| View previous topic :: View next topic |
| Author |
Message |
Darkshine
n00b
Joined: 06 Nov 2007
Posts: 53
Location: Kiev, Micronesia
|
 Posted: Thu Aug 21, 2008 10:13 pm Post subject: long replies in a ping command |
 |
|
I have a router at home (gentoo 2.6.1 and 3 PCs, which are linked to a router via the switch. 2 PCs have good internet connections via router, but the last one (gentoo 2.6.24) have a good connection only sometimes.
When I'm trying to ping some host from that last PC, I get very long delay between resolving the pinged hostname and beginning sending a packets. Moreover, all ping-replies are printer rarely (delays are about 5 seconds):
$ ping slashdot.com
PING slashdot.com (216.34.181.45) 56(84) bytes of data.
64 bytes from slashdot.org (216.34.181.45): icmp_seq=1 ttl=237 time=157 ms
64 bytes from slashdot.org (216.34.181.45): icmp_seq=2 ttl=237 time=173 ms (DUP!)
...
sometimes echo-replies contain label "DUP!". When I'm stop pinging on the PC and start pinging on the router, ping works excellent - no DUPs, echo-replies are printed as usually - 1 time per second.
I can solve this problem very easy: just restart /etc/init.d/net.eth0 on a PC and ping (and all another network utilities) works ok. Where is a problem? It looks like the problem exaclty on my PC and it seems the problem is not in DNS. |
|
| Back to top |
|
 |
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54805
Location: 56N 3W
|
| Posted: Thu Aug 21, 2008 10:48 pm Post subject: |
|
|
Darkshine,
It sounds like it may be your network card, Does on the slow PC show any errors ?
Look at this bit | Code: | RX packets:17423 errors:0 dropped:0 overruns:0 frame:0
TX packets:22989 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000 |
errors:0 is good. Test that the network card is good by trying another one.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
Darkshine
n00b
Joined: 06 Nov 2007
Posts: 53
Location: Kiev, Micronesia
|
| Posted: Fri Aug 22, 2008 12:40 am Post subject: |
|
|
Interface eth0 on a local PC:
RX packets:124059 errors:0 dropped:0 overruns:0 frame:0
TX packets:94427 errors:0 dropped:0 overruns:0 carrier:2
collisions:0 txqueuelen:1000
Interface eth0 on a router (local network):
RX packets:4125106 errors:0 dropped:0 overruns:0 frame:0
TX packets:4556123 errors:421 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
Interface eth1 on a router (internet):
RX packets:4606608 errors:0 dropped:0 overruns:0 frame:0
TX packets:4301978 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
So, I see 421 errors on a router's interface. Does that mean I should throw router's ethernet card into a thrash?
If errors happened on a router side, why restarting network interface on a local side usually prevents to correct working?
thanks |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54805
Location: 56N 3W
|
| Posted: Fri Aug 22, 2008 12:01 pm Post subject: |
|
|
Darkshine,
Thats a very small number of errors in the total packet count and is unlikely to account for your problem.
Try different ports on your switch, try different cables between the 'slow' PC and the switch.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
Sysa
Apprentice
Joined: 16 Mar 2005
Posts: 161
Location: Europe
|
| Posted: Fri Aug 22, 2008 3:36 pm Post subject: |
|
|
| Darkshine wrote: | Interface eth0 on a local PC:
RX packets:124059 errors:0 dropped:0 overruns:0 frame:0
TX packets:94427 errors:0 dropped:0 overruns:0 carrier:2
collisions:0 txqueuelen:1000
Interface eth0 on a router (local network):
RX packets:4125106 errors:0 dropped:0 overruns:0 frame:0
TX packets:4556123 errors:421 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
Interface eth1 on a router (internet):
RX packets:4606608 errors:0 dropped:0 overruns:0 frame:0
TX packets:4301978 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
So, I see 421 errors on a router's interface. Does that mean I should throw router's ethernet card into a thrash?
If errors happened on a router side, why restarting network interface on a local side usually prevents to correct working?
thanks |
Check the cables (bad pairing?), speed/duplex settings for hosts/router/switch
_________________
RedHat -> SuSE -> Debian -> Gentoo |
|
| Back to top |
|
|
Darkshine
n00b
Joined: 06 Nov 2007
Posts: 53
Location: Kiev, Micronesia
|
| Posted: Fri Aug 22, 2008 10:43 pm Post subject: |
|
|
| Sysa wrote: |
Check the cables (bad pairing?), speed/duplex settings for hosts/router/switch
|
I have a D-Link 1005D switch with 5 ports. This switch is able to detect if the cable is "comp-comp", not "comp-switch", then switch will be functioning correctly anyway. So, it seems like not a bad pairing. Moreover, my cable from problem PC to switch is "comp-switch"
what you mean in "speed/duplex settings"? |
|
| Back to top |
|
|
Sysa
Apprentice
Joined: 16 Mar 2005
Posts: 161
Location: Europe
|
| Posted: Sat Aug 23, 2008 4:59 am Post subject: |
|
|
| Darkshine wrote: | | Sysa wrote: |
Check the cables (bad pairing?), speed/duplex settings for hosts/router/switch
|
I have a D-Link 1005D switch with 5 ports. This switch is able to detect if the cable is "comp-comp", not "comp-switch", then switch will be functioning correctly anyway. So, it seems like not a bad pairing. Moreover, my cable from problem PC to switch is "comp-switch"
what you mean in "speed/duplex settings"? |
"bad pairing" - if you connect cable pins in wrong order: Ethernet cables use twisted pairs ( 1-2, 3-6, 4-5, 7-8 ). General beginner's mistake is to connect ( 1-2, 3-4, 5-6, 7-8 ) pairs for self-made cables. A switch detects your connection (all LED's lights correct!) but you'll have a lot of transmission errors and unstable connection as a result.
Please see http://en.wikipedia.org/wiki/10BASE-T for details.
Speed [10/100/1000Mbps] and duplex mode [half,full] must be the same for both connected devices.
Autosensing sometimes works wrong way and better to set it manually for sure.
_________________
RedHat -> SuSE -> Debian -> Gentoo |
|
| Back to top |
|
|
Sysa
Apprentice
Joined: 16 Mar 2005
Posts: 161
Location: Europe
|
| Posted: Sat Aug 23, 2008 5:10 am Post subject: Re: long replies in a ping command |
|
|
| Darkshine wrote: | ...
When I'm trying to ping some host from that last PC, I get very long delay between resolving the pinged hostname and beginning sending a packets. Moreover, all ping-replies are printer rarely (delays are about 5 seconds):
$ ping slashdot.com
PING slashdot.com (216.34.181.45) 56(84) bytes of data.
64 bytes from slashdot.org (216.34.181.45): icmp_seq=1 ttl=237 time=157 ms
64 bytes from slashdot.org (216.34.181.45): icmp_seq=2 ttl=237 time=173 ms (DUP!)
...
I can solve this problem very easy: just restart /etc/init.d/net.eth0 on a PC and ping (and all another network utilities) works ok. Where is a problem? It looks like the problem exaclty on my PC and it seems the problem is not in DNS. |
1. Check eth0 settings: netmask and broadcast address, - they must be the same for all network segment members.
2. Check your connection - try to ping your router (gateway).
3. Check your DNS settings, try to ping pure IP address (e.g. 216.34.181.45) from the host.
_________________
RedHat -> SuSE -> Debian -> Gentoo |
|
| Back to top |
|
|
|
Networking & Security
