Remove the following lines from the helm chart template to fix the problem.

This is related to the following issue of the underlaying docker image.
pschiffe/docker-pdns#50

I'm looking the values.yaml file and I have many questions (maybe most of them are dumb but please don't blame me).

1. In the rows 21-31 there is:

image:
  repository: pschiffe/pdns-mysql
  tag: alpine
  pullPolicy: IfNotPresent
domain: external.local
master: "yes"
api: "yes"
version_string: "anonymous"
default_ttl: "1500"
soa_minimum_ttl: "1200"
default_soa_name: "ns1.external.local"

Are they duplicated?

2. In the docker repository pschiffe/pdns-mysql is written that pdns come out without mysql server. Why is used 127.0.0.1 as default database in powerdns instead of the deployed one (mariadb) with pvc?
3. In admin same question as above, why is 127.0.0.1 used for powerdns_host? Shouldn't be used the service of powerdns instead? Furthermore in the readme there is powerdnsadmin.powerdns_host which is wrong because in the values.yaml the value is (correctly) set to the service (though has a different name but I see in the deploy a coalesce)

Yesterday I tried this chart and many errors come out about databases, the admin panel received the error 400 too (I think for wrong host api). This is why today I read the default values in the repo and seems they are wrong.
Furthermore, editing the deployment of powerdns from db host 127.0.0.1 to the service, seems the mariadb was only populated for powerdns-admin and not for powerdns (there wasn't those default tables ), but maybe using the right configuration it is automatically performed somewhere (though I can't find where). This must be performed else every pdnsutil command in powerdns-deployment.yaml will fail.

Finally seems there are many values which could create problems if changed (for example the pvc just takes care of powerdns.mysql_user ignoring the admin.mysql_user value), but this is just for info, not really important.

Monday I will try to perform these change, but I would like to know if I am wrong.

Thank you

Edit: just found out the admin.powerdns_host value is never used, so it shouldn't cause the error 400. I will investigate

Hello,

Is this helm chart supposed to work out of the box, or do i need to do something else than installing it with helm? I've tried both test and stable, but i always get issues with login to the database:
[Warning] Access denied for user 'powerdns'@'127.0.0.1'

I've tried to use --set to define the password for the database, but it seems to be the same result. i did this just for testing:
➜ ~ helm install powerdns --set powerdns.mysql_rootpass=test,powerdns.mysql_pass=test,mariadb.mysql_rootpass=test,mariadb.mysql_pass=test,powerdnsadmin.mysql_pass=test https://raw.githubusercontent.com/aescanero/helm-powerdns/master/test/powerdns.

Am i supposed to use secrets or configmap to get this to work? Or what am i missing here?

An error occurs:

root@master0:/# kubectl -n powerdns logs pod/powerdns-5487d4fc86-n85tz
Sep 17 11:23:11 Listening on controlsocket in '/var/run/pdns.controlsocket'
Sep 17 11:23:11 Guardian is launching an instance
Sep 17 11:23:11 Loading '/usr/lib/pdns/pdns/libgmysqlbackend.so'
Sep 17 11:23:11 This is a guarded instance of pdns
Sep 17 11:23:11 Fatal error: Trying to set unknown parameter 'innodb-read-committed'
Sep 17 11:23:12 Our pdns instance exited with code 1, respawning
Sep 17 11:23:13 Guardian is launching an instance
Sep 17 11:23:13 Loading '/usr/lib/pdns/pdns/libgmysqlbackend.so'
Sep 17 11:23:13 This is a guarded instance of pdns
Sep 17 11:23:13 Fatal error: Trying to set unknown parameter 'innodb-read-committed'
Sep 17 11:23:14 Our pdns instance exited with code 1, respawning
Sep 17 11:23:15 Guardian is launching an instance
Sep 17 11:23:15 Loading '/usr/lib/pdns/pdns/libgmysqlbackend.so'
Sep 17 11:23:15 This is a guarded instance of pdns
Sep 17 11:23:15 Fatal error: Trying to set unknown parameter 'innodb-read-committed'

Hello,
im using the powerdns helm chart. The worker VM in my environment do not have IP V6 enabled

Im getting this errors:

 Jun 19 21:14:11 This is a guarded instance of pdns                                                                                                                                    │
│ Jun 19 21:14:11 UDP server bound to 0.0.0.0:53                                                                                                                                        │
│ Jun 19 21:14:11 IPv6 Address Family is not supported - skipping UDPv6 bind                                                                                                            │
│ Jun 19 21:14:11 TCP server bound to 0.0.0.0:53                                                                                                                                        │
│ Jun 19 21:14:11 Fatal error: Unable to acquire TCPv6 socket: Address family not supported by protocol                                                                                 │
│ Jun 19 21:14:12 Our pdns instance exited with code 1, respawning                                                                                                                      │
│ Jun 19 21:14:13 Guardian is launching an instance                                                                                                                                     │
│ Jun 19 21:14:13 Loading '/usr/lib/pdns/pdns/libgmysqlbackend.so'                                                                                                                      │
│ Jun 19 21:14:13 This is a guarded instance of pdns                                                                                                                                    │
│ Jun 19 21:14:13 UDP server bound to 0.0.0.0:53                                                                                                                                        │
│ Jun 19 21:14:13 IPv6 Address Family is not supported - skipping UDPv6 bind                                                                                                            │
│ Jun 19 21:14:13 TCP server bound to 0.0.0.0:53                                                                                                                                        │
│ Jun 19 21:14:13 Fatal error: Unable to acquire TCPv6 socket: Address family not supported by protocol                                                                                 │

Is there a way to limit powerdns 53 exposure on IP V4 only ?

Hey -- I've been running into this issue whereby PowerDNS does not restart properly after initial config. I get the following error, fix has been to delete the mariadb and powerdns containers and let it re-instantiate

May 18 13:31:45 Guardian is launching an instance
May 18 13:31:45 Loading '/usr/lib/pdns/pdns/libgmysqlbackend.so'
May 18 13:31:45 This is a guarded instance of pdns
May 18 13:31:45 UDP server bound to 0.0.0.0:53
May 18 13:31:45 UDPv6 server bound to [::]:53
May 18 13:31:45 TCP server bound to 0.0.0.0:53
May 18 13:31:45 TCPv6 server bound to [::]:53
May 18 13:31:45 PowerDNS Authoritative Server 4.2.1 (C) 2001-2019 PowerDNS.COM BV
May 18 13:31:45 Using 64-bits mode. Built using gcc 9.2.0 on Dec  3 2019 17:17:37 by buildozer@build-3-11-x86_64.
May 18 13:31:45 PowerDNS comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it according to the terms of the GPL version 2.
May 18 13:31:45 [webserver] Listening for HTTP requests on 0.0.0.0:8081
May 18 13:31:45 Polled security status of version 4.2.1 at startup, no known issues reported: OK
May 18 13:31:45 Master/slave communicator launching
May 18 13:31:45 Creating backend connection for TCP
May 18 13:31:45 About to create 3 backend threads for UDP
May 18 13:31:45 Exiting because communicator thread died with error: GSQLBackend unable to retrieve list of master domains: Could not prepare statement: select d.id, d.name, d.notified_serial, r.content from records r join domains d on r.name=d.name where r.type='SOA' and r.disabled=0 and d.type='MASTER': Table 'powerdns.records' doesn't exist
May 18 13:31:45 Done launching threads, ready to distribute questions
May 18 13:31:46 Guardian is killed, taking down children with us

I'm seeing this when I install the helm chart with crash loop back-off

helm install powerdns-release -f values.yaml aecharts/powerdns

using the values file from your repo and just changing the domain

❯ kgl powerdns-release-6c85b7898f-cpwwp powerdns-powerdnsadmin
===> Waiting for 127.0.0.1 MySQL service
Connection to 127.0.0.1 3306 port [tcp/mysql] succeeded!
Using base prefix '/usr'
New python executable in /opt/pdnsadmin/flask/bin/python3
Not overwriting existing python script /opt/pdnsadmin/flask/bin/python (you must use /opt/pdnsadmin/flask/bin/python3)
Installing wheel...done.
===> DB management
---> Running DB Migration
Usage: flask db migrate [OPTIONS]

Error: Could not locate a Flask application. You did not provide the "FLASK_APP" environment variable, and a "wsgi.py" or "app.py" module was not found in the current directory.
Usage: flask db upgrade [OPTIONS] [REVISION]

Error: Could not locate a Flask application. You did not provide the "FLASK_APP" environment variable, and a "wsgi.py" or "app.py" module was not found in the current directory.
===> Update PDNS API connection info
Connection problem (1146, "Table 'powerdns.setting' doesn't exist")
Connection problem (1146, "Table 'powerdns.setting' doesn't exist")
[2020-06-19 08:25:17 +0000] [23] [INFO] Starting gunicorn 19.7.1
[2020-06-19 08:25:17 +0000] [23] [INFO] Listening at: http://0.0.0.0:9191 (23)
[2020-06-19 08:25:17 +0000] [23] [INFO] Using worker: sync
[2020-06-19 08:25:17 +0000] [25] [INFO] Booting worker with pid: 25
[2020-06-19 08:25:17 +0000] [25] [ERROR] Exception in worker process
Traceback (most recent call last):
File "/usr/lib/python3.7/site-packages/gunicorn/arbiter.py", line 578, in spawn_worker
worker.init_process()
File "/usr/lib/python3.7/site-packages/gunicorn/workers/base.py", line 126, in init_process
self.load_wsgi()
File "/usr/lib/python3.7/site-packages/gunicorn/workers/base.py", line 135, in load_wsgi
self.wsgi = self.app.wsgi()
File "/usr/lib/python3.7/site-packages/gunicorn/app/base.py", line 67, in wsgi
self.callable = self.load()
File "/usr/lib/python3.7/site-packages/gunicorn/app/wsgiapp.py", line 65, in load
return self.load_wsgiapp()
File "/usr/lib/python3.7/site-packages/gunicorn/app/wsgiapp.py", line 52, in load_wsgiapp
return util.import_app(self.app_uri)
File "/usr/lib/python3.7/site-packages/gunicorn/util.py", line 352, in import_app
import(module)
ModuleNotFoundError: No module named 'app'

helm install powerdns \   19:56  27.02.20  56.14G RAM
--set domain=domain.com
https://raw.githubusercontent.com/aescanero/helm-powerdns/master/test/powerdns.tgz
Error: failed to download "https://raw.githubusercontent.com/aescanero/helm-powerdns/master/test/powerdns.tgz" (hint: running helm repo update may help)

Hello @aescanero!

I would like to contribute to your chart. Would you like to add a license file to your repo?

It would be nice if you'd decide use the same MIT license as in powerdns-admin.

Regards

Alex

Hi,

I try using this Chart with custom image repositories (I need them, because my k8s runs on ARM64 devices). The documented configuration parameters powerdns.image.repository and powerdnsadmin.image.repository do not work. Looking into the chart I figured out that powerdns.image.repository needs to be changed into image.repository and powerdnsadmin.image.repository into admin.image.repository.

Furthermore the default value for mariadb.persistence.enabled changed from true to false which is not reflected in the documentation.

Hi,

I tried to set up powerdns on my (ARM64 based) k3s Cluster, following https://www.disasterproject.com/powerdns-overkubernetes-helm-v3/
When I run helm install pdns aecharts/powerdns all seems to run well but all containers except the mariadb one in that pod keep crashing. When I try to take a look at the logs (kubectl logs -p podname containername) I only get the following:
standard_init_linux.go:211: exec user process caused "exec format error"

Could this be due to my cluster running on Arm64?

As I'm fairly new to this, any help on this would be highly appreciated

Hi,
when I tried deploying the chart, the admin app could not connect to the api.
Finally I had a look powerdns-release pod logs.
It showed something along the lines of invalid option innodb-read-committed.
The powerdns docs have the name of the option as 'gmysql-innodb-read-commited'

After changing line 76 of powerdns-deployment.yaml to - name: PDNS_gmysql_innodb_read_committed, I was able to get a running system.

Any thoughts on how to proceed?

Thanks
Christian

I applied with helm.
helm install powerdns-release -f values.yaml aecharts/powerdns
and I got this error.

I can access database service ip through worker node. Someone please help me?