On Wed, Feb 09, 2011 at 12:22:58PM +0100, Martin Pelikan wrote: > On Wed, Jan 26, 2011 at 04:44:30PM +0100, Martin Pelikan wrote: > > Hello tech@, claudio@ in particular, > > this diff corrects the behavior of ospf6d when: > > - vlan/vether/? interface is added with some inet6 address > > - a configured interface is removed (please test with usb devs) > > In both cases the daemon crashes. Also, it contains various fixes > > > > Desired behavior: > > - remove if -> withdraw all its routes, but keep running > > - add previously configured if -> announce all its routes back > > - add unconfigured if -> get everyone to know about it without crashing > > Even in the recent -beta one can crash ospf6d easily, just by typing > # ifconfig vlan7 vlan 7 vlandev <your-if> > even in the simplest of possible setups. > > I run this diff on a production router where vlans are being added > dynamically from admin's web page, without any problem for two weeks. > Is there any chance people won't have to patch tne 4.9-release? :-) > What's the problem? Is it too complicated or have I missed something > again? > > Thanks for any feedback.
I would say that this got delayed because claudio was (or still is?) on vacation. I could review this. However, it's very close to release now so I won't bother with it now unless a couple of people have tested your diff to see if there are any unexpected side effects. Given the lack of responses it seems nobody else has tested it? I don't have an ospf6 test setup anymore, and never had one of reasonable size (only virtual machines which became useless for several months while qemu was broken). So for me, it's important that a lot of ospf6d users test diffs in their production setup (or close approximations of that setup). Especially while claudio is away... So I suppose that this will be looked at after release, and those running ospf6d on -current will get to test it then. Sorry.
