external-dns updates incomplete and infinite loop when ingress has several external addresses (cloudflare provider)

Created on 26 Aug 2018  路  19Comments  路  Source: kubernetes-sigs/external-dns

I think this issue is the same as described in:
https://kubernetes.slack.com/archives/C771MKDKQ/p1509029114000201

Even if there are no changes, and DNS looks correctly, I see and endless stream of updates:

time="2018-08-26T18:01:17Z" level=debug msg="Endpoints generated from ingress: logging/logging-kibana: [kibana.k8s.davidkarlsen.com 0 IN A 192.168.3.3;192.168.3.4;192.168.3.5 kibana.k8s.davidkarlsen.com 0 IN A 192.168.3.3;192.168.3.4;192.168.3.5]"
time="2018-08-26T18:01:17Z" level=debug msg="Removing duplicate endpoint kibana.k8s.davidkarlsen.com 0 IN A 192.168.3.3;192.168.3.4;192.168.3.5"
time="2018-08-26T18:01:23Z" level=info msg="Changing record." action=UPDATE record=kibana.k8s.davidkarlsen.com ttl=1 type=A zone=c1c51c73b8d326febfe9c5fab78ccf43
time="2018-08-26T18:01:29Z" level=info msg="Changing record." action=UPDATE record=kibana.k8s.davidkarlsen.com ttl=1 type=TXT zone=c1c51c73b8d326febfe9c5fab78ccf43

note "A 192.168.3.3;192.168.3.4;192.168.3.5 "

I think there is some flaw since the record generated from ingress has three addresses, but there is only one (the first one) in DNS:

dig kibana.k8s.davidkarlsen.com

; <<>> DiG 9.11.3-1ubuntu1.1-Ubuntu <<>> kibana.k8s.davidkarlsen.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 55558
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 13, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
; COOKIE: 713665436d83e2f8826fc84c5b82ebe92ef3a082a79cb40f (good)
;; QUESTION SECTION:
;kibana.k8s.davidkarlsen.com.   IN      A

;; ANSWER SECTION:
kibana.k8s.davidkarlsen.com. 299 IN     A       192.168.3.3

;; AUTHORITY SECTION:
com.                    162976  IN      NS      m.gtld-servers.net.
com.                    162976  IN      NS      b.gtld-servers.net.
com.                    162976  IN      NS      h.gtld-servers.net.
com.                    162976  IN      NS      i.gtld-servers.net.
com.                    162976  IN      NS      l.gtld-servers.net.
com.                    162976  IN      NS      a.gtld-servers.net.
com.                    162976  IN      NS      f.gtld-servers.net.
com.                    162976  IN      NS      g.gtld-servers.net.
com.                    162976  IN      NS      e.gtld-servers.net.
com.                    162976  IN      NS      k.gtld-servers.net.
com.                    162976  IN      NS      j.gtld-servers.net.
com.                    162976  IN      NS      d.gtld-servers.net.
com.                    162976  IN      NS      c.gtld-servers.net.

;; Query time: 199 msec
;; SERVER: 192.168.3.2#53(192.168.3.2)
;; WHEN: Sun Aug 26 20:03:19 CEST 2018
;; MSG SIZE  rcvd: 324

note only one A record is in DNS.

And indeed if I change the other externalAddresses for the ingress to only have one address the excessive updates do not happen.

I have tested with v0.5.4 and v0.5.5 - both have the same problem.

picture davidkarlsen

Most helpful comment

This looks interesting, I will take at it, if no one already had

picture ideahitme on 31 Aug 2018
👍4

All 19 comments

This looks interesting, I will take at it, if no one already had

ideahitme picture ideahitme on 31 Aug 2018
👍4

That would be great - I don't have go skills myself - and on Slack maintainers indicated they did not have the time for it

davidkarlsen picture davidkarlsen on 1 Sep 2018

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

fejta-bot picture fejta-bot on 25 Apr 2019

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

fejta-bot picture fejta-bot on 25 May 2019

/remove-lifecycle rotten

davidkarlsen picture davidkarlsen on 25 May 2019

Hello @davidkarlsen, did you find a solution?
I'm hitting the Cloudflare API req limit because of these constant calls.

AndresPineros picture AndresPineros on 14 Aug 2019

Not to my knowledge. This is still a bug. I had to work around it.

davidkarlsen picture davidkarlsen on 14 Aug 2019

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

fejta-bot picture fejta-bot on 12 Nov 2019

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

fejta-bot picture fejta-bot on 12 Dec 2019

Not stale

davidkarlsen picture davidkarlsen on 12 Dec 2019

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

fejta-bot picture fejta-bot on 11 Jan 2020

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

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.

k8s-ci-robot picture k8s-ci-robot on 11 Jan 2020

/reopen

linki picture linki on 14 Jan 2020

@linki: Reopened this issue.

In response to this:

/reopen

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.

k8s-ci-robot picture k8s-ci-robot on 14 Jan 2020

/remove-lifecycle rotten

linki picture linki on 14 Jan 2020

Not to my knowledge. This is still a bug. I had to work around it.

Can you share how you work around it? thanks

eric4545 picture eric4545 on 2 Apr 2020

This should be fixed in 0.7.2, could you confirm?

sheerun picture sheerun on 4 Jun 2020

No response, so I'll consider this fixed. If not, please create another issue with steps to reproduce, or ideally a test in cloudflare_test.go. The tests are really easy to write :)

/close

sheerun picture sheerun on 27 Jun 2020

@sheerun: Closing this issue.

In response to this:

No response, so I'll consider this fixed. If not, please create another issue with steps to reproduce, or ideally a test in cloudflare_test.go. The tests are really easy to write :)

/close

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.

k8s-ci-robot picture k8s-ci-robot on 27 Jun 2020
Was this page helpful?
0 / 5 - 0 ratings

Related issues

wndhydrnt picture wndhydrnt  路  3Comments
szuecs picture szuecs  路  4Comments
Gnnng picture Gnnng  路  3Comments
naveeng68 picture naveeng68  路  4Comments
dryewo picture dryewo  路  3Comments