XC says:
??
FA_Frey_XC says:
Those are on two seperate servers
FA_Frey_XC says:
LOL
Bearslayer says:
There is a 'test' server up right now to see what is going on
FA_Frey_XC says:
Did someone actually think that could cause a problem ?
Bearslayer says:

FA_Frey_XC says:
"Test" what, did they move it off THOR ?
Bearslayer says:
IE You really need to chime in...

We knew the serverkit was not crashing. I have 3/4 of a clue OK?    

[Hexx]

So the possibilites are:

1) To move the server to DH's box tonight to eliminate connection as the issue and we'll know pretty quickly if its the db thats bejiggered.

2) Put current db back in and let er rip on thor.

3) Leave it as it to determine if connection issues with MCI have stopped.

What's yer call Dizzy?

I would vote to give Frey's connection another chance... 

As acting Alliance RM (and therefore kinda Dizzy's boss) I'd second it.
Try it on Thor . (and advance the year to 75 for lost time)

[FA Frey XC]

Aww guys, it's no big deal

I appreciate your show of support, but I want this campaign to run its course....

We knew the serverkit was not crashing. I have 3/4 of a clue OK?    

Bonk, please accept my public apology, I didn't mean it like this.

I was really more upset at MCI and no one giving me a ring-a-ding, it came out wrong, and again - I apologize and Bonkster, let me tell you, I think you have much more than a clue......

So, whatever you guys decide, if you want to move it just in case, I'm cool with that.

I can't be responsible for MCI, but I can make sure you guys understand I won't take it personal if you decide to move it over to another box..

just in case.



Regards,

Frey

[Bonk]

That's perfectly OK, Frey. Apology accepted. You're entitled to kick me around somewhat considering some of the headaches I've given you over the years...     Anyway, I know you didn't mean it like that, I understand your frustration completely.

Ummm....  Frey said that they are 100% right now....

Roger, we're back up on the db saved at 7:14 CST this morning, still on Thor. So get busy! 

[Capt_Bearslayer_XC]

SWEET!!!!

[Capt_Bearslayer_XC]

That's perfectly OK, Frey. Apology accepted. You're entitled to kick me around somewhat considering some of the headaches I've given you over the years... Anyway, I know you didn't mean it like that, I understand your frustration completely.

Ummm.... Frey said that they are 100% right now....

Roger, we're back up on the db saved at 7:14 CST this morning, still on Thor. So get busy!

Umm... you need to let Frey know where the back up copy is....

[Bonk]

That's perfectly OK, Frey. Apology accepted. You're entitled to kick me around somewhat considering some of the headaches I've given you over the years... Anyway, I know you didn't mean it like that, I understand your frustration completely.

Ummm.... Frey said that they are 100% right now....

Roger, we're back up on the db saved at 7:14 CST this morning, still on Thor. So get busy!

Umm... you need to let Frey know where the back up copy is....

Um, I'm confused, But I'll PM him the location of the db backups...

[FA Frey XC]

No need, I think Bear was just trying to play MSN relay

[el-Karnak]

Do it workie now?

[FPF-DieHard]

Was stable as hell last night

[Bonk]

Coolness! 

I'm tempted to run another SQL load test once its finished, with the # of mission choices per move setting back at 2. To see if that was the problem while on SQL as well...

[762_XC]

NP, just disguise it as a real campaign. 

[Bonk]

  Smartypants... 

[el-Karnak]

We did the network bugged server disguised as a real campaign, so we may as well do the SQL dyna disguised as a campaign too.