Bug 488176 - pppoe(rp-pppoe) cannot consistenly establish a connection
Summary: pppoe(rp-pppoe) cannot consistenly establish a connection
Keywords:
Status: CLOSED NEXTRELEASE
Alias: None
Product: Fedora
Classification: Fedora
Component: rp-pppoe
Version: 10
Hardware: x86_64
OS: Linux
low
medium
Target Milestone: ---
Assignee: Than Ngo
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2009-03-03 00:14 UTC by C.S.D.
Modified: 2009-03-18 19:05 UTC (History)
2 users (show)

Fixed In Version: 3.10-1.fc10
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2009-03-18 19:05:37 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)

Description C.S.D. 2009-03-03 00:14:12 UTC
Description of problem:

Trying to connect to my ISP using the mentioned version of rp-pppoe has a very low rate of success. It's like a wheel of fortune actually.

Now, let's move on to the subject:

- pppoe-discovery shows five distinct Access-Concentrators when called.

PPPoE connections would constantly fail so I decided to take a look at what was happening using tcpdump. Next is what I observed:

1. PPPoE connection *fails* when the PPPoE communication goes like this:

PADI..., PADO, PADO , PADR , PADO , PADI ...-> FAILED

...and this happened too often.

2.PPPoE connection *succeeds* when the PPPoE communication goes like this:

PADI..., PADO, PADR, PADS, <session stuff> ... -> OK

                   or

PADI..., PADO, PADR, PADO, PADO, PADS, PADO,...<session stuff>...-> OK

...and I had to be very lucky to get one of these sequences :).

A wild guess would be that it fails when it receives a group of two (or more?) PADO =before= sending PADR or I could simply be an isolated case :).

Anyway, I got 3.10 from http://www.roaringpenguin.com/, built and installed the rpm package, created the symlinks(adsl-*) and it seems not to have this issue although it takes a little bit longer to connect than I expected - doesn't bother me.

I will not post /etc/sysconfig/network-scripts/ifcfg-ppp0 because it was the same all the time. The only tweaking was regarding SYNCHRONOUS parameter in the early stages of trying to connect with PPPoE. It was eventually left to "no" for the whole run.

Network card is, as shown by lspci:
Broadcom Corporation NetLink BCM5787 Gigabit Ethernet PCI Express

dmesg:
tg3.c:v3.94 (August 14, 2008)
tg3 0000:03:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
tg3 0000:03:00.0: setting latency timer to 64
tg3: eth0: Link is up at 100 Mbps, full duplex.
tg3: eth0: Flow control is off for TX and off for RX.

IPV6 is disabled on the system, network interface is brought up without any IP address and there are no delays when bringing it up.

I also tried using a WindowsXP machine trying to determine if it is a ISP related problem, but it successfully connected about ten times in a row and it was pretty fast.

Version-Release number of selected component (if applicable):

rp-pppoe:
- rp-pppoe-3.8-5.fc10.x86_64

ppp:
- ppp-2.4.4-8.fc10.x86_64

kernel:
- 2.6.27.15-170.2.24.fc10.x86_64

How reproducible:

every time

Steps to Reproduce:
1.
2.
3.
  
Actual results:

PPPoE has an issue, seemingly when a group of two PADO responses are received before sending PADR.

Expected results:
To fix this if it really is a problem (it is for me).

Additional info:

Comment 1 Fedora Update System 2009-03-18 10:48:36 UTC
rp-pppoe-3.10-1.fc10 has been submitted as an update for Fedora 10.
http://admin.fedoraproject.org/updates/rp-pppoe-3.10-1.fc10

Comment 2 Fedora Update System 2009-03-18 19:05:32 UTC
rp-pppoe-3.10-1.fc10 has been pushed to the Fedora 10 stable repository.  If problems still persist, please make note of it in this bug report.


Note You need to log in before you can comment on or make changes to this bug.