eni: Fix Cilium overallocating network interfaces #15911
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
gandro
added
release-note/bug
gandro
force-pushed
the
pr/gandro/fix-eni-creation-loop
branch
2 times, most recently
from
9563b6d
to
1eb7545
Compare
This commit modifies the CreateNetworkInterface mock API to mirror what the actual EC2 API implementation does for the `toAllocate` value, i.e. only use this number for secondary IP addresses, as a primary address is always implicitly allocated. Signed-off-by: Sebastian Wicki <sebastian@isovalent.com>
This fixes a bug where Cilium wrongly overestimated the amount of available ENI IP addresses. This bug was introduced when we removed the primary ENI IP address from the IPAM pool, but forgot to adjust the number of addresses used to compare with the AWS instance limits. This led to the operator overestimating the number of available IP addresses by one. This in turn could lead to the operator first failing to allocate more IPs (because it exceeded the limit) and then unnecessarily creating a new ENI to fulfill the allocation request. Fixes: 7c1bb35 ("aws/ec2: Exclude primary ENI IP from IPAM pool") Fixes: cilium#15877 Signed-off-by: Sebastian Wicki <sebastian@isovalent.com>
gandro
force-pushed
the
pr/gandro/fix-eni-creation-loop
branch
from
1eb7545
to
2f60008
Compare
|
test-me-please |
1 similar comment
|
test-me-please |
tgraf
approved these changes
Apr 28, 2021
|
retest-1.16-netnext Edit: Was https://jenkins.cilium.io/job/Cilium-PR-K8s-1.16-net-next/406/ #15455 |
|
retest-gke Edit: Cilium was sigkilled, might have been a issue with the cluster - retesting https://jenkins.cilium.io/job/Cilium-PR-K8s-GKE/5331/ |
|
retest-1.16-netnext Seems like it hit a network fluke: https://jenkins.cilium.io/job/Cilium-PR-K8s-1.16-net-next/408/ |
|
test-1.19-5.4 |
|
test-1.16-netnext |
gandro
added
the
kind/regression
christarazi
approved these changes
May 3, 2021
|
Marking this ready to merge, the CI failure is in the datapath which is not affected by this very ENI-specific PR. Created an issue for it #15998 |
gandro
added
the
ready-to-merge
maintainer-s-little-helper
bot
moved this from Needs backport from master
to Backport pending to v1.8
in 1.8.10
maintainer-s-little-helper
bot
moved this from Needs backport from master
to Backport pending to v1.9
in 1.9.7
maintainer-s-little-helper
bot
moved this from Backport pending to v1.8
to Backport done to v1.8
in 1.8.10
maintainer-s-little-helper
bot
moved this from Backport pending to v1.8
to Backport done to v1.8
in 1.8.10
maintainer-s-little-helper
bot
moved this from Backport pending to v1.9
to Backport done to v1.9
in 1.9.7
maintainer-s-little-helper
bot
moved this from Needs backport from master
to Backport pending to v1.10
in 1.10.0-rc2
maintainer-s-little-helper
bot
moved this from Backport pending to v1.10
to Backport done to v1.10
in 1.10.0-rc2
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This fixes a bug where Cilium wrongly overestimated the amount of available
ENI IP addresses. This bug was introduced when we removed the primary
ENI IP address from the IPAM pool, but forgot to adjust the number of
addresses used to compare with the AWS instance limits.
This led to the operator overestimating the number of available IP
addresses by one. This in turn could lead to the operator first failing
to allocate more IPs (because it exceeded the limit) and then
unnecessarily creating a new ENI to fulfill the allocation request.
Fixes: 7c1bb35 ("aws/ec2: Exclude primary ENI IP from IPAM pool")
Fixes: #15877