[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Help-gnunet] Not better since patch 0.4.6c :-(
From: |
Christian Drechsler |
Subject: |
Re: [Help-gnunet] Not better since patch 0.4.6c :-( |
Date: |
Fri, 6 Sep 2002 19:17:22 +0200 (CEST) |
hi!
On Fri, 6 Sep 2002, Christian Muellner wrote:
> As in http://www.cryss.net/gnunet/stats/gnunet-stats.html
> my stats since yesterday are not better.
> NO recievied querys ans NO octets send.
strange. a sample from here:
address@hidden GNUnet-0.4.6]$ gnunet-stats
Node-to-Node (udp):
#HELO received: 50 send: 115
#SKEY received: 14 send: 73
#PING received: 17 send: 26
#PONG received: 0 send: 17
#3QUERY received: 166031 send: 71774
#CONTENT received: 4351 send: 3054
#TIMESTAMP received: 0 send: 0
#SEQUENCE received: 13288 send: 4474
#NOISE received: 5627k send: 3064k
#HANGUP received: 0 send: 0
#octets received: 21248k send: 6517k
Server-Client (tcp):
#QUERY received: 5668 send: 0
#CONTENT received: 0 send: 3529
#INSERT received: 0 send: 0
#INDEX received: 0 send: 0
#LISTFILE received: 0 send: 0
#FILEINDEX received: 0 send: 0
#STATREQ. received: 1 send: 0
#STATREPLY received: 0 send: 0
Server Statistics:
Shared files : 744
Size of shared data: 223596k
Connected hosts : 7
Uptime : 817s
ok, i have to add that i'm downloading a ~140MB file at the moment, so
there is much traffic, anyway. but compare tcp-#QUERIES to udp-#3QUERIES.
i also have to add that this was a kill and start again; there's not
always 7 hosts after 817 sec. normal here is 5-15 hosts, 19 being the
highest i saw - same as christian grothoff.
btw, i just patched from 0.4.6 to 0.4.6c. wow! *smile* ok, there are no
longer any upstream values of 171 K. X-) now most of the time it's around
12 K of 16 K theoretical bandwidth. that's nice, i can even use the
computer for other net stuff. ;-)
those high values ... i think it's like network overclocking. ;-) it
actually seems to have worked - but i certainly don't know if those udp
packages ever reached their destination. anyway, the link became very
unstable and reconnected every 20-40 minutes. so it's for sure better this
way. and: the download speed is _way_ bigger now than it was before. i
can't give sensible figures now as the download started with the old
gnunetd and had big pauses (reconnects) in between. but it seems as if the
speed is on an average of at least 4000 bps, maybe more. great! ;-)
regards, zottel