hi all
I've been recently getting a large number of disconnects from ETG and am trying to narrow down the cause. I've employed packet sniffers both directly on the machine running mIRC and at the edge-most location on my local network (outside of my router/firewall, between it & my PPPoE DSL modem).
My understanding of the IRC protocols is somewhat limited, but I'm learning. I think I have a decent understanding of TCP/IP, ethernet, etc...
With mIRC in the recent weeks I've had 20+ disconnects/day, although I've also had some connection problems with other applications & games... just not as frequent. Via the sniffer, these disconnects show up as a CLIENT initiated TCP FIN/ACK packet. These finish packets are exactly 2 minutes after mIRC sent a PING: TIMEOUTCHECK, but the server failed to return a PONG message in time. However, the server DID send a TCP ACK to the original ping, so I know at that level, it got it...
So I'm wondering if that PING: TIMEOUTCHECK packet is from the new feature in 6.17 called "check for timed out connections". So in my case, the ETG servers aren't responding to this request or their response is being eaten by my ISP. Also, if this new ping/pong (in addition to the usual SERVER GENERATED ping/pong) is just giving mIRC another reason to close the connection.. it might just hurt more than it helps! I don't know if I'd enabled this option at some point in the past or if it's defaulted to ON.
I also get the more typical [10053] Software caused connection abort errors also, so maybe my DSL provider just sucks...
