LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Slackware (http://www.linuxquestions.org/questions/slackware-14/)
-   -   ping delay network issue (http://www.linuxquestions.org/questions/slackware-14/ping-delay-network-issue-4175423761/)

riwi 08-24-2012 05:51 AM

ping delay network issue
 
I've bought a new intel dn2800mt mini-itx board with atom proc and intel e1000e onboard nic and installed slackware32 13.37 just like I did on 2 other PC's.

Any PC on my network I try to ping from the atom PC has strange delays in ping :
Code:

root@riwiatom:~# ping riwilinux
PING riwilinux (192.168.1.3) 56(84) bytes of data.
64 bytes from riwilinux (192.168.1.3): icmp_req=1 ttl=64 time=2.44 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=2 ttl=64 time=0.635 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=3 ttl=64 time=0.635 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=4 ttl=64 time=0.629 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=5 ttl=64 time=356 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=6 ttl=64 time=0.272 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=7 ttl=64 time=0.633 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=8 ttl=64 time=0.623 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=9 ttl=64 time=346 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=10 ttl=64 time=0.640 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=11 ttl=64 time=0.101 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=12 ttl=64 time=14.9 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=13 ttl=64 time=125 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=14 ttl=64 time=0.650 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=15 ttl=64 time=0.626 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=16 ttl=64 time=4.02 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=17 ttl=64 time=264 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=18 ttl=64 time=0.635 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=19 ttl=64 time=296 ms

When I tcpdump on the destination PC I see that the received icmp request is answered without delay (ie always < 1ms)
I have swapped network cables.
Both PCs connect to the same switch. I also tried a different switch.
I updated the e1000e driver on the atom PC from 1.2.7-k2 to 2.0.0.1.
I recompiled the kernel for Atom processor and included the e1000e driver in the kernel now.

Yesterday I booted with a Live CD "backtrack5 linux" Ubuntu based and then the ping problem is gone.
This leads me to believe that it must be some slackware/driver issue

What could this be?

abrinister 08-24-2012 10:48 PM

Do you have linux-firmware installed? It provides a bunch of binary blobs for network cards and such.

Alex Brinister

riwi 08-25-2012 02:59 PM

Quote:

Originally Posted by abrinister (Post 4763680)
Do you have linux-firmware installed? It provides a bunch of binary blobs for network cards and such.

Alex Brinister

Thanks, I'll check that out.
I also noticed that the backtrack-linux cd has a newer kernel than I am running with on my stock 13.37 (2.6.37.6)
Maybe I should try to upgrade to slackware-current. I have always run with 'released' slackware versions without updating much inbetween major releases.

abrinister 08-25-2012 04:14 PM

Yeah that kernel is pretty old. I'm running Arch and my kernel is version 3.4.9. Kernel updates usually help when it comes to drivers.

Alex Brinister

riwi 10-07-2012 05:09 PM

Quote:

Originally Posted by riwi (Post 4764172)
Thanks, I'll check that out.
I also noticed that the backtrack-linux cd has a newer kernel than I am running with on my stock 13.37 (2.6.37.6)
Maybe I should try to upgrade to slackware-current. I have always run with 'released' slackware versions without updating much inbetween major releases.

I've upgrade to slackware32 14.0. So now I am running 3.2.29

Linux riwiatom 3.2.29-smp #1 SMP Mon Sep 17 13:02:10 CDT 2012 i686 Intel(R) Atom(TM) CPU N2800 @ 1.86GHz GenuineIntel GNU/Linux

Unfortunately not a complete solution but I do see improvement.
No ping delay varies between 0.600 ms (mostly) to 5-15 ms occasionally
So not as bad as before.

Code:

root@riwiatom:# ping riwilinux
PING riwilinux (192.168.1.3) 56(84) bytes of data.
64 bytes from riwilinux (192.168.1.3): icmp_req=1 ttl=64 time=1.78 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=2 ttl=64 time=0.732 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=3 ttl=64 time=6.77 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=4 ttl=64 time=9.50 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=5 ttl=64 time=6.47 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=6 ttl=64 time=0.673 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=7 ttl=64 time=12.3 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=8 ttl=64 time=0.700 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=9 ttl=64 time=0.801 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=10 ttl=64 time=0.693 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=11 ttl=64 time=3.65 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=12 ttl=64 time=9.02 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=13 ttl=64 time=4.67 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=14 ttl=64 time=4.04 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=15 ttl=64 time=0.701 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=16 ttl=64 time=0.666 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=17 ttl=64 time=2.64 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=18 ttl=64 time=1.12 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=19 ttl=64 time=0.692 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=20 ttl=64 time=0.680 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=21 ttl=64 time=0.691 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=22 ttl=64 time=3.99 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=23 ttl=64 time=0.696 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=24 ttl=64 time=0.684 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=25 ttl=64 time=2.72 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=26 ttl=64 time=5.69 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=27 ttl=64 time=5.61 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=28 ttl=64 time=0.681 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=29 ttl=64 time=0.697 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=30 ttl=64 time=3.17 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=31 ttl=64 time=10.9 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=32 ttl=64 time=0.667 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=33 ttl=64 time=0.672 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=34 ttl=64 time=0.717 ms
64 bytes from riwilinux (192.168.1.3): icmp_req=35 ttl=64 time=4.49 ms

I have tried the huge and the generic kernel

intel e1000e driver :
root@riwiatom:# ethtool -i eth0
driver: e1000e
version: 1.5.1-k
firmware-version: 2.1-2
bus-info: 0000:01:00.0

Maybe it is something external. But I can't explain why ping is stable < 1ms when I boot with some other distro from a stick and not with slack13.37 or slack14.0


All times are GMT -5. The time now is 03:58 PM.