Note: This is a beta release of Red Hat Bugzilla 5.0. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Also email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback here.
Bug 164509 - ifup-eth code order for wireless cards breaks Master mode configuration
Summary: ifup-eth code order for wireless cards breaks Master mode configuration
Status: CLOSED DUPLICATE of bug 122801
Alias: None
Product: Fedora
Classification: Fedora
Component: initscripts
Version: 4
Hardware: All
OS: Linux
Target Milestone: ---
Assignee: Bill Nottingham
QA Contact: Brock Organ
Depends On:
Blocks: FC5Target
TreeView+ depends on / blocked
Reported: 2005-07-28 13:13 UTC by Brett L. Trotter
Modified: 2014-03-17 02:55 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Last Closed: 2005-10-03 18:04:52 UTC

Attachments (Terms of Use)

Description Brett L. Trotter 2005-07-28 13:13:35 UTC
with initscripts-8.11.1-1

When using a wireless card in Master mode if you add BRIDGE=br0 to ifcfg-ath0 
(with the appropriate ifcfg-br0, etc setup) the wireless card goes into managed 
mode. Removing BRIDGE=br0 causes the wireless card to come up in master mode as 
desired but obviously is not in the bridge. This is due to the order things 
happen in ifup-eth.

I moved the wireless check before the bridging section and modified the 
bridging section as such:

# is the device wireless? If so, configure wireless device specifics
is_wireless_device ${DEVICE} && . ./ifup-wireless

if [ -n "${BRIDGE}" -a -x /usr/sbin/brctl ]; then
        if [ ! -n "${MODE}" ]; then
                /sbin/ip addr flush dev ${DEVICE} 2>/dev/null
                /sbin/ip link set dev ${DEVICE} up
      /usr/sbin/brctl addif ${BRIDGE} ${DEVICE}
      # Upon adding a device to a bridge,
      # it's necessary to make radvd reload its config
      [ -r /var/run/radvd/ ] && kill -HUP 
`cat /var/run/radvd/`
      exit 0

this seems to fix the problem.

Comment 1 Brett L. Trotter 2005-07-28 13:14:53 UTC
Actually, now that i look at it, i think the only modification necessary is the 
moving of the wireless check - the extra if within the bridging section fails 
to bring the interface up now. Sorry!

Comment 2 Lee Ryman 2005-10-02 16:11:29 UTC
I believe some clarification is needed here. If a ethernet device is bridged (by
having a "BRIDGE=br0" or similar entry in its ifcfg-eth* script), then nothing
after line 79 in the ifup-eth script is executed because of the `exit 0` after
the radvd reload. This includes, amongst other things, the check by the ifup-eth
script to see if the device has wireless extensions and if so, the execution of
ifup-wireless. Therefore none of your wireless parameters are actually
propergated to the device.

As to what the sollution is, I'm not entirely sure just moving the wireless
check is sufficient. I have only a limited knowledge in these scripts; can any
of the experts please determin if anything after the bridge check is also
neccessary to be invoked? If so, either the checks need to be reordered or
something needs to be done about the exit 0 on line 79.

Kind regards.

Comment 3 Bill Nottingham 2005-10-03 18:04:52 UTC

*** This bug has been marked as a duplicate of 122801 ***

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