PDA

View Full Version : Web browsing via 56K modem faster than 1.5Mb broadband



satimis
02-06-2006, 08:07 AM
Hi folks,

Knoppix LiveCD 4.0.2
1.5 Mb ADSL broadband via modem

I have been spendig couple days to find out the cause of slow web browsing on 1.5Mb broadband connection without result. It took prolonged time to connect a website. The strange thing was web browsing via 56K modem connection worked even better, no need to wait.

I changed the network card. ISP even sent his technical staff for site visit checking the cable without any discovery. On the same PC I have other 2 HDs mounted on mobile rack, running WinXP and FC3 respectively. Comparing all of them, web browsing on WinXP worked fastest, not necessary waiting for connecting after pressing <RETURN> and FC3 also not bad, no need waiting for prolonged time to connect. Knoppix LiveCD was the slowest on connecting. I have to sit and wait.

All connection were tested with Firefox as browser.

I tested Knoppix LiveCD on another PC. The same result was found. 56K modem connection performed better than 1.5Mb broadband connection.

Kindly advise. TIA

B.R.
satimis

Harry Kuhman
02-06-2006, 05:02 PM
Knoppix LiveCD 4.0.2
1.5 Mb ADSL broadband via modem
..... I have other 2 HDs mounted on mobile rack, running WinXP and FC3 respectively. ....

If I understand you correctly you are talking about one computer booting any of 3 different OS, and not 3 different computers.

Although Knoppix will be slower tp start firefox since reading a cd is slower than a hard disk, browsing should not be slower once the browser is loaded. I have 3 thought to share on this.
1. read my generic answer #4 (http://www.knoppix.net/wiki/User:Harry_Kuhman).

2. If your system is extremely memory starved it might be thrashing itself to death while running Firefox. You gave no cpu or memory information, so we can't help ypu. (Even when you gave info, like telling us you had a DSL modem, you omitted the make and model, It's almost always better to give as much information as you can in these cases.)

3. If it's not a memory and swapping issue then this sounds to me like a DNS server issue. How have you configured Knoppix without a router? Are you sure the DNS server is right? Try pinging sites by URL and by IP address and see if there is an extreme difference in how long it akes for you to get a response whn using URLs (not how long the ping is, but if it actually takes much longer with a stopwatch to get a response).

satimis
02-07-2006, 12:35 PM
Hi Harry,

Tks for your advice.

Knoppix LiveCD 4.0.2
1.5 Mb ADSL broadband via modem


If I understand you correctly you are talking about one computer booting any of 3 different OS, and not 3 different computers.No, on the same PC. WinXP and FC3 are running on separate HD.

I have my problem partitially solved with following step;
One of the nameserver assigned died.

# cat /etc/resolv.conf
nameserver 202.123.77.196
nameserver 202.123.77.213


# ping -c 3 202.123.77.196
PING 202.123.77.196 (202.123.77.196): 56 data bytes

--- 202.123.77.196 ping statistics ---
3 packets transmitted, 0 packets received, 100% packet loss
# ping -c 3 202.123.77.213

PING 202.123.77.213 (202.123.77.213): 56 data bytes
64 bytes from 202.123.77.213: icmp_seq=0 ttl=252 time=17.8 ms
64 bytes from 202.123.77.213: icmp_seq=1 ttl=252 time=18.2 ms
64 bytes from 202.123.77.213: icmp_seq=2 ttl=252 time=18.2 ms

--- 202.123.77.213 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 17.8/18.0/18.2 ms

After reverting their order, web browsing was much faster than before.


However each time booting up Knoppix LiveCD 4.0.2, /etc/resolv.conf is an empty file with no entry.

# cat /etc/resolv.conf

# Insert nameserver here
# nameserver 127.0.0.1


After setting up ADSL broadband connection, following nameservers are assigned automatically;
nameserver 202.123.77.196
nameserver 202.123.77.213

But "nameserver 202.123.77.196" is not active. Each time I have to revert their order.

How to get it corrected? ISP only provides dynamic IP. DNS server is in ISP's office.

Regarding disable ipv6 in the kernel, I have to run
echo "alias net-pf-10 off" >> /etc/modprobe.conf
echo "alias ipv6 off" >> /etc/modprobe.conf

and reboot Knoppix LiveCD afterwards to take effect. But once reboot all entries disappear. Is there any suggestion? Tks.

Other informations are as follows;

CPU AMD Athlon 1.4
RAM 512MB
ADSL Modem NEC, Model ATU R-110, MAC Add 00-01-38-02 -09 -3C


How have you configured Knoppix without a router? As follows;
Connect PC to ADSL modem and run;

Kstart --> Systems --> ADSL/PPPOE configuration
then following its instruction.


Try pinging sites by URL and by IP address and see if there is an extreme difference in how long it akes for you to get a response whn using URLs (not how long the ping is, but if it actually takes much longer with a stopwatch to get a response).[/list]
# ping -c 8 64.233.187.99
PING 64.233.187.99 (64.233.187.99): 56 data bytes
64 bytes from 64.233.187.99: icmp_seq=0 ttl=239 time=310.2 ms
64 bytes from 64.233.187.99: icmp_seq=1 ttl=239 time=316.2 ms
64 bytes from 64.233.187.99: icmp_seq=2 ttl=239 time=301.1 ms
64 bytes from 64.233.187.99: icmp_seq=3 ttl=239 time=315.4 ms
64 bytes from 64.233.187.99: icmp_seq=4 ttl=239 time=302.7 ms
64 bytes from 64.233.187.99: icmp_seq=5 ttl=239 time=316.4 ms
64 bytes from 64.233.187.99: icmp_seq=6 ttl=239 time=309.5 ms
64 bytes from 64.233.187.99: icmp_seq=7 ttl=239 time=308.9 ms

--- 64.233.187.99 ping statistics ---
8 packets transmitted, 8 packets received, 0% packet loss
round-trip min/avg/max = 301.1/310.0/316.4 ms

# ping -c 8 google.com
PING google.com (64.233.167.99): 56 data bytes
64 bytes from 64.233.167.99: icmp_seq=0 ttl=239 time=269.6 ms
64 bytes from 64.233.167.99: icmp_seq=1 ttl=239 time=271.3 ms
64 bytes from 64.233.167.99: icmp_seq=2 ttl=239 time=269.3 ms
64 bytes from 64.233.167.99: icmp_seq=3 ttl=239 time=274.2 ms
64 bytes from 64.233.167.99: icmp_seq=4 ttl=239 time=268.7 ms
64 bytes from 64.233.167.99: icmp_seq=5 ttl=239 time=269.6 ms
64 bytes from 64.233.167.99: icmp_seq=6 ttl=239 time=268.5 ms
64 bytes from 64.233.167.99: icmp_seq=7 ttl=239 time=269.2 ms

--- google.com ping statistics ---
8 packets transmitted, 8 packets received, 0% packet loss
round-trip min/avg/max = 268.5/270.0/274.2 ms

# traceroute 64.233.187.99
traceroute to 64.233.187.99 (64.233.187.99), 30 hops max, 38 byte packets
1 203.88.164.1 (203.88.164.1) 17.639 ms 17.132 ms 18.970 ms
2 202-123-94-73.ihenderson.com (202.123.94.73) 17.405 ms 17.613 ms 17.741 ms
3 202-123-94-89.ihenderson.com (202.123.94.89) 17.911 ms 22.485 ms 17.237 ms
4 203.131.241.121 (203.131.241.121) 19.631 ms 19.564 ms 19.924 ms
5 ge-0-0-0.r01.newthk01.hk.bb.verio.net (203.131.240.92) 20.727 ms 18.758 ms 18.696 ms
6 p4-7-2-3.r21.tokyjp01.jp.bb.verio.net (129.250.4.101) 72.798 ms 86.215 ms 86.170 ms
7 p64-2-1-0.r21.sttlwa01.us.bb.verio.net (129.250.4.185) 195.309 ms 197.576 ms 198.522 ms
8 xe-0-2-0.r20.sttlwa01.us.bb.verio.net (129.250.4.16) 182.989 ms 198.315 ms 183.735 ms
9 p16-6-0-0.r04.sttlwa01.us.bb.verio.net (129.250.2.69) 182.162 ms 181.433 ms 182.016 ms

# traceroute google.com
traceroute: Warning: google.com has multiple addresses; using 64.233.187.99
traceroute to google.com (64.233.187.99), 30 hops max, 38 byte packets
1 203.88.164.1 (203.88.164.1) 17.730 ms 18.438 ms 23.988 ms
2 202-123-94-73.ihenderson.com (202.123.94.73) 17.041 ms 16.996 ms 16.942 ms
3 202-123-94-41.ihenderson.com (202.123.94.41) 17.046 ms 17.807 ms 17.160 ms
4 203.131.241.121 (203.131.241.121) 19.895 ms 19.830 ms 18.473 ms
5 ge-0-0-0.r01.newthk01.hk.bb.verio.net (203.131.240.92) 18.845 ms 18.062 ms 20.691 ms
6 p4-7-2-3.r21.tokyjp01.jp.bb.verio.net (129.250.4.101) 86.301 ms 73.380 ms 74.348 ms
7 p64-2-1-0.r21.sttlwa01.us.bb.verio.net (129.250.4.185) 197.531 ms 195.366 ms 196.299 ms
8 xe-0-2-0.r20.sttlwa01.us.bb.verio.net (129.250.4.16) 195.067 ms 195.594 ms 198.543 ms
9 p16-6-0-0.r04.sttlwa01.us.bb.verio.net (129.250.2.69) 183.127 ms 181.962 ms 182.268 ms
10 so-3-3-1.cr1.sea1.us.above.net (64.125.12.29) 182.231 ms 182.857 ms

Is there a way checking the speed of ADSL broadband? I'm interested to know the actual speed of my 1.5Mb ADSL broadband connection. TIA

B.R.
satimis

Harry Kuhman
02-07-2006, 06:29 PM
Is there a way checking the speed of ADSL broadband? I'm interested to know the actual speed of my 1.5Mb ADSL broadband connection. TIA
Yup, it sounded line a dns issue. The simple fix would once again be to use a router, almost all (except some Dlinks) let you put your own choice for dns server in their settings to over ride what they get from the ISP. Do this once and everything is fixed. Otherwise you could run netcardconfig and give knoppix your own dns setting. There are likely other ways you could change this too; I just don't know them.

As to checking speed, I generally just tend to watch throughput when downloading from a newsgroup. But broadbandreports.com (formerly dslreports.com) used to offer some speed monitoring tools. I haven't used them in ages, but I rather expect the tools are still there.

satimis
02-08-2006, 05:20 AM
Hi Harry,

Tks for your advice.

What I can't resolve is on Win PC Internet browsing was not affected disregarding the dead primary nameserver. I made following text on a Win PC

C:\WINDOWS>ping 202.123.77.196

Pinging 202.123.77.196 with 32 bytes of data:

Reply from 202.123.95.81: TTL expired in transit.
Reply from 202.123.95.81: TTL expired in transit.
Reply from 202.123.95.81: TTL expired in transit.
Reply from 202.123.95.81: TTL expired in transit.

Ping statistics for 202.123.77.196:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 0ms, Average = 0ms

C:\WINDOWS>ping 202.123.77.213

Pinging 202.123.77.213 with 32 bytes of data:

Reply from 202.123.77.213: bytes=32 time=28ms TTL=252
Reply from 202.123.77.213: bytes=32 time=13ms TTL=252
Reply from 202.123.77.213: bytes=32 time=14ms TTL=252
Reply from 202.123.77.213: bytes=32 time=28ms TTL=252

Ping statistics for 202.123.77.213:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 13ms, Maximum = 28ms, Average = 20ms
The statistics on the dead nameserver was also % loss. But the response was from another nameserver.

On complaining to my ISP, he said the primary nameservers was checked active there. Could you please help me to understand this happening. TIA

Others noted with tks.

BR
satimis

Harry Kuhman
02-08-2006, 06:10 AM
On complaining to my ISP, he said the primary nameservers was checked active there. Could you please help me to understand this happening. TIA
I don't have any info on your ISP, but I know mine is an awful ISP that frequently has DNS problems (I've been having one for the past several days and unfortunately my router, a Dlink, does not let me in replace the DNS setting) and they are either too ignorant to even know they have probelms or they flat out lie about it. Do you have any reason to think this is not the same case for you?

As to why any OS would behave differently than another, this doesn't come as a shock to me at all. I imagine it all relates to how each system deals with both being given multiple DNS servers and also with how they process errors. It's not unlikely that one OS may switch to an alternate server faster than another.

satimis
02-08-2006, 09:41 AM
Hi Harry,

Your advice noted with tks

satimis

satimis
02-20-2006, 01:18 PM
Hi Harry and folks,

Today my ISP on his email in reply to my complaints said that 202.123.77.196 (primary nameserver) belongs to Proxy DNS. Therefore, the Firewall will block the IP so that I cannot get any success result on the ping test. For 202.123.77.213 and 202.123.77.209 (the final nameservers assigned by them), they are not Proxy DNS, and therefore I can find the result on the ping test.

I fail to understand why some nameservers belonging to Proxy DNS behind firewall, as my ISP claimed, another not belonging to Proxy DNS without firewall protection. Why “traceroute� test on 202.123.77.196 showed “HTTP/1.1 502 Gateway Error�? If only because of non Proxy DNS which is not the cause of slow Internet browsing why my ISP finally re-assigned another new set of nameservers? Only to satisfy my curiosity on pinging the nameservers?

Besides I haven't reported to them both "202.123.77.213 and 202.123.77.209" went through "ping" and "traceroute" tests.

Please help me to understand. TIA.

B.Regards
satimis

Harry Kuhman
02-20-2006, 09:03 PM
Today my ISP on his email in reply to my complaints said that 202.123.77.196 (primary nameserver) belongs to Proxy DNS. Therefore, the Firewall will block the IP so that I cannot get any success result on the ping test. For 202.123.77.213 and 202.123.77.209 (the final nameservers assigned by them), they are not Proxy DNS, and therefore I can find the result on the ping test.

I fail to understand why some nameservers belonging to Proxy DNS behind firewall, as my ISP claimed, another not belonging to Proxy DNS without firewall protection. Why “traceroute� test on 202.123.77.196 showed “HTTP/1.1 502 Gateway Error�? If only because of non Proxy DNS which is not the cause of slow Internet browsing why my ISP finally re-assigned another new set of nameservers? Only to satisfy my curiosity on pinging the nameservers?

Besides I haven't reported to them both "202.123.77.213 and 202.123.77.209" went through "ping" and "traceroute" tests.
Please help me to understand. TIA.
Thanks for the update. This confirms that it was a DNS issue. If I'm understanding properly and the DNS settings came from your ISP by DHCP the I would write back to them and ask why they were giving out a DNS server address that woul not work for you. I'm also not at all clear where "the firewall" is or why it blocked that address, but the ISP certainly shouldn't be assigning addresses that they know will be blocked. I would even consider asking for a refund of DSL fees over the time this was done and the service was useless.

Do I understand correctly that the ISP has fixed this and is now no longer assigning the bad DNS address and your problems are resolved?

I would have to see the entire traceroute to understand it enough to try to explain it, but it likely stops at "the Firewall". As to why the ISP is changing, they should have never had a DNS server in their assignment at al if they were blocking access to it. It will slow down everyone, but as you saw the amount of slow down depends on how many times the OS tries to get a response fron the primary DNS server and how long it waits before going to the backup DNS server, so some systems respond even worse than others to a badly configured primary DNS server.

satimis
02-22-2006, 12:52 PM
Hi Harry,


Do I understand correctly that the ISP has fixed this and is now no longer assigning the bad DNS address and your problems are resolved?Now they assigned another new set of nameservers. Both went through "ping" and "traceroute" tests.

My ISP tried to reject the bad nameservers assigned previously not working, saying that they only could not be pinged because behind firewall and were working.

I wonder why previously they assigned 2 nameservers on connecting, both behind firewall as claimed by them. Not until I made further complaint, then they assigned another new set of nameservers. Both went through "ping" and "traceroute" tests without bouncing between routers like playing tennis. Let me waiting for prolonged time before a webpage popup/connected.

I'm trying to discover the cause of the problem, if possible. Let them pay attention to it. I'm still bound by them with a 12 month contract, with only one (1) month elapsed.

I made further tests on the bad nameservers as follows using "dig"

1)
# dig @202.123.77.196 www.google.com

; <<>> DiG 9.2.5 <<>> @202.123.77.196 www.google.com
; (1 server found)
;; global options: printcmd
;; connection timed out; no servers could be reached

Still failed to connect the server.

2)
# dig @202.123.77.198 www.google.com
got the same printout


3)
# dig @202.123.77.213 -x 202.123.77.196

; <<>> DiG 9.2.5 <<>> @202.123.77.213 -x 202.123.77.196
; (1 server found)
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 34353
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 2

;; QUESTION SECTION:
;196.77.123.202.in-addr.arpa. IN PTR

;; ANSWER SECTION:
196.77.123.202.in-addr.arpa. 86400 IN PTR 202-123-77-196.ihenderson.com.

;; AUTHORITY SECTION:
77.123.202.in-addr.arpa. 86400 IN NS ns02.ihenderson.com.
77.123.202.in-addr.arpa. 86400 IN NS ns01.ihenderson.com.

;; ADDITIONAL SECTION:
ns02.ihenderson.com. 3600 IN A 202.123.77.213
ns01.ihenderson.com. 3600 IN A 202.123.77.209

;; Query time: 29 msec
;; SERVER: 202.123.77.213#53(202.123.77.213)
;; WHEN: Wed Feb 22 18:45:35 2006
;; MSG SIZE rcvd: 181

4)
# dig @202.123.77.213 -x 202.123.77.198
got the same printout except;-
id: 75
;; Query time: 21 msec
;; WHEN: Wed Feb 22 18:48:23 2006

Both of them referred to the same server;-
;; SERVER: 202.123.77.213#53(202.123.77.213)

202.123.77.213 is one of the nameservers assigned on connecting recently. Both of them are working properly and went through "ping" and "traceroute" tests.

B.R.
satimis