TXT records with multiple targets are not handled properly

[golx]

What happened:
I have multiple TXT records for one of my domains, some of them are not managed by external-dns TXT registry. This results in TXT endpoints with several targets coming from the DNS provider, but TXT registry just takes the first target.
https://github.com/kubernetes-sigs/external-dns/blob/master/registry/txt.go#L107
This results in either:

• A record not updated by external-dns as it fails to "see" the TXT record it looks for, as the target is not the first one in the list
• or TXT record not related to external-dns removed by external-dns if external-dns managed TXT target is the first one in the list

What you expected to happen:
TXT registry should iterate over TXT targets instead of just picking the first one.

How to reproduce it (as minimally and precisely as possible):
Create a TXT record in addition to the one created by external-dns.

Environment:

• External-DNS version (use external-dns --version): 0.11.1
• DNS provider: DigitalOcean
• Others:

[cuppett]

I think I see this as well in AWS.

I have a zone with multiple, existing TXT entries on the root domain:

        {
            "Name": "example.com.",
            "Type": "TXT",
            "TTL": 86400,
            "ResourceRecords": [
                {
                    "Value": "\"v=spf1 a mx a:completeupdates.com ~all\""
                },
                {
                    "Value": "\"google-site-verification=pKt0HRP...aomM\""
                },
                {
                    "Value": "\"google-site-verification=wsObL_...3gWME-Bj5E\""
                }
            ]
        },

and in the updates it's trying to make I get this in the log:

time="2022-07-30T00:59:57Z" level=info msg="Desired change: CREATE example.com A [Id: /hostedzone/ZZZZZZZZZZZ]"
time="2022-07-30T00:59:57Z" level=info msg="Desired change: CREATE example.com TXT [Id: /hostedzone/ZZZZZZZZZZZ]"
time="2022-07-30T00:59:57Z" level=error msg="Failure in zone example.com. [Id: /hostedzone/ZZZZZZZZZZZ]"
time="2022-07-30T00:59:57Z" level=error msg="InvalidChangeBatch: [Tried to create resource record set [name='example.com.', type='TXT'] but it already exists]\n\tstatus code: 400, request id: f31254aa-a72f-4e80-8008-c6fdda55774e"

[nickmonad]

I believe I am also seeing this issue, or one quite similar. If we have a TXT record created ahead of time (for example, as required by mailgun, when our application needs to send mail, TXT myapp v=spf1 include:mailgun.org ~all), external-dns reports,

level=warning msg="Preexisting records exist which should not exist for creation actions." dnsName=myapp.domain.com domain=domain.com recordType=TXT

and will continue to create new TXT records containing "heritage" information indefinitely at every polling loop.

[pjelar]

I'm seeing this issue using:

Environment:
External-DNS version (use external-dns --version): 0.13.1
DNS provider: DigitalOcean

I only have one nginx ingress.

spec:
ingressClassName: nginx
rules:

• host: api.mydns.com
  http:
  paths:
  • backend:
    service:
    name: myservice
    port:
    number: 8000
    path: /
    pathType: Prefix

The dns is created correctly but on every loop it creates another TXT record and complains that one already exists.

[cuppett]

I found this workaround:

#449 (comment)

Using a --text-prefix that helps disambiguate and separate these out, it creates predictability here (and a few more records).

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle rotten
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Reopen this issue with /reopen
• Mark this issue as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close not-planned

[k8s-ci-robot]

@k8s-triage-robot: Closing this issue, marking it as "Not Planned".

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Reopen this issue with /reopen
• Mark this issue as fresh with /remove-lifecycle rotten
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close not-planned

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.