Really weird internet lag

Crone · 2012-05-11 23:16:56

Hello everyone

So I have a wireless network consisting of 7 network devices, one of which is the arch box. The other devices run Android,iOs,Windows 7.

Every single device has no internet problems but the arch system. Let me note that this is new, I used to connect to the same network without any troubles.
By internet lag I mean:

1) Pages take too long to open or the connection resets. Happens both on Mozilla and Chromium
2) Pinging servers (etc google) takes a lot of time.
3) Download speed (torrents/pacman) is fine but sometimes it randomly drops to 0 and does never come back

Here is some information about my system:

I connect to a wireless WPA/WPA2 encrypted network through wicd.

My rc.conf:

#
# /etc/rc.conf - Main Configuration for Arch Linux
#
# See 'man 5 rc.conf' for more details
#

# LOCALIZATION
# ------------
HARDWARECLOCK="UTC"
TIMEZONE="Europe/Athens"
KEYMAP="us"
CONSOLEFONT=
CONSOLEMAP=
LOCALE=
DAEMON_LOCALE="yes"
USECOLOR="yes"

# HARDWARE
# --------
MODULES=(fglrx snd-usb-audio)
USEDMRAID="no"
USEBTRFS="no"
USELVM="no"

# NETWORKING
# ----------
HOSTNAME=GliderVillage

interface=wlan0
address=
netmask=
broadcast=
gateway=

NETWORK_PERSIST="no"

# DAEMONS
# -------
#
DAEMONS=(dbus syslog-ng wicd crond xkbevd wireless-regdom)

My hosts:

#
# /etc/hosts: static lookup table for host names
#

#<ip-address>	<hostname.domain.org>	<hostname>
127.0.0.1	localhost.localdomain	localhost GliderVillage
::1		localhost.localdomain	localhost

# End of file

My netconfig (untouched)

#
# The network configuration file. This file is currently only used in
# conjunction with the TI-RPC code in the libtirpc library.
#
# Entries consist of:
#
#       <network_id> <semantics> <flags> <protofamily> <protoname> \
#               <device> <nametoaddr_libs>
#
# The <device> and <nametoaddr_libs> fields are always empty in this
# implementation.
#
udp        tpi_clts      v     inet     udp     -       -
tcp        tpi_cots_ord  v     inet     tcp     -       -
udp6       tpi_clts      v     inet6    udp     -       -
tcp6       tpi_cots_ord  v     inet6    tcp     -       -
rawip      tpi_raw       -     inet      -      -       -
local      tpi_cots_ord  -     loopback  -      -       -
unix       tpi_cots_ord  -     loopback  -      -       -

Additionally I use an ALFA card with rtl8187 driver. Have also tested with a NETGEAR card, same issue.
My router is brand new.
Let me say that the problem appears the moment I connect to the network.

Thanks

Edit: I can provide you with a wireshark packet dump if needed

Last edited by Crone (2012-05-11 23:23:06)

ewaller · 2012-05-12 00:07:57

I would start with some sanity checks. Look at the output of iwconfig and be sure you are really associating with your router, and not the open one across the street. Check your signal level while you are at it.
Look at the output of ip route and make sure your default gateway is that of your router and is through wlan0.
Why does your /etc/rc.conf "interface" line point to wlan0. I think that should be eth0.
Is your eth0 attached to the same router to which your trying to connect wirelessly? It should not be.

Gcool · 2012-05-12 05:59:31

Also consider testing with wifi power management disabled if you have it enabled at the moment (iwconfig will show you the current status).

Crone · 2012-05-12 11:28:01

Yes I am 100% sure that I am associating with my router, I have been using my routers control panel to check things constantly. Here is iwconfig's output:

 
lo        no wireless extensions.

wlan0     IEEE 802.11bg  ESSID:"Namsayin"  
          Mode:Managed  Frequency:2.447 GHz  Access Point: 00:05:59:4A:AF:B5   
          Bit Rate=18 Mb/s   Tx-Power=30 dBm   
          Retry  long limit:7   RTS thr:off   Fragment thr:off
          Power Management:off
          Link Quality=66/70  Signal level=-44 dBm  
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:10   Missed beacon:0

eth0      no wireless extensions.

Power management is disabled as you can see.

ip route looks fine:

default via 192.168.2.1 dev wlan0  metric 303 
192.168.2.0/24 dev wlan0  proto kernel  scope link  src 192.168.2.5  metric 303

Also, /etc/rc.conf interface was set to wlan0 while I was installing arch and I think its right. I am not physically connecting to the router through eth0 (router is far far away from me).

Thanks for the replies

Edit: By the way since I made this post I removed wireless-regdom from my daemon array (I use it to change my regulatory domain in order to transmit with my wireless card to higher dbm). Since then things are going much better. I am gonna stay like that for a couple of days so I can know for sure though. Maybe the fact that I connect to the network first and then change the reg domain is a problem? (wicd daemon is first)

Last edited by Crone (2012-05-12 11:44:51)

Strike0 · 2012-05-12 13:51:41

Crone wrote:

Edit: By the way since I made this post I removed wireless-regdom from my daemon array (I use it to change my regulatory domain in order to transmit with my wireless card to higher dbm). Since then things are going much better. I am gonna stay like that for a couple of days so I can know for sure though. Maybe the fact that I connect to the network first and then change the reg domain is a problem? (wicd daemon is first)

I would not think that is a problem, if the link stays up. Your iw output, however, shows only 18Mb/s albeit a high link quality which is a bit strange.
In any case I would avoid manipulating standard dbm parameters. If that is what you write you do, that may well be your problem. There is a common misconception that increasing TX power results in a stronger signal. It won't since the router's TX power does not change. Better link quality is best achieved with better antennae. But your link quality looks good anyway IMO.

Crone · 2012-05-12 14:32:18

Hmm I will change the txpower back to 20dbm, but I am pretty sure this is not the issue since I had the same problem on a NETGEAR card which cannot physically transmit that high. As to why I have set it to 30dbm its because this way I can "see" more wireless networks. I don't know the technical details, all I know is that a regular card detects 15 wifi networks while the alfa card detects about 30 or more. Anyway ill reset it to 20 after a couple of days

ewaller · 2012-05-12 15:42:54

As moderator, I have to question one thing. Are you trying to force your system into a state that is not in compliance with the electromagnetic emissions regulations where you live? Or are you in a locale that permits the higher output power?

If it is the former, please don't. First, it is bad manners; it will drive up the signal to noise ratio, may cause your signal to splatter out of band, and it degrades your neighbor's performance. Second, it adds credence to the argument that, "We cannot allow open drivers because hackers will use the software radios in ways that are not allowed". Third, it is illegal, which brings me to our policy

If it is the latter, carry on.

Last edited by ewaller (2012-05-12 15:43:10)

Strike0 · 2012-05-12 18:46:21

Crone wrote:

..this way I can "see" more wireless networks. I don't know the technical details, all I know is that a regular card detects 15 wifi networks while the alfa card detects about 30 or more. ..

Yeah - have you tested that with putting the Alfa antenna onto your netgear card also? ;-)

Your card will see more networks yes, but noise will be higher as well - distorting the link/throughput. Further, the more important reasons and negative side effects ewaller pointed out rightfully. CRDA's are there to adhere for important reasons.

Crone · 2012-05-12 22:07:26

Oh man I changed this for a couple of days for pen testing, into my own networks. If you believe that is relevant to my internet connection feel free to ban me/lock the thread.

I shall say that when I started the thread I had no intention to lure you to anything illegal. In my country i think that yes this is a bit of illegal (transmitting to 30dbm). I think 27dbm is the maximum. And no I am not in a locale allowing such things. If we had such a place I ensure you i would be there.
I am just doing lab tests people, its a hobby.

Anyway feel free to take any action needed as a modarator regarding this topic.

Thanks for the replies again btw.

Last edited by Crone (2012-05-12 22:16:16)

ewaller · 2012-05-13 03:32:15

Crone wrote:

I shall say that when I started the thread I had no intention to lure you to anything illegal. In my country i think that yes this is a bit of illegal (transmitting to 30dbm). I think 27dbm is the maximum. And no I am not in a locale allowing such things.
...
Anyway feel free to take any action needed as a modarator regarding this topic.

No, that is not the intent. I actively encourage anything that contributes to learning, even short term experiments that may fall out of legal bounds in the name of education.

Our influence is such that I do not want to promote extra-legal behavior.

fukawi2 · 2012-05-13 10:00:20

Please post the output of:

dig www.google.com

(You will need to install `dnsutils` package if you don't already have it)

Crone · 2012-05-13 12:33:27

ewaller wrote:

No, that is not the intent. I actively encourage anything that contributes to learning, even short term experiments that may fall out of legal bounds in the name of education.
Our influence is such that I do not want to promote extra-legal behavior.

I completely understand and respect that.

 dig www.google.com

output:

; <<>> DiG 9.9.0 <<>> www.google.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 60555
;; flags: qr aa rd ad; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1
;; WARNING: recursion requested but not available

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;www.google.com.			IN	A

;; ADDITIONAL SECTION:
www.google.com.		240	IN	A	173.194.35.147

;; Query time: 12 msec
;; SERVER: 192.168.2.1#53(192.168.2.1)
;; WHEN: Sun May 13 18:30:45 2012
;; MSG SIZE  rcvd: 59

Internet is still stable btw since I removed wireless-regdom. I guess I'll re-enable it now to see if it is of relevance.

Edit: Nope wireless-regdom isn't it. Internet still stable. This was the only modification I made since the post.

Last edited by Crone (2012-05-13 12:43:23)

fukawi2 · 2012-05-13 23:47:00

Crone wrote:

 dig www.google.com 
output:
......
;; Query time: 12 msec
.......

I was curious if it was a DNS issue, but this time does not seem unreasonable. I can't think of any other suggestions at the moment, sorry.

EDIT:
Although, this part is interesting...

;; WARNING: recursion requested but not available

Your internal DNS server should be offering you recursion; I'm not sure how it got an answer without recursion being offered. Have you tried other DNS servers? (Google: 8.8.8.8 and 8.8.4.4 or OpenDNS 208.67.222.222 and 208.67.220.220)

Last edited by fukawi2 (2012-05-13 23:49:25)

Crone · 2012-05-14 00:26:22

Hmmm. I used to have a no-ip host assigned to my dynamic ip address 2 weeks ago. I also configured my router to use the no-ip service. However I have this service disabled the past two weeks and also I removed my host from no-ip.

I will try google dns servers and post back.

Edit:

i digged google a couple of times before changing dns here is the output: (The Warning is gone)

; <<>> DiG 9.9.0 <<>> www.google.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 8912
;; flags: qr rd ra; QUERY: 1, ANSWER: 6, AUTHORITY: 4, ADDITIONAL: 5

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;www.google.com.			IN	A

;; ANSWER SECTION:
www.google.com.		1800	IN	CNAME	www.l.google.com.
www.l.google.com.	81	IN	A	173.194.35.180
www.l.google.com.	81	IN	A	173.194.35.176
www.l.google.com.	81	IN	A	173.194.35.177
www.l.google.com.	81	IN	A	173.194.35.178
www.l.google.com.	81	IN	A	173.194.35.179

;; AUTHORITY SECTION:
google.com.		165080	IN	NS	ns1.google.com.
google.com.		165080	IN	NS	ns2.google.com.
google.com.		165080	IN	NS	ns3.google.com.
google.com.		165080	IN	NS	ns4.google.com.

;; ADDITIONAL SECTION:
ns1.google.com.		146500	IN	A	216.239.32.10
ns2.google.com.		59295	IN	A	216.239.34.10
ns3.google.com.		168253	IN	A	216.239.36.10
ns4.google.com.		154790	IN	A	216.239.38.10

;; Query time: 47 msec
;; SERVER: 192.168.2.1#53(192.168.2.1)
;; WHEN: Mon May 14 06:42:09 2012
;; MSG SIZE  rcvd: 279

Now I changed the dns to 8.8.8.8 but I am maybe doing something wrong because dig times out. However I can browse the internet.
I checked "Use Static DNS" and "Use Global dns" and set DNS 1 to 8.8.8.8 on wicd. Left the rest blank

dig www.google.com
; <<>> DiG 9.9.0 <<>> www.google.com
;; global options: +cmd
;; connection timed out; no servers could be reached

In any case I am good right now so I should not bother you guys anymore. I will come back here if the issue comes up again (we will have more clues that way).

Last edited by Crone (2012-05-14 00:57:00)

Arch Linux

#1 2012-05-11 23:16:56

Really weird internet lag

#2 2012-05-12 00:07:57

Re: Really weird internet lag

#3 2012-05-12 05:59:31

Re: Really weird internet lag

#4 2012-05-12 11:28:01

Re: Really weird internet lag

#5 2012-05-12 13:51:41

Re: Really weird internet lag

#6 2012-05-12 14:32:18

Re: Really weird internet lag

#7 2012-05-12 15:42:54

Re: Really weird internet lag

#8 2012-05-12 18:46:21

Re: Really weird internet lag

#9 2012-05-12 22:07:26

Re: Really weird internet lag

#10 2012-05-13 03:32:15

Re: Really weird internet lag

#11 2012-05-13 10:00:20

Re: Really weird internet lag

#12 2012-05-13 12:33:27

Re: Really weird internet lag

#13 2012-05-13 23:47:00

Re: Really weird internet lag

#14 2012-05-14 00:26:22

Re: Really weird internet lag

Board footer