Dropping Pings

[themacks]

Pings to the Raspberry Pi fail intermittently. I'm currently running the latest kernel from the package manager. Any ideas on what is causing it or what can be done to prevent it?

245 packets transmitted, 240 packets received, 2.0% packet loss
round-trip min/avg/max/stddev = 1.203/6.675/95.676/10.166 ms

Linux raspi 3.1.9-18-ARCH+ #1 PREEMPT Sat Jun 2 20:07:09 UTC 2012 armv6l GNU/Linux

[harryths]

windows wired network:

Code: Select all

Packets: Sent = 245, Received = 245, Lost = 0 (0% loss),
Minimum = 0ms, Maximum = 1ms, Average = 0ms

ubuntu wireless network:

Code: Select all

245 packets transmitted, 245 received, % packet loss, time 244432ms
rtt min/avg/max/mdev = 1.917/37.027/2017.673/224.761

I'm surprised by the max, this happened a couple of times during pinging

Were you on a wireless machine??

[thogue]

I figured I would bump this thread.

I am having ping drop issue about .2 - .4 % packet loss. Most recent Arch as of 2 days ago. Hardwired. Confirmed switch and network equipment is 200% percent. No tx/rx errors at either side, etc.

Anyone else experiencing issues?

[KnightMB]

Just tried on the lastest updated image. Using Pi vB

Ethernet

Code: Select all

282 packets transmitted, 282 received, 0% packet loss, time 281010ms
rtt min/avg/max/mdev = 0.339/0.455/1.358/0.075 ms

[thogue]

well, totally different pi/sdcard/cable/network/psu/fresh install. I still seeing .1 percent. However this is my first test here with this setup could be a random loss of a single ping. I will try a few more tests.

Loss Count
0.1% 7836 1.1 0.9 0.9 18.7 0.2


From what I have seen a slight bit of load can make a difference. ie, running nothing on the rpi and pinging it vs, say running MTR inside screen.

Yeah it seems it doesn't appear until the 1000 or 2000 count sometimes.

on an off-topic note I have been enjoying Arch very much otherwise. Just want to get any network issues figured out before I attempt to go live with hosting e-mail/websites from it.

[thogue]

Looks like it could be related to power issues on the other Pi (known blown poly).
Finally a good test on the new setup.
0.0% 10304 1.1 0.9 0.8 17.0 0.4

[thogue]

swapped out the pi today with a model B.

Much better now

0.0% 2098 1.2 1.0 0.9 6.0 0.3