Hi, I have problem with ethernet.
One day Beaglebone ethernet work correctly but after that I can't get
access to ethernet.
During boot I will got message:
[ 4.031072] PHY 0:00 not found
[ 4.034313] PHY 0:01 not found
ifup eth0 give results:
root@beaglebone:~# ifdown eth0
root@beaglebone:~# ifup eth0
[ 714.253219] PHY 0:00 not found
[ 714.256579] PHY 0:01 not found
udhcpc (v1.19.3) started
Sending discover...
Sending discover...
Sending discover...
No lease, failing
RJ45 connector orange led blink couble times durin boot. It is same is
cable connected or not. If bone is long time without power the orange
led is on at start up boot but go off during boot. At run time leds
are off, is cable connected or not.
I have tried with power and without it. Several different Ångström
images eq. 16.11, 13.12 and 22.12 dated versions, but no working
ethernet.
I found couple similar treads by google but no help from those.
Have anybody any idea what I should do that ethernet will work again?
Or is the board broken
Something new, the board was one day without power and now network
works a couple minutes. After that ethernet miss, I used ping to see
condition of ethernet. After rebooting no ethernet.
That can be PCB issue...
Well, yes. On my BeagleBone the ethernet connector LED is orange most
of the time regardless of whether or not there is a cable plugged in.
I tried different cables and routers but no "sustainable" change.
On rare occasions the green LED starts to blink very quickly (~ 5 Hz)
after plugging (ethernet autonegotiation?) a cable. But it seems to
have no positive effect on the software.
ifup eth0 gives following result:
root@beaglebone:~# ifup eth0
[ 83.679984]
[ 83.679992] CPSW phy found : id is : 0x7c0f1
[ 83.686743] PHY 0:01 not found
udhcpc (v1.19.3) started
Sending discover...
Sending discover...
Sending discover...
No lease, failing
I have the 2011.12.13 Angstrom image running.
In another thread I read the BeagleBone is "picky" with regard to the
ethernet cable used. Any details about when a cable is "good" from the
BeagleBone's perspective?
flobe: I have same Angstrom image than you at now. I read some were
with it the ethernet should work.
I can see that on you board ethernet PHY 0:00 is found.
[ 83.679992] CPSW phy found : id is : 0x7c0f1
My situatuin is still:
[ 4.420247] PHY 0:00 not found
I tried with several cables and different switch ports, but no
difference. But seems that omap chip didn't find PHY-chip.
Situation is strange because ethernet work short period if board is
unpowered over 12 hours. One possibility with my case is that clock
source (crystal Y4) of PHY have bad stability. But I don't have tools
to measure it. Like "Short period aging of crystal" I don't know what
is right word for that.
Last time when ehernet work I tried to use cold spray to see if any
effect and ethernet stop working immediately.
I received a board a couple of days ago, with exactly the same
behavior as Kari. When I first powered it up the ethernet port worked
fine, but after maybe 10 minutes the link went down. To begin with I
could restore link by rebooting the board. but after about an hour it
required a power cycle, and now it does not work at all.
I have also tried with several cables and different switch ports, to
no avail.
I also get the 'PHY 0:00 not found' error message, both when booting,
and when issuing ab ifup command.
I received a board a couple of days ago, with exactly the same
behavior as Kari. When I first powered it up the ethernet port worked
fine, but after maybe 10 minutes the link went down. To begin with I
could restore link by rebooting the board. but after about an hour it
required a power cycle, and now it does not work at all.
I have also tried with several cables and different switch ports, to
no avail.
I also get the 'PHY 0:00 not found' error message, both when booting,
and when issuing ab ifup command.
Are you sure your power supply is capable of delivering enough
current?
Yes. I started using the board powered from USB, and to begin with the
ethernet port worked fine. After I began to see errors, I tried using
a 5V 3A supply, but this did not change anything.
I also have this problem. I am running Angstrom-Cloud9-IDE-eglibc-ipk-
v2011.12-core-beaglebone-2011.12.26.img and getting the same results.
Ethernet stops working after a while.
root@beaglebone:~#ifup
eth0
[ 7849.112585] PHY 0:01 not
found
udhcpc (v1.19.3)
started
Sending
discover...
Sending
discover...
Sending
discover...
No lease, failing
I have tried using two different 2A 5V power supplies, running off of
USB power, and a bunch of ethernet cables to no avail.
Some more information: I haven't checked the PHY supply, because it is
shared with a number of other devices which are still working fine
(USB2412HUB, FT223, EEPROM), so I have assumed it is OK. I checked the
ethernet crystal (Y4), and it is oscillating at 25MHz with a good
amplitude. I also checked whether the PHY is sending link pulses on
pair A, and it is not. So, at least in my case, it looks like the PHY
is toast.
I guess I'll be returning my board. It is rather worrying that there
are already this many cases of problems with the PHY.
I see the same behavior as Kari: if I leave the board powered down for
a while (hours) then the ethernet will work for a short while.
Another thing I have noticed is that the yellow LED in the RJ45 jack
will turn on when the board is powered up. If it stays on, then the
ethernet port will work when I insert a cable. But usually it just
blinks, as if the PHY is powering up, and immediately powering down
again.
My first BeagleBone had the same issues: PHY 0:00, no link at all after it run for ca. 12 hrs. It went even worse: even the LED went mad without even booted, later it stopped booting at all (You may find some postings in the list archive):
You won’t be receiving any feedback anytime soon. Watterrott holds all the RMAs until they have several and then they ship them for repair. We won’t be getting it for months.
We learned that a long time ago. we are working to get the board back for Watterott as soon as we can. We really need to see what is going on from a HW standpoint.
Would you be willing to request an RMA and give us a chance to look a the board? So far, nobody will let us see one. We will replace it with a new board.
