ddns-updater spamming AdGuard Home DNS with requests

[SnowB0und14]

I got ddns-updater up and running. It works great! However I was checking the logs on my AdGuard Home and noticed that dens-updater was spamming dns requests to the domain specified to be updated. 21,000 times in ~24 hours. I checked the config file and everything is still set to the defaults. I'm running the latest version on TrueNAS Scale. Anything else I could check? I didn't want to file a bug report yet, because I'm not 100% sure it isn't something I did during setup.

[qdm12]

• every minute, the healthcheck resolves your configured domain names and check it matches the last public ip address in the internal database (updates.json)

  ddns-updater/Dockerfile

  Line 68 in 71191fa

  HEALTHCHECK --interval=60s --timeout=5s --start-period=10s --retries=2 CMD ["/updater/ddns-updater", "healthcheck"]

  ddns-updater/internal/health/check.go

  Line 42 in 71191fa

  lookedUpNetIPs, err := resolver.LookupIP(ctx, "ip", hostname)

• every period (5 min by default), it resolves the domain names configured to check it matches the current public ip address

  ddns-updater/internal/update/service.go

  Line 217 in 71191fa

  recordIPv4s, recordIPv6s, err := s.lookupIPsResilient(ctx, hostname, tries)

That should sum up to 1728 resolutions per 24h: pretty high, but also nearly "free" to do in order to operate correctly. Now why you have 21000 I don't know. Check in your adguard logs how periodic it is perhaps?

[SnowB0und14]

@qdm12 Thanks for your response! It does 2 requests every 10 seconds. One to a A record and one to a AAAA. Could it not be happy that AdGuard Home is sending it cached responses, I believe it keeps a cache for the predetermined ttl in the response.

EDIT: I just noticed a setting called HTTP timeout and the default in TrueNAS is 10s. That leads me back to, maybe ddns-updater doesn't want/like the cached DNS records from AdGuard Home?

EDIT2: I disabled caching on AdGuard Home, no change.

[qdm12]

Can you share your settings tree at the top of your logs?

[SnowB0und14]

2024-12-25 06:13:14.608574+00:00├── HTTP client
2024-12-25 06:13:14.608582+00:00|   └── Timeout: 1m0s
2024-12-25 06:13:14.608594+00:00├── Update
2024-12-25 06:13:14.608601+00:00|   ├── Period: 5m0s
2024-12-25 06:13:14.608607+00:00|   └── Cooldown: 5m0s
2024-12-25 06:13:14.608613+00:00├── Public IP fetching
2024-12-25 06:13:14.608619+00:00|   ├── HTTP enabled: yes
2024-12-25 06:13:14.608625+00:00|   ├── HTTP IP providers
2024-12-25 06:13:14.608634+00:00|   |   └── all
2024-12-25 06:13:14.608641+00:00|   ├── HTTP IPv4 providers
2024-12-25 06:13:14.608647+00:00|   |   └── all
2024-12-25 06:13:14.608654+00:00|   ├── HTTP IPv6 providers
2024-12-25 06:13:14.608660+00:00|   |   └── all
2024-12-25 06:13:14.608669+00:00|   ├── DNS enabled: yes
2024-12-25 06:13:14.608675+00:00|   ├── DNS timeout: 3s
2024-12-25 06:13:14.608684+00:00|   └── DNS over TLS providers
2024-12-25 06:13:14.608690+00:00|       └── all
2024-12-25 06:13:14.608696+00:00├── Resolver: use Go default resolver
2024-12-25 06:13:14.608705+00:00├── Server
2024-12-25 06:13:14.608711+00:00|   ├── Listening address: :30007
2024-12-25 06:13:14.608717+00:00|   └── Root URL: /
2024-12-25 06:13:14.608723+00:00├── Health
2024-12-25 06:13:14.608729+00:00|   └── Server listening address: 127.0.0.1:9999
2024-12-25 06:13:14.608739+00:00├── Paths
2024-12-25 06:13:14.608745+00:00|   ├── Data directory: /updater/data
2024-12-25 06:13:14.608752+00:00|   ├── Config file: /updater/data/config.json
2024-12-25 06:13:14.608758+00:00|   └── Umask: --------w-
2024-12-25 06:13:14.608764+00:00├── Backup
2024-12-25 06:13:14.608773+00:00|   ├── Period: 24h0m0s
2024-12-25 06:13:14.608780+00:00|   └── Directory: /updater/data
2024-12-25 06:13:14.608786+00:00└── Logger
2024-12-25 06:13:14.608792+00:00├── Level: info
2024-12-25 06:13:14.608798+00:00└── Caller: hidden

[SnowB0und14]

I opened an issue on the TrueNAS side I found a few environmental values that don't change in the settings tree when you modify them. I suspect the issue is there.

I had a separate question that is related to the issue I’m having. Could it be that TrueNAS is setting up IPv4, IPV6 and IPV4/6 modes for the update, that could be why it’s trying so many times? After reading the docs it looks like you’re supposed to pick one?

[SnowB0und14]

Ok so a little update. It looks like there was a config issue on the TrueNAS app side. Looks like they are fixing that now Issue 1250

I discovered this because I was trying to reduce the DNS requests to 1 a minute by increasing the rate limit from the default 3s to 1m and it wasn't reflecting in the settings tree.

[SnowB0und14]

I updated to the latest version on TureNAS and the DNS timeout config issue is fixed. Still makes 2 requests every 10 seconds. Anything else I could check?

[SnowB0und14]

Could it be with how the TrueNAS container is set up? Is there anything I could check with that?

[dthreee]

Have you found any solution to this? I am also seeing DNS requests every 10 seconds despite what values I put in the update period, timeout, etc. I am running a Pi-hole and the logs are just overfilled with the same requests coming from the ddns-updater. I don't need such frequent updates as my ip rarely changes, usually only when my internet goes down.

[SnowB0und14]

@dthreee what are you running ddns-updater on? I'm trying to find out if it's a TrueNAS implementation of docker issue or if its something else.

[dthreee]

@SnowB0und14
TrueNAS Scale, ElectricEel-24.10.1
ddns-updater installed via the Apps panel.

[SnowB0und14]

@dthreee Here is a link to the issue I opened up over on the TrueNAS GitHub. They closed it because they fixed the environmental variable issue I found. I think there is still an underlying problem with TrueNASs implementation of ddns-updater, as I'm still getting 2 DNS requests every 10 seconds. You helped me narrow down that its not adguard home causing the issue.

[SnowB0und14]

@dthreee
Did you ever figure this out? The latest update on the TrueNAS side didn't change anything.

[SnowB0und14]

Someone over on the TrueNAS issue I had opened, mentioned docker health checks. The health check timing coincides with the excessive DNS queries I was seeing (every 10 seconds). Should I open an issue over here for that? I don't want to bog anyone down with something that's not for-sure a bug.

EDIT: it looks like as I was typing this there are some proposed changes on the TrueNAS side of things. Maybe that will fix it!