ViolentCrimes

I'd say it's getting there, but not ready, for Debian at least. We run 2.0.18.0 on a single Debian box with a couple of TF2 services. We've spent a lot of time upgrading, downgrading, diagnosing, dealing with services disconnected from TCADMIN, services doubling up, etc. 2.0.22.0 seems a lot better in that regard, but it's a bit early to tell. We're waiting for a TF2 lockup issue to get resolved (might be TCADMIN, might be TF2) before we can run it.

I wouldn't use mani if you paid me for one. I know it is not the gameserver itself issue when ran by command line and not tcadmin no crashes. I reverted tcadmin to the older version and now no crashes aswell. TF2 - Linux.xml

TCAdmin Version: 2.0.22 Operating System: Debian Squeeze, 64 bit, Stock Kernel customized for 1000Hz Machine Specs: Dual X5472 @ 3.00GHz, 8GB Database Type: mysql Mono: 2.6.7 Linux 64 Bits (Debian Squeeze default package) Just upgraded to 2.0.22.0. When I start TF2 servers from tcadmin, they lock up within a couple of minutes after starting; Processes are not terminated, TCADMIN will say it's online, but querying won't work, and players can't connect. Empty servers don't seem to lock up. Nothing in the tf2 log files. As a test: - Starting a TF2 server from the commandline as test, doesn't lock up (tested for an hour). - Did a full clean restart of all the services, still a lockup. - Reverted back to 2.0.18.0, servers no longer lock up. Any ideas?

This issue was fixed for us after upgrading from 2.0.18.0 to 2.0.22.0.

Upgrading to 2.0.22 solved it, without a need for a mono update. Thanks!

I'll test tomorrow.

I have been using Debian as the master for 2-3 months with the beta.

Anything yet?

Has anyone figured out how to query this game yet?

It's a production machine with other mono apps on it; I prefer to stick with the Debian squeeze stable packages for mono. I could set up a vmware image with my setup and 2.0.20, see if it crashes at all or if it's specific to the production machine, but it'll be a while before I have the time. Any other Linux testers that have run into this issue or am I the only one? Do you run any tests on Linux before releasing? On what kind of platform?

TCAdmin Version: 2.0.19 and 2.0.20 Operating System: Debian Squeeze, 64 bit, Stock Kernel customized for 1000Hz Machine Specs: Dual X5472 @ 3.00GHz, 8GB Database Type: mysql Mono: 2.6.7 Linux 64 Bits (Debian Squeeze default package) == Observed: After starting monitor-console or monitor-service, it boots and crashes (10% of the time) or coredumps mono (90% of the time) right after startup. It will not start properly. Steps to reproduce: Upgrade old 2.0.18 version (not the later released 2.0.18_1 or whatever update) to 2.0.19. Run monitor. I've tried all combinations of upgrading via web or via command line tool. Upgrading to 2.0.19 or 2.0.20. Repeating command line tool upgrade, etc. See attachment for coredump stacktrace and mono stacktrace. I've rolled back to 2.0.18 for now. Really want to escape that version. D: tcadmin2-2.0.19-monitor-crash-mono-coredump-stacktrace.txt tcadmin2-2.0.19-monitor-crash-stacktrace.txt

Will us beta testers who have been testing and reporting bugs be getting a discount at all for v2?

http://clientforums.tcadmin.com/showthread.php?t=8012

TCAdmin Version: 2.0.18.0 Operating System: Debian Squeeze, 64 bit, Stock Kernel customized for 1000Hz Machine Specs: Dual X5472 @ 3.00GHz, 8GB Database Type: mysql Found TCAdmin filled up a partition with 15GB Monitor logs containing the following over and over: 05/15/2011 11:39:05 System System.Net.Sockets.SocketException: interrupted at System.Net.Sockets.Socket+SocketAsyncResult.CheckIfThrowDelayedException () [0x00000] in <filename unknown>:0 at System.Net.Sockets.Socket.EndReceive (IAsyncResult asyncResult, System.Net.Sockets.SocketError& errorCode) [0x00000] in <filename unknown>:0 at System.Net.Sockets.Socket.EndReceive (IAsyncResult result) [0x00000] in <filename unknown>:0 at CleverComponents.InetSuite.CommandConnection.EndRead (IAsyncResult ar) [0x00000] in <filename unknown>:0 at CleverComponents.InetSuite.?.. (IAsyncResult .) [0x00000] in <filename unknown>:0 Process: violent 31324 76.6 1.8 781960 150892 ? Sl May14 1407:56 /usr/bin/mono /usr/lib/mono/2.0/mono-service.exe -d:/home/violent/tca2beta/Monitor /home/violent/tca2beta/Monitor/TCAdminMonitor.exe

For some reason it doesn't stop/start the server when you run a steam update.

What Game Compatibility does it use.

I use 8% for a full 16 person server. Dual Quadcore 2.33

does it use the .pid files in monitor/services? I noticed those contained a pid of 0 (only looked after restarting the service manager, so no idea what set it to 0).

Picking the right query protocol solved the issue. thanks.

For our own personal goals 32 characters would suffice to uniquely describe a server.

What do you mean? Like a copy from /var/log/daemon? Apr 18 14:56:13 banhammer mdadm[1160]: NewArray event detected on md device /dev/md0 Apr 18 14:56:13 banhammer mdadm[1160]: NewArray event detected on md device /dev/md1 Apr 18 14:56:14 banhammer acpid: starting up with netlink and the input layer Apr 18 14:56:14 banhammer acpid: 1 rule loaded Apr 18 14:56:14 banhammer acpid: waiting for events: event logging is off Apr 18 14:56:14 banhammer dhclient: Internet Systems Consortium DHCP Client 4.1.1-P1 Apr 18 14:56:14 banhammer dhclient: Copyright 2004-2010 Internet Systems Consortium. It solves the issue by listing service as a column. Right now I could just combine and sort logs with scripts after the fact, it's just a convenience to be able to make a live tail and grep.

TCAdmin Version: 2.0.18.0 Operating System: Debian Squeeze, 64 bit, Stock Kernel customized for 1000Hz Machine Specs: Dual X5472 @ 3.00GHz, 8GB Database Type: mysql Sometimes serviceman-service just stops (without any error logs as far as I can see, unless they are stored /outside/ the serviceman log dir). The game services continue running, but after restarting serviceman-service it starts all the game services another time, doubling them up. I seem to remember something about serviceman being changed in 2.0.19.0, but I no longer seem to see that version in the auto update.

so it's a feature? Is there a roadmap for all completed and uncompleted features?

TCAdmin Version: 2.0.18.0 Operating System: Debian Squeeze, 64 bit, Stock Kernel customized for 1000Hz Machine Specs: Dual X5472 @ 3.00GHz, 8GB Database Type: mysql For some reason TCADMIN opens up data sockets outside the specified port range. TCadmin2 is set up to ftp control channel 8821, with passive Ports: 5000-5100. I've seen tcadmin opening ports ranging from 30000 to 58000. netstat: tcp 0 0 serverip:8821 clientip:60385 ESTABLISHED 9263/mono tcp 0 0 serverip:8821 clientip:60371 ESTABLISHED 9263/mono tcp 0 0 serverip:47960 clientip:60386 ESTABLISHED 9263/mono I'm not using any ftp connection tracking modules.

Seems a non-issue, given how daemon.log, user.log, dmesg or any other system log deals with it. I guess I can combine it myself with some scripts.