Michael Rogers wrote:
> Another factor might be vendor selection: my peer quickly reaches a
> state where 60% of its neighbours are either LimeWire or BearShare,
> at which point around half of the new connections it makes will be
> dropped straight away because of the vendor. Although I like this
> feature, it must be causing quite a lot of load on the network.

It's pointless. Something like 80-90% would be alright but I simply
recommend to disable it i.e., allow 100%.
 
> Use surplus bandwidth = yes
> Prefer compressed connections = yes
> Use IP Type of Service = no
> General incoming traffic = 5 KiB/s
> General outgoing traffic = 5 KiB/s
> Ultrapeer mode incoming traffic from leaves = 5 KiB/s
> Ultrapeer mode outgoing traffic to leaves = 5 KiB/s
> HTTP cumulative download rate = 245 KiB/s
> HTTP cumulative upload rate = 20 KiB/s

Configuring the bandwidth limits is just bad. There should be a total
limit and that's it especially with "use surplus" all these detailed
settings are rather useless. I wouldn't configure anything but the
cumulative rates and leave the rest undefined.

> (By the way it's always puzzled me why the total limit doesn't seem
> to include the ultrapeer mode leaf traffic. Another thing that
> puzzles me: when I specify "Try to keep at least 20 and allow at
> most 35 total connections up", I can see 66 active ultrapeer
> connections (28 black, which I assume means outgoing, and 38 grey).

This is supposed to be fixed in 0.96. I don't know why you are seeing
this. However, grey means leaf and black means ultrapeer.  I would
think you have 28 ultrapeers and 38 leaves which means the configured
limit is respected.

> With my current settings (at least 20 and at most 35 connections,
> use 4 connections to connect more quickly),

The default of 40 for quick connections is really over the top. 10 to
20 should really be sufficient. As you say it just provokes timeouts.
One reason why some might consider such a high number necessary is the
default vendor limit which makes it incredibly hard to get enough
connections whereas it works in an instant without. And yes the
handshaking takes a lot of resources and traffic compared to stable
connections.

> gnet traffic is at about 2 KiB/s in and 1 KiB/s out, and leaf
> traffic is at about 0.2 KiB/s in and 1 KiB/s out. HTTP is using the
> rest. The message queues are pretty much empty. If I increase the
> settings to at least 32, at most 40 connections then sooner or later
> it gets into another storm

32 is really nominal, you don't need more or less. The storm you are
seeing might be caused by UDP traffic especially the vast and ever
increasing amount of spam.

-- 
Christian

Attachment: pgpWXnB5KxlV7.pgp
Description: PGP signature

Reply via email to