On Sun, 2021-01-03 at 09:44 -0500, Sam Varshavchik wrote: > Michael H. Warfield writes: > > > On Sat, 2021-01-02 at 19:57 -0500, Sam Varshavchik wrote: > > > Michael H. Warfield writes: > > > > Fedora 34 and the Fedora 33). All on the same network. All > > > > with > > > > the > > > > same name servers for resolution. > > > > /etc/resolv.conf > > > > -- > > > > # Generated by NetworkManager > > > > search wittsend.com > > > > nameserver 8.8.8.8 > > > > nameserver 1.1.1.1 > > > > nameserver 10.205.38.3 > > > > -- > > > Is /etc/resolv.conf a symlink, and if so where is it pointing. > > > > NO! That's the first thing I checked and FIXED. Again, same thing > > in > > F32 and F33. What's changed? > > If it weren't for the fact that you've confirmed that you fixed this, > this > would be the prime suspect. Even though it's shown contents don't > indicate > the system is pointing to its proxy, who knows what that beast does > sometimes. Maybe it swipes the real system resolver at some point, > and > writes it out for its own, so the symlink points to its copy of > networkmanager's. And since all services – as I understand – run in > their > own container it's possible that the misbehaving services' containers > have > resolv.conf pointing to the systemd proxy. It looks like the "resolve [!UNAVAIL=return]" stanza on the "hosts:" line in /etc/nsswitch.con might be the guilty party. Ed Greshko pointed this out to me. Because it was a "long period" failure, I'm going to test for a few days before passing judgement but it's looking good so far, since I removed that stanza. It directly relates to the systemd resolver which is not running on any of those systems. > On one of my servers I have named-chroot running, its ipv4.dns is set > to its > own IP address, so networkmanager's resolv.conf writes out > "nameserver" with > its own IP address, where bind happily lives. I had to turn off > systemd- > resolved after it started to periodically and repeatedly TEMPFAIL > everything, after some period of uptime. I don't understand how > that > fracking thing can lose track of bind running on the same server. Par for the course I'm afraid... Thanks everyone for the feedback and clues. I've at least got my thread to pick at. I'll report back in a few days. Regards, Mike -- Michael H. Warfield (AI4NB) | (o) +1 706 850-8773 | mhw@xxxxxxxxxxxx /\/\|=mhw=|\/\/ | (c) +1 678 463-0932 | http://www.wittsend.com/mhw/ ARIN whois: MHW9-ARIN | An optimist believes we live in the best of all PGP Key: 0xC0EB9675674627FF | possible worlds. A pessimist is sure of it!
Attachment:
signature.asc
Description: This is a digitally signed message part
_______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@xxxxxxxxxxxxxxxxxxxxxxx
