Closed ThaDaVos closed 7 months ago
Hi there, you need to modify your unbound.conf
and remove the path of the chroot parameter or just comment the line out using a of the entry: #
in frontchroot=""
Ah, I guessed it would be something simple - but couldn't find it... thanks for the heads up
Will try when I get home - had to quickly disassemble my whole DNS architecture to get it running again remotely
It sounds like you use watchtower. I suggest better not.
Here's a good explanation why on the Pi-hole Github readme
I hope I was able to solve your issue. Congrats for opening the 50th issue! 🎉
I'm here if you need more assistance.
...by the way, if you have a username="_unbound"
in your unbound.conf
comment this out remove this as well...
Sadly both don't resolve the problem, error messages changes though - the path has now "unbound.d" appended
I have built a new image, are you on 1.19.1-2 already? If so, please show your compose and configs @ThaDaVos
Yeah I've pulled the latest just a few minutes ago, my compose:
I see that username
is uncommented again - but I tried commenting it out
Uhm... I hope your file isn't loaded? https://github.com/madnuttah/unbound-docker/blob/main/unbound/root/usr/local/unbound/unbound.conf
Something overrides the unbound.conf. Can you comment it out remove it in the compose for testing with my vanilla config? My def. config doesn't load files via include.
It's important that port 53 ain't used as well as localhost. Chroot and username should be empty.
I just renamed mine and copied yours into it's place - now it boots but says this:
Feb 14 19:16:10 unbound[1:0] warning: unbound is already running as pid 1.
Feb 14 19:16:10 unbound[1:0] notice: init module 0: validator
Feb 14 19:16:10 unbound[1:0] notice: init module 1: iterator
Feb 14 19:16:10 unbound[1:0] info: start of service (unbound 1.19.1).
Feb 14 19:16:12 unbound[1:0] error: could not open /iana.d/root.zone.tmp1: No such file or directory
So there's a difference between yours and mine which causes the issue
Got it fixed - chroot
has to be ""
and not commented out
Ok, good to know. So everything is fine again?
Hey @ThaDaVos, is it running as expected for you again?
@madnuttah Just to be sure, chroot: "" AND username: "", correct?
Yes @nick-diama. Please open a new issue in case of further problems. Thanks.
@madnuttah The container now reports as unhealty but I am unable to find logs.
Hey @ThaDaVos, is it running as expected for you again?
It's running now, except that it doesn't get healthy (using the new healthcheck script mentioned in https://github.com/madnuttah/unbound-docker/issues/49)
But I haven't waited long to check - but after a few minutes it was still on "Starting"
I swear, I've changed the healthcheck script in my local git and pushed the commits. There's the wrong healthcheck script, though. I'll upload a fix.
So @ThaDaVos, it's done. 🤞
~Yeah it's done~
I'll check at home if the healthcheck works
Take your time and thanks for your patience.
Hi, can the issue be closed @ThaDaVos?
It takes a moment, but it gets healthy
Describe the bug A clear and concise description of what the bug is.
Latest pushed image crashes with
fatal error: unable to chroot to /usr/local/unbound: Operation not permitted
Whole mounted folder is chowned with 1000:1000 - used to work before, but new image just fails to startTo Reproduce Steps to reproduce the behavior:
Expected behavior A clear and concise description of what you expected to happen.
Continueing to work
Screenshots If applicable, add screenshots to help explain your problem.
Please complete the following information:
Additional context Add any other context about the problem here.