Servers keep dropping off of Master List

[a girl]

I'm not sure if this is a Minit only problem or if other people are experiencing this issue... but anyway, any help will be much appreciated.

Ever since we switched over to the new master list, our (Minit) servers tend to drop off of the master list... randomly. It doesn't leave an error code in the logs and it doesn't display any such "can not connect to master server" error. When this happens, I've verified that the master broadcast is on... but the servers just aren't being listed. To fix this problem, usually toggling the master broadcast off then back on again fixes the problem (ie !master !master). However, this seems to be happening to our Assault server at least twice a day. An experimental work around that RainbowDash came up with is a script that auto cycles the master broadcast every 30 minutes... although this seems to be somewhat effective, the servers still drop on occasion.

If this is a Minit only problem, then it must be a script or some pyspades file that we have changed on our end that is causing the problem... another possibility that we just ruled out today was that our mini server list that Topo created for us by grabbing the data off of our server list pages might be trying to update at the exact same time and possibly breaking something... but as we just disabled this today and it didn't solve the problem, I'm ruling it out as the culprit. Any thoughts or ideas? I might have to download a fresh copy of master.py or something.

[TEFLON]

This happens all the time to me

[MKU]

I think it has something to do with the fact that Minit is using an older version of the pyspades branch when it should be on the pysnip branch. I admin a server named SuperCool Build and it does this all the time.

[TEFLON]

Im using pysnip latest and it does this everyday..

I'm too tired of restarting it since i don't even play this game...

once its fixed let me know

[topo]

I think it has something to do with the fact that Minit is using an older version of the pyspades branch when it should be on the pysnip branch. I admin a server named SuperCool Build and it does this all the time.

Actually, that's incorrect. Minit uses its own branch of pysnip commit a808462ccd67.

[TEFLON]

The problem looks like its fixed for me

[Sasquatch]

I had this issue and I talked to the guys at Ramnode and they said for whatever the server here requires something to be sent back to assume you are active. Make a file and call it .screenrc and put these two lines of code in it.

Code: Select all

hardstatus alwayslastline
hardstatus string '%{= kG}[ %{G}%H %{g}][%= %{= kw}%?%-Lw%?%{r}(%{W}%n*%f%t%?(%u)%?%{r})%{w}%?%+Lw%?%?%= %{g}][%{B} %m-%d %{W}%c:%s %{g}]'

Then go pack into putty and run your server. It will look a little different because it will keep a clock running when sends data back. idk how or why it works but it does, my server hasn't shut off since using this.

[StackOverflow]

I have been tweaking things with the master server for the last few weeks to help resolve the issue.

According to Enari, topo had found a problem with either onectf/babel that was resulting in an error, which was the cause for the Minit server(s) dropping off the last. I'm not sure what exactly the issue was.

For anybody else having any issues, some server information would be useful, along with any error logs associated with the master server.

[topo]

The babel+onectf error caused babel to basically freeze, so it's no wonder the master connection wasn't being maintained. That was local to one server and seems to be resolved now. I haven't noticed any of our servers drop off lately. \o/

[Sasquatch]

I take it back, mine dropped of again =/

[StackOverflow]

I take it back, mine dropped of again =/

Can you give some details pertaining to server setup (name, country, gamemode, port, etc). Also, what is your network setup, ie are there multiple servers being hosted on the same connection?

[Sasquatch]

I take it back, mine dropped of again =/

Can you give some details pertaining to server setup (name, country, gamemode, port, etc). Also, what is your network setup, ie are there multiple servers being hosted on the same connection?

Well, I run my server on RamNode, configure it through Putty. It's on a linux system. (Ubuntu 11.4 I think.) Usually it's just named Sasquatch 24/7, it runs CTF only. It doesn't have anything to do with game mode incompatibility I don't think, because it's stayed up for days at a time and cycled through all the maps. The server is in Atlanta, GA. Link to it in my sig (I'll start it up again in a second.)

[Sasquatch]

Strangely enough, when I re-opened putty, my server shows the last thing to happen in the server, and then Killed, which I'm assuming means the server connection was killed.

Code: Select all

2013-02-03 07:58:17+0400 [-] megatron3000 disconnected!
Killed

[Handles]

I'm not sure why it does this. It's never happened to me

[TEFLON]

its happening again, no errors in console..

im tired of having to check up on it