[SUPPORT] pihole for unRaid - Spants repo


Recommended Posts

4 hours ago, mintjberry said:

Terrible update, can we roll back?

 

According to this

https://github.com/diginc/docker-pi-hole/releases

 

you SHOULD be able to edit the container to have this as the repository

diginc/pi-hole:v3.3

 

I'm not 100% sure that will work, basing this off of my Ombi container that is linuxserver/ombi:65

from here: https://github.com/linuxserver/docker-ombi/releases

 

As I'm running the version listed as tag 65 due to a bug introduced after that.

 

EDIT: To add to this I would suggest you submit bug reports with logs so they can get this corrected, as this container is simply a template to use diginc's pihole container.

 

https://github.com/diginc/docker-pi-hole/issues -- report issues here.

Edited by CorneliousJD
Link to comment
12 hours ago, WannabeMKII said:

Just to add, I've had to kill the docker as it was using huge amounts of CPU and also crashing the unraid GUI.

 

Anyone else having the same issue, or is it isolated to me?

 

Hoping for a fix, or is there a way to roll back?

 

The update I received overnight appears to have fixed everything for me. Thanks for the prompt resolution!

Link to comment

So just to verify, is it possible to run Pihole docker in bridge mode on the unRAID host IP?  I assume I would need to assign the unRAID GUI to a port other than 80, correct? 

 

I really do not want to run it that way, but, I am unable to assign *any* docker its own IP address without generating lots of ip/macvlan related call traces in the syslog.  Pihole was running on its own IP address until it generated over a dozen call traces yesterday and completely locked up my server overnight making my whole network inaccessible this morning.

 

I have since disabled Pihole and things have settled down (no more call traces).

Link to comment
33 minutes ago, Hoopster said:

So just to verify, is it possible to run Pihole docker in bridge mode on the unRAID host IP?  I assume I would need to assign the unRAID GUI to a port other than 80, correct? 

 

I really do not want to run it that way, but, I am unable to assign *any* docker its own IP address without generating lots of ip/macvlan related call traces in the syslog.  Pihole was running on its own IP address until it generated over a dozen call traces yesterday and completely locked up my server overnight making my whole network inaccessible this morning.

 

I have since disabled Pihole and things have settled down (no more call traces).

i just made a debian vm and running this way

 

 

Link to comment
9 minutes ago, ijuarez said:

i just made a debian vm and running this way

 

 

 

This is also how I used to run it. I just had a debian VM setup with command line only and no GUI for being as lightweight as possible.

It worked well, but I've since migrated to the docker container and I run it with its own IP... sounds like you have some deeper issue if running a container with its own IP is causing call traces. 

 

Post diagnostics and logs in the general support forums maybe? If you can get that sorted out then this docker container should work for you.

Link to comment
23 minutes ago, CorneliousJD said:

 

This is also how I used to run it. I just had a debian VM setup with command line only and no GUI for being as lightweight as possible.

It worked well, but I've since migrated to the docker container and I run it with its own IP... sounds like you have some deeper issue if running a container with its own IP is causing call traces. 

 

Post diagnostics and logs in the general support forums maybe? If you can get that sorted out then this docker container should work for you.

 

 

I had the docket running and rebooted to address another issue. somehow the docker continue to work but since it ran on port 80 the unraid gui was inaccessible and that did not work for me. I don't to want to change the default unraid ports so a vm is good for me.

Link to comment
1 hour ago, CorneliousJD said:

Post diagnostics and logs in the general support forums maybe? If you can get that sorted out then this docker container should work for you.

 

I have posted about this several times in general support and defects boards and uploaded diagnostics.  bonienl (docker networking author) says he cannot reproduce the problem and suggested it may be specific to my hardware.  Limetech says it is not a general problem with unRAID.  Yesterday another user posted about the same issue, call traces with ip/macvlan, while using Pihole with its own IP.  He has very different hardware.

 

 

Docker container worked for a little over a week without issue (despite generating call traces during that time).  However, it started generating numerous call traces yesterday and totally locked up last night.

Edited by Hoopster
Link to comment
17 hours ago, Hoopster said:

 

I have posted about this several times in general support and defects boards and uploaded diagnostics.  bonienl (docker networking author) says he cannot reproduce the problem and suggested it may be specific to my hardware.  Limetech says it is not a general problem with unRAID.  Yesterday another user posted about the same issue, call traces with ip/macvlan, while using Pihole with its own IP.  He has very different hardware.

 

That's super weird. I have a Supermicro server and haven't had any call traces like that so that's really odd, sorry you're having such trouble. Hopefully more people report it if/when it occurs so that it can get the attention it needs then. 

 

In the meantime I would suggest setting up a quick debain VM, if you're not familiar with the linux terminal you can install a GUI and still run pihole with a few simple commands on the terminal. The GUI will help you with things like setting a static IP for the VM, etc. 

 

This is what I was doing up until I decided to give the docker container another go and got that working so I killed off my VM.

Link to comment
2 hours ago, CorneliousJD said:

 

In the meantime I would suggest setting up a quick debain VM

 

Yes, a VM is certainly a possibility.  I prefer to use docker containers, when possible, simply to avoid the need to have a VM running 24x7.  I am still experimenting with potential solutions to the call trace problem by trying different docker networking settings, NICs, etc. trying to find the source of the call traces.  I hate unsolved mysteries :)

Link to comment
4 hours ago, Hoopster said:

 

Yes, a VM is certainly a possibility.  I prefer to use docker containers, when possible, simply to avoid the need to have a VM running 24x7.  I am still experimenting with potential solutions to the call trace problem by trying different docker networking settings, NICs, etc. trying to find the source of the call traces.  I hate unsolved mysteries :)

 

I'm with you! I know we're staying off the pihole container support here but willing to help provide logs/sepcs/etc from my build that's not having any issues if needed anywhere, feel free to PM me if ever needed to drop more logs into something. I have a 4U supermicro w/ 4x onboard gigabit NIC ports.

Link to comment
16 minutes ago, spants said:

It is strange! I used to run it on the unraid ip with the unraid port changed. I then moved to separate ips and that works great for me. I do have two nics installed.... 

 

Yeah, this is not Pihole related at all as the call traces occur with any docker to which I assign an IP.  So far, however (fingers crossed), I have been running the Pihole docker on its own IP for 9 hours without a call trace.  I disabled LAN1 motherboard NIC in the BIOS and enabled LAN2 NIC.

Link to comment
45 minutes ago, rmilyard said:

I am also getting call traces running PiHole.  I had docker disabled for 24hrs and no calls.  Turned back on docker and had some within 10 mins.

 

I am almost certain this is not a Pihole issue.  It is just an issue with static IP addresses assigned to a docker.  It could be any docker, not just Pihole.  I have had the same issue with three separate docker containers.  All three generated the same ip/macvlan call traces.  As soon as I disabled the dockers or removed the IP address assignment, the call traces stopped.

 

I have been running Pihole with a static IP address for almost 30 hours now with NO call traces.  Since very few have this issue and lots of unRAID users are successfully running Pihole and other dockers that have given me problems when assigned a static IP address, I assumed the issue had something to do with my hardware/BIOS configuration.

 

I have two integrated Intel NICs on my ASRock C236 WSI motherboard.  LAN1 was eth0 and LAN2 was eth1.  I was getting call traces with LAN1 enabled and serving as the eth0 NIC which unRAID used to connect to the LAN and as br0 for the dockers.  I disabled LAN1, enabled LAN2 (it is now eth0/br0) and disabled every feature in the BIOS I was not using.  Since changing the NIC unRAID uses, I have had no call traces for 30 hours and counting (fingers crossed).

 

Does your MB have two or more NICs?  If so, you might try changing the one unRAID uses and see what that does for your call traces.

Edited by Hoopster
Link to comment

This container seems to be locking up my whole Docker - the Docker page in unraid wouldn't load.  Tried stopping Docker from the settings tab, no luck.

 

SSHing I see this:

 

root@Tower:~# docker ps
CONTAINER ID        IMAGE                   COMMAND             CREATED             STATUS                PORTS               NAMES
62518c9bb052        diginc/pi-hole:latest   "/s6-init"          6 days ago          Up 4 days (healthy)                       pihole

root@Tower:~# docker stop pihole
# Hangs for a few minutes.  Eventually ctrl-c out

root@Tower:~# docker kill pihole
# Hangs forever

 

Link to comment
On 4/16/2018 at 5:09 PM, JustinAiken said:

This container seems to be locking up my whole Docker - the Docker page in unraid wouldn't load.  Tried stopping Docker from the settings tab, no luck.

 

Did you set it up using the SpaceInvader One approach? I had to hard-reboot my server, and after a few days it ended up the same way. I'm wondering if a log or something is causing the container to lock up after a few days.

Link to comment
2 hours ago, coppit said:

 

Did you set it up using the SpaceInvader One approach? I had to hard-reboot my server, and after a few days it ended up the same way. I'm wondering if a log or something is causing the container to lock up after a few days.

I had the same problem. Created and vm running Debian. Also changing the unraid http to something other that 80 fixed all fo my issues. 

Link to comment
17 hours ago, coppit said:

 

Did you set it up using the SpaceInvader One approach? I had to hard-reboot my server, and after a few days it ended up the same way. I'm wondering if a log or something is causing the container to lock up after a few days.

 

Yep, walked through the space invader video.

 

First time I was using the container heavily, got a hard lock.

Next I tried just running it for a few days, but not using it for DNS; so it was running 24/7, but handling 0 requests.

After a few days, the docker container showed as "healthy", but was unkillable.

 

Anecdotally speaking, every hardlock a docker container has ever done to my unRaid has been a container using the s6 system (gogs used to lock mine up occasionally, but it quit using s6, and hasn't locked since)

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.