Thread: Occasional Drop in connection speed + high latency/some packet loss

Today (8 july) right after 2:10am speeds suddenly dropped.

Latencies to USA went up from 280-290 to about 480-510 accompanied with heavy packet loss.

Code:

Reply from 216.239.122.178: bytes=32 time=494ms TTL=244
Reply from 216.239.122.178: bytes=32 time=488ms TTL=244
Reply from 216.239.122.178: bytes=32 time=492ms TTL=244
Reply from 216.239.122.178: bytes=32 time=489ms TTL=244
Reply from 216.239.122.178: bytes=32 time=488ms TTL=244
Reply from 216.239.122.178: bytes=32 time=489ms TTL=244

normal time=280ms

SNR up/down :14 dB /21.5 dB
Atten up/down: 7.0 dB /14.0 dB

using opendns
bridged mode (cannot switch to always on since other ppl use the PC during the day)

plan: BSNL Home 1000



usually speed of 2.5 mbps, latency 280-300 here too

o would just like to add that this problem seems to happen more often during the night, but it also happens during the evening, early afternoon and morning too; it's not really dependant on peak usage or anything. Unless congestion is random ...

and also at times when this happens reconnecting to bsnl multiple times gives a better routing with normal latency, but then again this doesn't always work.

ping to another US server at around 5:16am:

Code:

reply from 63.241.193.18: bytes=32 time=510ms TTL=241
reply from 63.241.193.18: bytes=32 time=507ms TTL=241
reply from 63.241.193.18: bytes=32 time=510ms TTL=241
reply from 63.241.193.18: bytes=32 time=509ms TTL=241
request timed out.
request timed out.
request timed out.
request timed out.
request timed out.
request timed out.
request timed out.
request timed out.
request timed out.
request timed out.
request timed out.
reply from 63.241.193.18: bytes=32 time=288ms TTL=239
reply from 63.241.193.18: bytes=32 time=288ms TTL=239
reply from 63.241.193.18: bytes=32 time=288ms TTL=239
reply from 63.241.193.18: bytes=32 time=287ms TTL=239
reply from 63.241.193.18: bytes=32 time=297ms TTL=239
reply from 63.241.193.18: bytes=32 time=291ms TTL=242
reply from 63.241.193.18: bytes=32 time=289ms TTL=242
reply from 63.241.193.18: bytes=32 time=295ms TTL=242
reply from 63.241.193.18: bytes=32 time=292ms TTL=241
reply from 63.241.193.18: bytes=32 time=295ms TTL=241
reply from 63.241.193.18: bytes=32 time=291ms TTL=241
reply from 63.241.193.18: bytes=32 time=290ms TTL=242
reply from 63.241.193.18: bytes=32 time=290ms TTL=242
reply from 63.241.193.18: bytes=32 time=290ms TTL=242
request timed out.
request timed out.
reply from 63.241.193.18: bytes=32 time=286ms TTL=244
reply from 63.241.193.18: bytes=32 time=280ms TTL=244
reply from 63.241.193.18: bytes=32 time=277ms TTL=244
reply from 63.241.193.18: bytes=32 time=280ms TTL=244
reply from 63.241.193.18: bytes=32 time=282ms TTL=244

screenshot from d3tr (a 3d traceroute tool; does a traceroute to a server over a period of time and builds a graph)[ 3d Traceroute ]

here it is from about 10 minutes ago:

as you can see the routing is horribly screwed up, with latency fluctuating wildly every 5-10 minutes settling between the 280-310 ms range and then to the unusable 500+ ms range.
I also did a ping using colasoft ping tool (it gives a graph of pings to diff websites over time) - www.colasoft.com/ping_tool



here's another ping from a minute ago :

Code:

Reply from 63.241.193.18: bytes=32 time=299ms TTL=238
Reply from 63.241.193.18: bytes=32 time=300ms TTL=238
Request timed out.
Reply from 63.241.193.18: bytes=32 time=299ms TTL=238
Reply from 63.241.193.18: bytes=32 time=299ms TTL=238
Reply from 63.241.193.18: bytes=32 time=304ms TTL=238
Reply from 63.241.193.18: bytes=32 time=305ms TTL=238
Reply from 63.241.193.18: bytes=32 time=303ms TTL=238
Request timed out.
Request timed out.
Reply from 63.241.193.18: bytes=32 time=304ms TTL=238
Reply from 63.241.193.18: bytes=32 time=302ms TTL=238
Reply from 63.241.193.18: bytes=32 time=305ms TTL=238
Reply from 63.241.193.18: bytes=32 time=302ms TTL=238
Reply from 63.241.193.18: bytes=32 time=301ms TTL=238
Reply from 63.241.193.18: bytes=32 time=300ms TTL=238
Request timed out.
Reply from 63.241.193.18: bytes=32 time=303ms TTL=238
Reply from 63.241.193.18: bytes=32 time=302ms TTL=238
Reply from 63.241.193.18: bytes=32 time=301ms TTL=238
Reply from 63.241.193.18: bytes=32 time=301ms TTL=238
Reply from 63.241.193.18: bytes=32 time=302ms TTL=238
Reply from 63.241.193.18: bytes=32 time=299ms TTL=238
Reply from 63.241.193.18: bytes=32 time=490ms TTL=238
Request timed out.
Reply from 63.241.193.18: bytes=32 time=303ms TTL=238
Reply from 63.241.193.18: bytes=32 time=302ms TTL=238

WHAT THE HELL IS BSNL DOING ?

i wanted to download the latest patch for starcraft 2 beta but it's proving to be impossible.

cheap/bad routing ? some kind of maintainance ? random network congestion for unknown reasons ? some kind of traffic shaping ? throttling ?

Exactly same problem, again, today (July 11th); right after 2am.

490+ ms latencies (from the usual of 270ms)
occasional packet loss.

does know one else experience high latency/slow speed/packet drops ?

wow man,such a detailed analysis,you really know what you are doing.i am also facing the same problems since yesterday afternoon (panipat,karnal area,haryana circle).only thing i can do is sleep off the anger and frustration.

Originally Posted by iorhld

sleep off the anger and frustration

haha that's what i do too.