bad packet?

[mikepeck]

[Jul 11 21:14:16 UTC] FlushRC5::Bad packet acknowledgement. (0.168)
This is the error I get when I try to flush my work. I have about 5100 work units to flush but no luck yet. I am attempting to flush to Dnet, I still have never been able to use the teamanandtech proxy. Not sure what's happening there. Any advice/suggestions you have will be a great help. Thanks.

 

[Batti]

Are you behind a firewall, by any chance?

 

[Russ]

mikepeck,

I get that occasionally when I flush my off line machines. I don't know what causes it, but I just start the flush over and it works out fine.

Russ, NCNE

 

[mikepeck]

Not behind a firewall, but using a proxy server for my 3 home machines. I have flushed before to Dnet, but never to teamanadtech. It has never worked. Here is the problem... I have the files stored on my athlon machine and just run the program over the network on the other machines (a PII 333 laptop and PII 300 desktop.) The PII 300 desktop is the machine that has the net connection and the proxy software is running on. The reason I keep the dnet prog on my athlon machine is because it is my machine and is running 24/7. I was able to flush to dnet if I flush from the machine that has the net connection and it worked this time for some reason. But I do not really want to do this everytime that I flush..
Do you think it is possible to have the Dnet prog on my Athlon machine and use the proxy to flush to teamanandtech (I kinda like the stats idea). If so, here is the current setup if it would help. I connect to the proxy using the PII's ip and port 6588. It seems to flush fine using the port 2064 of Dnet's, so I don't see my teamanandtech would be much different. Is there an IP address for teamanandtech's proxy, that might help.

Thanks for your help.. and i can't believe how fast people respond in this forum.. You guys must have no lives, or just get paid to surf the web.. Kinda like i do...

 

[ViRGE]

Mike, try using jator.2y.net as your keyserver. It listens on port 6588.

 

[mikepeck]

Virge,
Here is the response I get when I use Jators (tried on both ports..)[Jul 11 21:52:17 UTC] Input buffers are full (or projects are closed).
No fetch required.

Haven't tried flushing to it yet, I dont' have anything to flush (I just took a big dump at Dnet)

 

[ViRGE]

Give it another shot in a bit. As long as you have nothing to fetch or flush, there's no way to test it.

 

[mikepeck]

I do have something to fetch though. I can fetch from Dnet just fine.. Just no where else

 

[ViRGE]

Humm, that one has me honestly stumped then. I'll see if I can find a solution, but in the mean time, just be glad you can contact Dnet.

 

[Ken g6]

Let me see if I've got this straight: you have no problem flushing from your computers to your own PProxy, right? The problem is when you try to flush your PProxy to Mika's, right? Then you probably need to ask somebody who has a PProxy, like BPhantom or Jator, how to fix this.

 

[mindless]

This probably isnt going to help one bit, but my pproxy flushes through Mika just fine. But my setup is a bit different, both my machines run rc5 off their local drives and then flush to my pproxy on the machine that is connected to the net.

In case thats not clear. Ive got a celeron and P3, the Celly has the connection to the net and runs the pproxy and RC5 from its own drive, my P3 runs its own copy of RC5 from its own drive and flushes to my pproxy on the celly. The pproxy then flushes to Mika's on 2064, or whatever the default port is.

 

[Jal]

I get a whack of bad blocks occasionally.. For me , I find that if I process random blocks, all the clients dumping to the same buff-out file. I figure it must be colliding with other clients when they process 1 block at a time, it is easy to find another client saving results at the same time. (I think I lost 15K blocks over the weekend because of it)