Bug 1918568 - Seeing `rule:delete_security_group_rule is disallowed by policy` in Kuryr controller after upgrading to Octavia Train (OSP13z13)
Summary: Seeing `rule:delete_security_group_rule is disallowed by policy` in Kuryr con...
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Networking
Version: 3.11.0
Hardware: Unspecified
OS: Unspecified
medium
medium
Target Milestone: ---
: 3.11.z
Assignee: Michał Dulko
QA Contact: Itzik Brown
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2021-01-21 04:31 UTC by Mohammad
Modified: 2024-06-13 23:58 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: No Doc Update
Doc Text:
Clone Of:
Environment:
Last Closed: 2021-04-21 20:21:34 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
Kuryr controller logs (30.14 KB, text/plain)
2021-01-21 04:59 UTC, Mohammad
no flags Details


Links
System ID Private Priority Status Summary Last Updated
Github openshift kuryr-kubernetes pull 463 0 None open Bug 1918568: Fix LB handling with ACLs and ERROR state 2021-02-25 15:53:20 UTC
Github openshift kuryr-kubernetes pull 467 0 None open Bug 1918568: Raise ResourceNotReady when deleting ERROR LB 2021-03-01 12:22:43 UTC
Github openshift kuryr-kubernetes pull 481 0 None open Bug 1918568: Fix TypeError on ERROR LB removal 2021-03-16 10:08:24 UTC
Red Hat Product Errata RHBA-2021:1147 0 None None None 2021-04-21 20:21:40 UTC

Description Mohammad 2021-01-21 04:31:50 UTC
Description of problem: We are seeing `rule:delete_security_group_rule is disallowed by policy` in Kuryr controller after upgrading to Octavia Train (OSP13z13).


Version-Release number of selected component (if applicable): 3.11.306 (and 3.11.272)


How reproducible: Upgrade OSP13 to z13 on two OpenShift clusters using Kuryr with versions 3.11.272 and 3.11.306.


I'll add more details in the BZ after creation.

Comment 1 Mohammad 2021-01-21 04:35:55 UTC
In our environments, there are two different versions of the Kuryr controller:

3.11.272 (which doesn't have support for Octavia Train OSP13z13)
3.11.306 (which does have support for Octavia Train OSP13z13)

These two environments have recently seen their Octavia upgraded as part of the OpenStack OSP13z13 upgrade.

which has led us to see the following errors:
=================================================================================
3.11.272 Kuryr controller (which doesn't know how to handle Octavia Train):

2021-01-20 07:13:54.353 1 ERROR kuryr_kubernetes.controller.drivers.lbaasv2 [-] Failed
when creating security group rule to enable routes for listener
prodnamespace/dart-master:TCP:5001.: NotFound: Security group
43daaf4a-193b-430d-a5bc-477db91a4028 does not exist Neutron server returns request_ids:
['req-5eb0c37f-3e11-4f9e-b73e-b29e12117b70']


AND

-----------------------------------------------------------------------------------
3.11.306 Kuryr controller (which does know how to handle Octavia Train):

kuryr-controller-59f678d96c-zxc4j.logs:2021-01-19 04:25:29.955 1 ERROR kuryr_kubernetes.handlers.logging [-] Failed to handle event {u'object': {u'kind': u'Endpoints', u'subsets': [{u'addresses': [{u'ip': u'XXX.XXX.7.47', u'targetRef': {u'kind': u'Pod', u'resourceVersion': u'74682408', u'namespace': u'testnamespace', u'name': u'redis-ha-server-0', u'uid': u'9a27ba5f-5546-11eb-aeac-fa163ee18f34'}, u'nodeName': u'app-node-4.openshift-test1-dcb1.csda.gov.au'}], u'ports': [{u'protocol': u'TCP', u'name': u'exporter', u'port': 9121}, {u'protocol': u'TCP', u'name': u'server', u'port': 6379}, {u'protocol': u'TCP', u'name': u'sentinel', u'port': 26379}]}], u'apiVersion': u'v1', u'metadata': {u'name': u'redis-ha-announce-0', u'labels': {u'release': u'redis-ha', u'app': u'redis-ha', u'heritage': u'Helm', u'chart': u'redis-ha-4.4.4', u'app.kubernetes.io/managed-by': u'Helm'}, u'namespace': u'testnamespace', u'resourceVersion': u'74682410', u'creationTimestamp': u'2020-07-10T01:43:13Z', u'annotations': {u'openstack.org/kuryr-lbaas-spec': u'{"versioned_object.data": {"ip": "XXX.XXX.185.165", "lb_ip": null, "ports": [{"versioned_object.data": {"name": "server", "port": 6379, "protocol": "TCP", "targetPort": "redis"}, "versioned_object.name": "LBaaSPortSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.1"}, {"versioned_object.data": {"name": "sentinel", "port": 26379, "protocol": "TCP", "targetPort": "sentinel"}, "versioned_object.name": "LBaaSPortSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.1"}, {"versioned_object.data": {"name": "exporter", "port": 9121, "protocol": "TCP", "targetPort": "exporter-port"}, "versioned_object.name": "LBaaSPortSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.1"}], "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "security_groups_ids": ["2cb04b99-ffcc-46f9-8c52-f598973e56b7", "1107252a-5a37-41a5-a5b6-467fc1b21c5b"], "subnet_id": "3dd8da27-8f31-46a6-bce1-67c04eca0a66", "type": "ClusterIP"}, "versioned_object.name": "LBaaSServiceSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}', u'openstack.org/kuryr-lbaas-state': u'{"versioned_object.data": {"listeners": [{"versioned_object.changes": ["id"], "versioned_object.data": {"id": "c6428296-8345-4bcb-9d64-0d28244a0ff0", "loadbalancer_id": "4211d51c-6d3f-4925-9c06-a77070140bdb", "name": "testnamespace/redis-ha-announce-0:TCP:6379", "port": 6379, "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "protocol": "TCP"}, "versioned_object.name": "LBaaSListener", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}, {"versioned_object.changes": ["id"], "versioned_object.data": {"id": "f3ba3229-d81e-43b8-b7d7-53509ec38819", "loadbalancer_id": "4211d51c-6d3f-4925-9c06-a77070140bdb", "name": "testnamespace/redis-ha-announce-0:TCP:26379", "port": 26379, "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "protocol": "TCP"}, "versioned_object.name": "LBaaSListener", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}], "loadbalancer": {"versioned_object.data": {"id": "4211d51c-6d3f-4925-9c06-a77070140bdb", "ip": "XXX.XXX.185.165", "name": "testnamespace/redis-ha-announce-0", "port_id": "9b993cce-b2b3-4d2a-85c4-f52e70cc586b", "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "provider": "octavia", "security_groups": ["2cb04b99-ffcc-46f9-8c52-f598973e56b7", "1107252a-5a37-41a5-a5b6-467fc1b21c5b"], "subnet_id": "3dd8da27-8f31-46a6-bce1-67c04eca0a66"}, "versioned_object.name": "LBaaSLoadBalancer", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.3"}, "members": [{"versioned_object.data": {"id": "d297498a-ee70-44a3-9b67-3108c5be4765", "ip": "XXX.XXX.7.9", "name": "testnamespace/redis-ha-server-0:6379", "pool_id": "aae59e24-5a5c-4204-87e9-78abc3b4743b", "port": 6379, "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "subnet_id": "3dd8da27-8f31-46a6-bce1-67c04eca0a66"}, "versioned_object.name": "LBaaSMember", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}, {"versioned_object.data": {"id": "5952dd1f-47f7-48a9-9b27-cc2244693e7a", "ip": "XXX.XXX.7.9", "name": "testnamespace/redis-ha-server-0:26379", "pool_id": "9e993033-eed8-4c99-9f98-c6b281c6726c", "port": 26379, "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "subnet_id": "3dd8da27-8f31-46a6-bce1-67c04eca0a66"}, "versioned_object.name": "LBaaSMember", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}], "pools": [{"versioned_object.changes": ["id"], "versioned_object.data": {"id": "aae59e24-5a5c-4204-87e9-78abc3b4743b", "listener_id": "c6428296-8345-4bcb-9d64-0d28244a0ff0", "loadbalancer_id": "4211d51c-6d3f-4925-9c06-a77070140bdb", "name": "testnamespace/redis-ha-announce-0:TCP:6379", "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "protocol": "TCP"}, "versioned_object.name": "LBaaSPool", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.1"}, {"versioned_object.changes": ["id"], "versioned_object.data": {"id": "9e993033-eed8-4c99-9f98-c6b281c6726c", "listener_id": "f3ba3229-d81e-43b8-b7d7-53509ec38819", "loadbalancer_id": "4211d51c-6d3f-4925-9c06-a77070140bdb", "name": "testnamespace/redis-ha-announce-0:TCP:26379", "project_id": "89cf74d67cef4c62ab1d1432996afaf1", "protocol": "TCP"}, "versioned_object.name": "LBaaSPool", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.1"}], "service_pub_ip_info": null}, "versioned_object.name": "LBaaSState", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}'}, u'selfLink': u'/api/v1/namespaces/testnamespace/endpoints/redis-ha-announce-0', u'uid': u'b7ba0623-c24e-11ea-8e7d-fa163ecbe187'}}, u'type': u'ADDED'}: Forbidden: rule:delete_security_group_rule is disallowed by policy
kuryr-controller-59f678d96c-zxc4j.logs:2021-01-19 04:25:29.955 1 ERROR kuryr_kubernetes.handlers.logging Forbidden: rule:delete_security_group_rule is disallowed by policy

=================================================================================

Comment 2 Mohammad 2021-01-21 04:37:44 UTC
Introduction

In my lab, I am using Kuryr controller 3.11.306, and I have created a test application:

[openshift@master-0 octaviatrainissue]$ oc get svc
NAME      TYPE        CLUSTER-IP        EXTERNAL-IP   PORT(S)   AGE
echo-09   ClusterIP   XXX.XXX.136.204   <none>        88/TCP    35d
echo-10   ClusterIP   XXX.XXX.144.5     <none>        88/TCP    1d

Service echo-09 was created when Octavia was OSP13z12 (or OSP13z11, I'm not sure) and echo-10 was created with the new Octavia OSP13z13:

echo-09:

$ oc get svc echo-09 -o yaml
apiVersion: v1
kind: Service
metadata:
  annotations:
    openstack.org/kuryr-lbaas-spec: '{"versioned_object.data": {"ip": "XXX.XXX..136.204",
      "lb_ip": null, "ports": [{"versioned_object.data": {"name": "http", "port":
      80, "protocol": "TCP", "targetPort": "8080"}, "versioned_object.name": "LBaaSPortSpec",
      "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version":
      "1.1"}], "project_id": "f7b96553d2fd4e26a05beb87c85c67c9", "security_groups_ids":
      ["adfc0f11-e15d-469e-af16-0c62826074df", "889474b8-4f3c-4d87-a18e-ab90934ca3de"],
      "subnet_id": "e7a1a65f-28c1-4f47-bcbd-bff7b35a3d45", "type": "ClusterIP"}, "versioned_object.name":
      "LBaaSServiceSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version":
      "1.0"}'
  creationTimestamp: "2020-12-16T05:22:48Z"
  labels:
    k8s-app: momoecho-09
  name: echo-09
  namespace: momo
  resourceVersion: "18507"
  selfLink: /api/v1/namespaces/momo/services/echo-09
  uid: bc3cbf29-3f5e-11eb-8692-fa163eee27a8
spec:
  clusterIP: XXX.XXX..136.204
  ports:
  - name: http
    port: 80
    protocol: TCP
    targetPort: 8080
  selector:
    k8s-app: momoecho-09
  sessionAffinity: None
  type: ClusterIP
status:
  loadBalancer: {}


echo-10:

$ oc get svc echo-10 -o yaml
apiVersion: v1
kind: Service
metadata:
  annotations:
    openstack.org/kuryr-lbaas-spec: '{"versioned_object.data": {"ip": "XXX.XXX..144.5",
      "lb_ip": null, "ports": [{"versioned_object.data": {"name": "http", "port":
      80, "protocol": "TCP", "targetPort": "8080"}, "versioned_object.name": "LBaaSPortSpec",
      "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version":
      "1.1"}], "project_id": "f7b96553d2fd4e26a05beb87c85c67c9", "security_groups_ids":
      ["adfc0f11-e15d-469e-af16-0c62826074df", "889474b8-4f3c-4d87-a18e-ab90934ca3de"],
      "subnet_id": "e7a1a65f-28c1-4f47-bcbd-bff7b35a3d45", "type": "ClusterIP"}, "versioned_object.name":
      "LBaaSServiceSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version":
      "1.0"}'
  creationTimestamp: "2021-01-20T02:55:00Z"
  labels:
    k8s-app: momoecho-10
  name: echo-10
  namespace: momo
  resourceVersion: "8432118"
  selfLink: /api/v1/namespaces/momo/services/echo-10
  uid: e2fa45d8-5aca-11eb-b6de-fa163e90af27
spec:
  clusterIP: XXX.XXX..144.5
  ports:
  - name: http
    port: 80
    protocol: TCP
    targetPort: 8080
  selector:
    k8s-app: momoecho-10
  sessionAffinity: None
  type: ClusterIP
status:
  loadBalancer: {}


Tests
=====

1. Change source port for both services from 80 to 88.
2. Change target port for both services from 8080 to 8080 (obviously the ports don't exist on the pods)

What appears in the logs:

Logs are showing:

=========================================================================================
[openshift@master-0 ~]$ oc logs -f kuryr-controller-5676f699f5-2c98h -n kuryr |grep echo

2021-01-21 04:15:10.604 1 WARNING kuryr_kubernetes.controller.drivers.lbaasv2 [-] Cannot find SG rule for 8e82fc5e-0e40-43f8-a230-b23dddc2967d (momo/echo-10:TCP:80) listener.
2021-01-21 04:30:00.769 1 ERROR kuryr_kubernetes.handlers.logging [-] Failed to handle event {u'object': {u'kind': u'Endpoints', u'subsets': [{u'addresses': [{u'ip': u'XXX.XXX..3.224', u'targetRef': {u'kind': u'Pod', u'resourceVersion': u'8168688', u'namespace': u'momo', u'name': u'echo-09-56fcff9f5d-z9bjt', u'uid': u'd123c963-553d-11eb-b6de-fa163e90af27'}, u'nodeName': u'app-node-1.openshift-dev2-dca1.csda.gov.au'}], u'ports': [{u'protocol': u'TCP', u'name': u'http', u'port': 8088}]}], u'apiVersion': u'v1', u'metadata': {u'name': u'echo-09', u'labels': {u'k8s-app': u'momoecho-09'}, u'namespace': u'momo', u'resourceVersion': u'8434389', u'creationTimestamp': u'2020-12-16T05:22:48Z', u'annotations': {u'openstack.org/kuryr-lbaas-spec': u'{"versioned_object.data": {"ip": "XXX.XXX..136.204", "lb_ip": null, "ports": [{"versioned_object.data": {"name": "http", "port": 88, "protocol": "TCP", "targetPort": "8088"}, "versioned_object.name": "LBaaSPortSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.1"}], "project_id": "f7b96553d2fd4e26a05beb87c85c67c9", "security_groups_ids": ["adfc0f11-e15d-469e-af16-0c62826074df", "889474b8-4f3c-4d87-a18e-ab90934ca3de"], "subnet_id": "e7a1a65f-28c1-4f47-bcbd-bff7b35a3d45", "type": "ClusterIP"}, "versioned_object.name": "LBaaSServiceSpec", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}', u'openstack.org/kuryr-lbaas-state': u'{"versioned_object.data": {"listeners": [{"versioned_object.changes": ["id"], "versioned_object.data": {"id": "855e0bbf-e6a7-4d84-8aca-fa399e8ab808", "loadbalancer_id": "76f00d36-f938-419d-8c20-090010af471c", "name": "momo/echo-09:TCP:80", "port": 80, "project_id": "f7b96553d2fd4e26a05beb87c85c67c9", "protocol": "TCP"}, "versioned_object.name": "LBaaSListener", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}], "loadbalancer": {"versioned_object.data": {"id": "76f00d36-f938-419d-8c20-090010af471c", "ip": "XXX.XXX..136.204", "name": "momo/echo-09", "port_id": "ca015a97-99a7-4919-9188-86473ce25daf", "project_id": "f7b96553d2fd4e26a05beb87c85c67c9", "provider": "amphora", "security_groups": ["adfc0f11-e15d-469e-af16-0c62826074df", "889474b8-4f3c-4d87-a18e-ab90934ca3de"], "subnet_id": "e7a1a65f-28c1-4f47-bcbd-bff7b35a3d45"}, "versioned_object.name": "LBaaSLoadBalancer", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.3"}, "members": [{"versioned_object.data": {"id": "891c92bc-e557-4258-b69c-0e40cbad407c", "ip": "XXX.XXX..3.224", "name": "momo/echo-09-56fcff9f5d-z9bjt:8080", "pool_id": "ee3c8afa-e06c-463b-b444-1a9f2f0f90e6", "port": 8080, "project_id": "f7b96553d2fd4e26a05beb87c85c67c9", "subnet_id": "e7a1a65f-28c1-4f47-bcbd-bff7b35a3d45"}, "versioned_object.name": "LBaaSMember", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}], "pools": [{"versioned_object.changes": ["id"], "versioned_object.data": {"id": "ee3c8afa-e06c-463b-b444-1a9f2f0f90e6", "listener_id": "855e0bbf-e6a7-4d84-8aca-fa399e8ab808", "loadbalancer_id": "76f00d36-f938-419d-8c20-090010af471c", "name": "momo/echo-09:TCP:80", "project_id": "f7b96553d2fd4e26a05beb87c85c67c9", "protocol": "TCP"}, "versioned_object.name": "LBaaSPool", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.1"}], "service_pub_ip_info": null}, "versioned_object.name": "LBaaSState", "versioned_object.namespace": "kuryr_kubernetes", "versioned_object.version": "1.0"}'}, u'selfLink': u'/api/v1/namespaces/momo/endpoints/echo-09', u'uid': u'bc3e498c-3f5e-11eb-bf6c-fa163e90af27'}}, u'type': u'MODIFIED'}: ResourceNotReady: Resource not ready: LBaaSLoadBalancer(id=76f00d36-f938-419d-8c20-090010af471c,ip=XXX.XXX..136.204,name='momo/echo-09',port_id=ca015a97-99a7-4919-9188-86473ce25daf,project_id='f7b96553d2fd4e26a05beb87c85c67c9',provider='amphora',security_groups=[adfc0f11-e15d-469e-af16-0c62826074df,889474b8-4f3c-4d87-a18e-ab90934ca3de],subnet_id=e7a1a65f-28c1-4f47-bcbd-bff7b35a3d45)

2021-01-21 04:30:00.769 1 ERROR kuryr_kubernetes.handlers.logging ResourceNotReady: Resource not ready: LBaaSLoadBalancer(id=76f00d36-f938-419d-8c20-090010af471c,ip=XXX.XXX..136.204,name='momo/echo-09',port_id=ca015a97-99a7-4919-9188-86473ce25daf,project_id='f7b96553d2fd4e26a05beb87c85c67c9',provider='amphora',security_groups=[adfc0f11-e15d-469e-af16-0c62826074df,889474b8-4f3c-4d87-a18e-ab90934ca3de],subnet_id=e7a1a65f-28c1-4f47-bcbd-bff7b35a3d45)

=========================================================================================

I will sanitise and attach these logs. This is after the kuryr controller crashed and restarted.

I haven't been able to replicate it where the controller goes into CrashLoopBackOff and won't start untill I delete the OCP service it is complaining about (which we found in our production environment).

Comment 4 Mohammad 2021-01-21 04:59:35 UTC
Created attachment 1749299 [details]
Kuryr controller logs

Comment 5 Mohammad 2021-01-21 07:34:47 UTC
Notice the status of the loadbalancer (echo-09 existed before the octavia upgrade and echo-10 is new):

$ openstack loadbalancer list |egrep 'echo-09|echo-10'
| 76f00d36-f938-419d-8c20-090010af471c | momo/echo-09                                             | f7b96553d2fd4e26a05beb87c85c67c9 | XXX.XXX.136.204 | PENDING_UPDATE      | amphora  |
| 912c24e0-c41a-4b11-9851-5f643bad61f2 | momo/echo-10                                             | f7b96553d2fd4e26a05beb87c85c67c9 | XXX.XXX.144.5   | ACTIVE              | amphora  |

The kuryr controller has restarted 18 times since logging this BZ hours agi:

[openshift@master-0 ~]$ oc get pods -n kuryr
NAME                                READY     STATUS    RESTARTS   AGE
kuryr-cni-ds-69lvb                  2/2       Running   0          6h
kuryr-cni-ds-7mfp6                  2/2       Running   0          6h
kuryr-cni-ds-8zgkf                  2/2       Running   0          6h
kuryr-cni-ds-fgzwc                  2/2       Running   0          6h
kuryr-cni-ds-frqcj                  2/2       Running   0          6h
kuryr-cni-ds-lncts                  2/2       Running   0          6h
kuryr-cni-ds-m4k7q                  2/2       Running   0          6h
kuryr-cni-ds-rzdlx                  2/2       Running   0          6h
kuryr-cni-ds-sds7d                  2/2       Running   0          6h
kuryr-cni-ds-vw7m7                  2/2       Running   0          6h
kuryr-cni-ds-wfg77                  2/2       Running   0          6h
kuryr-controller-5676f699f5-2c98h   1/1       Running   18         3h

Comment 6 Mohammad 2021-01-21 07:53:25 UTC
I have tested this for another old echo-08 and got the same result, the service is stuck in PENDING_UPDATE.

As advised by Luis, I am upgrading the Kuryr components to 3.11.346 to work from a version that has more fixes.

I am also checking if the Octavia Amphora VM image has been updated to correspond with the OSP13z13 update.

Comment 7 Mohammad 2021-01-21 08:12:05 UTC
To confirm, services before Octavia upgrade have their loadbalancers go into PENDING_UPDATE and get stuck there (e.g., echo-09) while services created after Octavia upgrade do go into PENDING_UPDATE but for a few seconds. I have confirmed this twice.

Comment 10 Mohammad 2021-02-04 05:23:13 UTC
Confirming that a failover of loadbalancers fixes this problem.

Comment 14 Michał Dulko 2021-03-01 12:21:56 UTC
Back to ASSIGNED, this needs one more change.

Comment 17 Itzik Brown 2021-03-16 09:55:39 UTC
Failed with 
v3.11.400
OSP13z13

Removed amphora tag from amphora image
$ openstack image unset --tag amphora-image

Created a service , the LB moved to error state.
Severalloadbalancers were created until added the tag back again and the a new LB one moved to active state but the ones in error state were not deleted

Controller log:
e-11eb-b778-fa163eff3f77'}}, u'type': u'ADDED'}: TypeError: 'LBaaSLoadBalancer' object has no attribute '__getitem__'                                                                          
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging Traceback (most recent call last):                                                                                           
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/handlers/logging.py", line 37, in __call__                         
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     self._handler(event)                                                                                                     
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/handlers/retry.py", line 78, in __call__                           
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     self._handler(event)                                                                                                     
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/handlers/k8s_base.py", line 75, in __call__                        
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     self.on_present(obj)                                                                                                     
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/handlers/lbaas.py", line 188, in on_present             
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     if self._sync_lbaas_members(endpoints, lbaas_state, lbaas_spec):                                                         
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/handlers/lbaas.py", line 275, in _sync_lbaas_members    
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     if self._sync_lbaas_pools(endpoints, lbaas_state, lbaas_spec):                                                           
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/handlers/lbaas.py", line 462, in _sync_lbaas_pools      
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     if self._sync_lbaas_listeners(endpoints, lbaas_state, lbaas_spec):                                                       
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/handlers/lbaas.py", line 522, in _sync_lbaas_listeners
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     if self._add_new_listeners(endpoints, lbaas_spec, lbaas_state):
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/handlers/lbaas.py", line 551, in _add_new_listeners
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     service_type=lbaas_spec.type)
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/drivers/lbaasv2.py", line 656, in ensure_listener
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     self._find_listener, _LB_STS_POLL_SLOW_INTERVAL)
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/drivers/lbaasv2.py", line 990, in _ensure_provisioned
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     self._wait_for_provisioning(loadbalancer, remaining, interval)
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging   File "/usr/lib/python2.7/site-packages/kuryr_kubernetes/controller/drivers/lbaasv2.py", line 1027, in _wait_for_provisioning
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging     loadbalancer['id'])
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging TypeError: 'LBaaSLoadBalancer' object has no attribute '__getitem__'
2021-03-16 08:44:26.540 1 ERROR kuryr_kubernetes.handlers.logging

Comment 20 Itzik Brown 2021-03-29 08:10:38 UTC
Verified with version:v3.11.405

1. Unset tags from the amphora imgae
2. openstack image unset --tag amphora-image <amphora-image-id>
3. Created deployment:
apiVersion: apps/v1
kind: Deployment
metadata:
  name: demo
  labels:
    app: demo
spec:
  replicas: 3
  selector:
    matchLabels:
      app: demo
  template:
    metadata:
      labels:
        app: demo
    spec:
      containers:
      - name: demo
        image: quay.io/kuryr/demo
        ports:
        - containerPort: 8080

4. Created a service:
apiVersion: v1
kind: Service
metadata:
  name: demo
labels:
  app: demo
spec:
  selector:                  
    app: demo
  ports:
  - port: 80
    protocol: TCP
    targetPort: 8080

5. Watched the loadbalancer created,moving to an error state and being deleted
6. Set the tag for the amphora image
    openstack image set --tag amphora-image <amphora-image-id>
7. Watched the loadbalancer is ACTIVE
8. Checked connectivity to the service

Comment 23 errata-xmlrpc 2021-04-21 20:21:34 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (OpenShift Container Platform 3.11.420 bug fix and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2021:1147


Note You need to log in before you can comment on or make changes to this bug.