martin f krafft wrote:
Okay, this helped me figure out the problem. I was running dnsmasq from an ifupdown up hook, and by the time it ran, the iface state was indeed still tentative, even though it got properly cleared a few seconds later. Inserting a sleep 5 before seems to fix this.
Cool!
I do wonder if it wouldn't be possible to properly fix this in dnsmasq itself? E.g. identify the tentative state, retry n times every second and only give up (and fail, instead of just silently starting without binding) only if it doesn't yield the desired result?
Yes, that would be a much better solution than the ignore-the-error bandaid that got put in long ago without really understanding the problem.
Thanks a lot, Simon, for your patience and sharpness in spotting the bug!
Ditto, we've enabled improvement in dnsmasq too. Cheers, Simon.
-- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
