OCPBUGS-98619: Bump frr-k8s MAX_FDS from 1024 to 65536 - #3054
Conversation
bfdd opens 5 (2 IPv4 + 3 IPv6) UDP sockets against each VRF even if BFD has not been explicitly enabled on them. When using RouteAdvertisements with OVN-K CUDNs, this limits the scalability to ~200 of them assuming a default file descriptor limit of 1024. Bump it to 65536 which is in any case a much more reasonable limit for production network daemons. Signed-off-by: Jaime Caamaño Ruiz <jcaamano@redhat.com>
|
@jcaamano: This pull request references Jira Issue OCPBUGS-98619, which is invalid:
Comment The bug has been updated to refer to the pull request using the external bug tracker. 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 openshift-eng/jira-lifecycle-plugin repository. |
|
Pipeline controller notification For optional jobs, comment This repository is configured in: LGTM mode |
WalkthroughThe FRR-K8s startup configuration now actively sets ChangesFRR-K8s startup configuration
Estimated code review effort: 1 (Trivial) | ~2 minutes Caution Pre-merge checks failedPlease resolve all errors before merging. Addressing warnings is optional.
❌ Failed checks (4 errors)
✅ Passed checks (20 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
|
/jira refresh |
|
@jcaamano: This pull request references Jira Issue OCPBUGS-98619, which is valid. The bug has been moved to the POST state. 3 validation(s) were run on this bug
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 openshift-eng/jira-lifecycle-plugin repository. |
venkataanil
left a comment
There was a problem hiding this comment.
I have tested with this change and able to create 250 CUDNs (i.e 250 VRFs on a worker node). Without this change, bfdd is crashing with below error for the same number of 250 CUDNs (i.e 250 VRFs)
2026/07/15 03:53:59 BFD: [XCQPZ-40HX4] echov6-socket: socket: Too many open files
2026/07/15 03:53:59 BFD: bfdd/bfd_packet.c:2140: bp_initv6_socket(): assertion (!"echov6-socket: socket: %s") failed
BFD: Received signal 6 at 1784087639 (si_addr 0x3e60000db44, PC 0x7f42a12be4bc); aborting...
BFD: /usr/lib64/frr/libfrr.so.0(zlog_backtrace_sigsafe+0x71) [0x7f42a1534d11]
BFD: /usr/lib64/frr/libfrr.so.0(zlog_signal+0xf5) [0x7f42a1534f15]
BFD: /usr/lib64/frr/libfrr.so.0(+0x101a55) [0x7f42a156fa55]
BFD: /lib64/libc.so.6(+0x3fc60) [0x7f42a1270c60]
BFD: /lib64/libc.so.6(+0x8d4bc) [0x7f42a12be4bc]
BFD: /lib64/libc.so.6(raise+0x16) [0x7f42a1270bb6]
BFD: /lib64/libc.so.6(abort+0xd3) [0x7f42a125a873]
BFD: /usr/lib64/frr/libfrr.so.0(_zlog_assert_failed+0xed) [0x7f42a159f05d]
BFD: /usr/libexec/frr/bfdd(bp_initv6_socket+0xe9) [0x55d104bcb2c9]
BFD: /usr/libexec/frr/bfdd(+0x274a8) [0x55d104bc44a8]
BFD: /usr/lib64/frr/libfrr.so.0(vrf_enable+0x59) [0x7f42a1586959]
BFD: /usr/lib64/frr/libfrr.so.0(+0x128a7b) [0x7f42a1596a7b]
BFD: /usr/lib64/frr/libfrr.so.0(+0x13ad3c) [0x7f42a15a8d3c]
BFD: /usr/lib64/frr/libfrr.so.0(event_call+0x9f) [0x7f42a1585ccf]
BFD: /usr/lib64/frr/libfrr.so.0(frr_run+0xd8) [0x7f42a152b798]
BFD: /usr/libexec/frr/bfdd(main+0x1e2) [0x55d104bb5a02]
BFD: /lib64/libc.so.6(+0x2a610) [0x7f42a125b610]
BFD: /lib64/libc.so.6(__libc_start_main+0x80) [0x7f42a125b6c0]
BFD: /usr/libexec/frr/bfdd(_start+0x25) [0x55d104bb5dd5]
BFD: in thread zclient_read scheduled from lib/zclient.c:4959 zclient_event()
|
/lgtm |
|
Scheduling required tests: Scheduling tests matching the |
|
/verified by @venkataanil |
|
@venkataanil: This PR has been marked as verified by 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 openshift-eng/jira-lifecycle-plugin repository. |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: fedepaol, jcaamano, jtaleric, venkataanil 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 |
|
/retest |
|
/test all |
|
/override ci/prow/e2e-aws-ovn-upgrade-ipsec |
|
@jcaamano: Overrode contexts on behalf of jcaamano: ci/prow/e2e-aws-ovn-upgrade-ipsec, ci/prow/e2e-metal-ipi-ovn-ipv6-ipsec, ci/prow/e2e-ovn-ipsec-step-registry 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-sigs/prow repository. |
|
/override ci/prow/e2e-aws-ovn-upgrade-ipsec |
|
@jcaamano: Overrode contexts on behalf of jcaamano: ci/prow/e2e-aws-ovn-upgrade-ipsec, ci/prow/e2e-metal-ipi-ovn-ipv6-ipsec, ci/prow/e2e-ovn-ipsec-step-registry 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-sigs/prow repository. |
|
/retest-required |
|
/override ci/prow/e2e-aws-ovn-fdp-qe https://github.com/openshift/openshift-tests-private/pull/30074 |
|
@jcaamano: Overrode contexts on behalf of jcaamano: ci/prow/e2e-aws-ovn-fdp-qe 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-sigs/prow repository. |
|
/override ci/prow/e2e-aws-ovn-fdp-qe |
|
@jcaamano: Overrode contexts on behalf of jcaamano: ci/prow/e2e-aws-ovn-fdp-qe 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-sigs/prow repository. |
|
/retest-required |
|
/retest-required |
|
@jcaamano: 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-sigs/prow repository. I understand the commands that are listed here. |
|
/override ci/prow/e2e-gcp-ovn-techpreview A job that we don't own with a 50% pass rate (periodic) cannot really be required. The problems cannot possibly be related to this change. |
|
@jcaamano: Overrode contexts on behalf of jcaamano: ci/prow/e2e-gcp-ovn-techpreview 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-sigs/prow repository. |
|
/retest-required |
5c7f864
into
openshift:master
|
@jcaamano: Jira Issue Verification Checks: Jira Issue OCPBUGS-98619 Jira Issue OCPBUGS-98619 has been moved to the MODIFIED state and will move to the VERIFIED state when the change is available in an accepted nightly payload. 🕓 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 openshift-eng/jira-lifecycle-plugin repository. |
|
/cherry-pick release-4.22 |
|
@jcaamano: new pull request created: #3090 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-sigs/prow repository. |
|
/jira backport release-4.21,release-4.20 |
|
@arkadeepsen: Missing required branches for backport chain:
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 openshift-eng/jira-lifecycle-plugin repository. |
|
/jira backport release-4.22,release-4.21,release-4.20 |
|
@arkadeepsen: The following backport issues have been created:
Queuing cherrypicks to the requested branches to be created after this PR merges: 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 openshift-eng/jira-lifecycle-plugin repository. |
|
@openshift-ci-robot: new pull request created: #3091 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-sigs/prow repository. |
|
@openshift-ci-robot: new pull request created: #3092 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-sigs/prow repository. |
|
@openshift-ci-robot: new pull request could not be created: failed to create pull request against openshift/cluster-network-operator#release-4.22 from head openshift-cherrypick-robot:cherry-pick-3054-to-release-4.22: status code 422 not one of [201], body: {"message":"Validation Failed","errors":[{"resource":"PullRequest","code":"custom","message":"A pull request already exists for openshift-cherrypick-robot:cherry-pick-3054-to-release-4.22."}],"documentation_url":"https://docs.github.com/rest/pulls/pulls#create-a-pull-request","status":"422"} 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-sigs/prow repository. |
bfdd opens 5 (2 IPv4 + 3 IPv6) UDP sockets against each VRF even if BFD has not been explicitly enabled on them. When using RouteAdvertisements with OVN-K CUDNs, this limits the scalability to ~200 of them assuming a default file descriptor limit of 1024. Bump it to 65536 which is in any case a much more reasonable limit for production network daemons.