Hi Andrew, On 18.06.2012 16:33, Andrew Shadura wrote:
> Michael, ADDRFAM=NetworkManager in the dispatcher script, why is it > there? I don't quite remember, are there any hook scripts which depend > on this exact value? Because actually when it's passed there, most of > the scripts will ignore the fact they're called. So either the script > should pass inet or inet6 depending on the type of the interface, or we > should check for it too in mountnfs (and, probably, somewhere else). Just to be clear here: Is this particular bug report about a/ NM using ADDRFAM=NetworkManager in the dispatcher script or b/ the /usr/lib/NetworkManager/ifblacklist_migrate.sh producing a broken/incorrect /e/n/i? I somehow got confused what the actual issue now is, which causes the problems with NFS mounts. For a/ there is already [1], so anything related to that should probably follow up there. What I basically need there, is someone with access to an IPv6 enabled network, who can test a few things for me. For b/, see the recent email I sent regarding netcfg and d-i. We can certainly try to fix the script to no longer produce such broken configuration or/and wait for the patches in d-i to land. What should we do with existing, incorrect /e/n/i files though? I'm very worried doing even more /e/n/i mangling and trying to fixup the file retroactively. What exactly are the consequences of having such a broken configuration? Michael [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=475188 -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
