ACL Indexes Merge Downstream: 31st May 2023#1684
ACL Indexes Merge Downstream: 31st May 2023#1684tssurya wants to merge 18 commits intoopenshift:masterfrom
Conversation
which mean reset, not ignore. When acls are updated on initial sync (by handling add events) if namespace log levels were reset, acls should also be updated with nil Severity. Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
intention from implementation. Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
Add new BuildACLFromDbIDs function, that should be the only one used in the end of acls indexes update. Add 2 owner Types for global and namespaced multicast resources. Generate ACL name based on dbIDs, don't include owner information to save some symbols. Add (objectIDs *DbObjectIDs) StringNoOwner() for this purpose. Add cleanup function for disabled multicast support and namespaced multicast cleanup. Move noneMatch to the gress_policy.go, where it is used. Add tests for multicast acl sync and cleanup. Rename address_set_syncer package to external_ids_syncer, add acl syncer sub-package. Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
1. Use dbIDs with new NetpolNodeOwnerType 2. Add aclSync for allow from node ACL 3. Rework AllowFromNode ACL unit tests to run everything for ipv4 and ipv6 Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
Move gress-related constants from policy.go to gress_policy.go. Use old externalIDs for syncs only. policy test "correctly retries recreating a network policy with the same name" last expected data was updated, since newly generated ACLs Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
Add NetpolNamespaceOwnerType for namespace-wide default deny acls. Add acl_sync PrimaryID check to make sure in case more than 1 ACLs have the same primaryID, only of them will be updated. This also eliminates the need for "deleting a network policy that failed half-way through creation succeeds" test, which is based on a "found multiple results for provided predicate" condition. Replace tests "stale ACLs should be cleaned up or updated at startup via syncNetworkPolicies" and "ACLs with long names and run syncNetworkPolicies" with the new "reconciles an existing networkPolicy updating stale ACLs with long names", and a part that deletes stale ARP ACLs is now updated and tested as a part of acl_sync. Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
Add EgressFirewallOwnerType for egress firewall ACLs. Update syncEgressFirewall to use passed egressFirewalls objects for cleanup, create updateEgressFirewallACLsDbIndex function that updates old formatted ACLs. Update egressfirewall_test.go, move a part of setup to BeforeEach. Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
1. rename BuildACLFromDbIndex to BuildACL, since it is the only method that should be used now. 2. Rename getACLMatchFromACLDir to getACLMatch, remove stale version. 3. update isEquivalentACL to checkACLPrimaryID, that matches on primaryID until we get client indexes. 4. Add acls.md doc explaining all acls that are used by ovn-k and their dependencies with examples 5. Update multicast docs Add missing links to README.md Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
Now Egress firewall ACLs are named "EF:<namespace>:<priority>" default deny netpol ACLs "NP:<namespace>:<direction>" gress ACLs "NP:<policyNamespace>:<policyName>:<direction>:<gressIdx>" Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
Acls new indexes
Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
…once Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
Signed-off-by: Nadia Pinaeva <npinaeva@redhat.com>
…t that. Spotted in upstream ovn-org/ovn CI when running against ovn versions <=22.09 which don't support component templates: https://github.com/ovn-org/ovn/actions/runs/4628617882 Reported error: failed to sync chassis: error: failed to get template var list: error: Wrong parameter type (*nbdb.ChassisTemplateVar): Model not found in Database Model Fixes: 4b3475a ("services: Use OVN template load balancers for NodePort services.") Signed-off-by: Dumitru Ceara <dceara@redhat.com>
services: Don't try to list/cleanup templates when OVN doesn't suppor…
Minor performance fixes
Use PrimaryID as a client index for ACL.
…cl-indexes-merge-downstream Conflicts: go-controller/pkg/libovsdb/libovsdb.go because openshift#1652 is merged already
331f2b2 to
2f2bc32
Compare
|
/hold cancel |
|
/retest unit |
|
@tssurya: The
The following commands are available to trigger optional jobs:
Use
DetailsIn response to this:
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. |
|
/test unit |
1 similar comment
|
/test unit |
|
unit test will fail till we get ovn-kubernetes/ovn-kubernetes#3505 in, I can get that in only when this merges since we don't want to disrupt merge order, so we are going to have to override that job :/ |
|
/test unit |
|
/test unit |
|
/retest |
|
/payload 4.14 nightly informing |
|
@tssurya: trigger 55 job(s) of type informing for the nightly release of OCP 4.14
See details on https://pr-payload-tests.ci.openshift.org/runs/ci/5619f840-ff81-11ed-9b89-6c0cf7df2d1e-0 |
|
/payload-abort |
|
/payload-aggregate periodic-ci-openshift-release-master-ci-4.14-e2e-gcp-ovn-upgrade 5 |
|
@tssurya: trigger 1 job(s) for the /payload-(job|aggregate) command
See details on https://pr-payload-tests.ci.openshift.org/runs/ci/173d7b50-ff82-11ed-89b0-105077cde5d5-0 |
|
/lgtm for conflicts around #1652 |
|
/test e2e-aws-ovn-hypershift |
|
/lgtm |
|
/hold |
|
/approve |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: jcaamano, npinaeva, tssurya The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/test e2e-aws-ovn-windows |
|
/hold cancel |
|
/test e2e-aws-ovn-windows |
|
/test e2e-vsphere-windows |
|
/hold |
|
don't remove hold here, we want to prioritize #1690 |
|
@tssurya: The following tests failed, say
Full PR test history. Your PR dashboard. DetailsInstructions 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. I understand the commands that are listed here. |
|
closed in favour of #1692 |
Uh oh!
There was an error while loading. Please reload this page.