Re: [hlds_linux] Source Engine Update Available
I'm actually seeing the same thing happening out of our 3140 servers as well, with the same errors, and they do not seem to reconnect to VAC. I was just about to throw in a report myself until I saw this thread. Popping open tcpdump didn't seem to show any communications attempted to anything identifying itself as a Valve server upon startup.A bit later on it did, but nothing that brought up anything of consequence. Jason, Alfred - any ideas? ics wrote: 3 of our servers are also VACless so i take it that VAC servers are once again down. If VAC goes down, servers should automatically connect back to it when its up but maybe there should be a manual command which admins could use to connect to vac servers. How frequently the servers are trying to connect now? version : 1.0.0.34/7 3140 insecure (secure mode enabled, disconnected from Steam3) We also get that same error in console time to time which is netconnectionbuffer.cpp (241) : Assertion Failed: cubReadRemaining == 0 -ics ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Uniform test structure? (was Re: [hlds_linux] Very high CPU usage since HLDS Engine upgrade 35)
Heya all, I participated in the swapbeta test that Alfred had going during Valve's move from an IPC architecture to a event-driven architecture, in hopes of making everything perform unilaterally and cross-platform between the hlds_l and hlds_win32 binaries. Unfortunatly, my test set was quite hasty, and we did not test all the possible configurations of hlds_i386 on our systems, the majority of which being dual Intel Xeon 3.0ghz w/ HT, 2GB RAM and SCSI HDDs. Since the release of Engine v35, we've had enough reports of 'registry' problems that I was forced to make a backup 'last known good' copy of hlds, just in case we had anymore trouble. These troubles, again, seem to revolve around our pingboosted 32-bit setups, as those without pingboost seem to have no trouble with the newest release. How about instead of posting a set of specs, we go one step further and begin to propose ways to effectively prod the binaries, from not just the server-side perspective but the player-side as well, collect reports from players on these servers, get as much data as possible together to bundle up and send to Alfred, for him to sift through and see if he can find any correlations? DISCUSS! Or I'll just force a testing structure idea on everyone that I've been throwing around. :) ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] choke problem
I just personally got done doing a good job at messing up a fresh copy of Debian Sarge w/ 2.6.8 by trying to upgrade it to the latest testing distribution, Etch, which runs 2.6.15 by default, if I recall correctly. I can't honestly tell you what's going on, because I'm personally not running Etch myself. But if I were me, I'd nuke the install, get a Sarge ISO, and install that instead. (Looks like 2.6.8-2-686 that's available in the repository for Sarge is KERNEL_HZ = 1000) [EMAIL PROTECTED] wrote: Hi again: My machine specs are: Linux Debian kernel 2.6.15 P4 extreme edition 3,4 2GB RAM Connnection type: T3 I dont think its a rate problem because all is ok until it reaches 15 users or more.And i have more machine with same specs and dont get choke. Thanks a lot __ LLama Gratis a cualquier PC del Mundo. Llamadas a fijos y móviles desde 1 céntimo por minuto. http://es.voice.yahoo.com ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
