[Support] *** An existing connection was forcibly closed by the remote host.

[RadoX 0]

I hate this problem more than anything.

Do you not get this message?

I know i´ve posted this before but this seems to be ignored by DC++ cos thay can´t find wuy this is happening. Older clients don´t have this problem. Alot of ppl are getting this error.

Client needs to have better connection to the hub, is there no way anybody could fix this.

[Lee 84]

Moved here Radox..

[Sidetrack 0]

This is getting to be a very big problem, and believe it or not, its NOT on your end probably. As a hub owner, I know that we tweak the hubsofts as much as we can in order to be able to have as many users as possible, using YNHUB, this problem has become even more evident lately so I noticed with my hub. I have about 630 K up in bandwidth.. and I push over 800 usrs, most of the time actually 900. I have gotten to know the software very well and can usually do very well at it. but there are times when it simply decides to not play nice. They say (they being ynhub forums) that for every 1 user, you should allocate 20k of bandwidth.. you do the math.. its not very much at all.. However, with tweaking.. you can bring it down to less than 1k. As long as everything is working right. its works flawlessly, but like me at the moment, is having a problem on our DSL line and a ton of line noise.. so what would normally work well, is failing miserably. (got a tech coming out today, probably remote terminal screwing up again). My point in this is that the hub owners are pushing the connection to hard, or something is screwed up. On a side note, but also a direct thing affecting this.. Many ppl are now using ynhub 1.33, and socket errors are far worse than they were using 1.31, but the additional commands and functionality that was added made running a hub far more efficient. so as a hub owner, we have to choose, a sacrifice for quality over accessibility. its a very hard decision. As a support hub owner, some commands simply will not work under 1.31. so we have no choice but to run 1.33. hope this helps

[RadoX 0]

Well if this is just a hub problem then wuy do older clients run better than the new?

I´m running a verlihub on 24shells hosting so the line should be fine but maybe the problem is that it´s hosted in USA not in Iceland where I live. But this happens in icelandic hubs too.

I have been trying to find some reason for this and all I can find is that the newer clients seems to be more fragile than the older ones, there are always some problems with noise or something not perfect in the line but the fact is that this did not happen so often with the older clients. I am on 8mb/832kb adsl2 line so the line speed is not the problem. I have seen this error when running antivirus scan and when my computer is running something heavy. But this does not happen all the time so I can´t find anything specific wrond other than it started to happen on the new clients.

Somehow I doubt that this will ever be fixed

P.S. This error came from DC++, this is not a specific ApexDC error.

[Zlobomir 3]

Maybe newer clients get fragile when switched to new connection display?