-
Notifications
You must be signed in to change notification settings - Fork 1.3k
feat(proxy, proxy-init): Combine proxy and proxy-init and use minimal base image #14577
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Changes from all commits
d676645
784f7b6
45dd1a9
9c16267
a465635
f0dc78a
a0e3bf7
308c94e
d418cad
28a4ed5
891e3d4
8660ab0
2f9637f
200deaa
3fbe0d1
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -349,14 +349,6 @@ proxyInit: | |
| # -- Log format (`plain` or `json`) for the proxy-init | ||
| # @default -- plain | ||
| logFormat: "" | ||
| image: | ||
| # -- Docker image for the proxy-init container | ||
| name: cr.l5d.io/linkerd/proxy-init | ||
| # -- Pull policy for the proxy-init container image | ||
| # @default -- imagePullPolicy | ||
| pullPolicy: "" | ||
| # -- Tag for the proxy-init container image | ||
| version: v2.4.3 | ||
|
Comment on lines
-352
to
-359
Member
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. cross-referencing linkerd/linkerd2-proxy#4333. |
||
| # -- Changes the default value for the nf_conntrack_tcp_timeout_close_wait | ||
| # kernel parameter. If used, runAsRoot needs to be true. | ||
| closeWaitTimeoutSecs: 0 | ||
|
|
||
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Assume that there are users that run the CNI plugin because they do not want containers with any CAP_NET_ADMIN.
In this situation they would now still pull that image. Is that a problem? Or that is ultimately limited by what caps are being set in the workload definition? Which brings me to the next question. Do we have any tests that exercise this image in CNI mode?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
My understanding is that this is okay because the proxy containers (which do not have the CAP_NET_ADMIN capability) never attempt to execute any of these binaries which have additional caps set on them. If it did, they would fail with a permission denied error. On the other hand, the proxy-init container does have the CAP_NET_ADMIN capability and can execute these binaries just fine. Same image, but different runtime capabilities and different binaries executed.
Security policies will be looking to make sure that the proxy container itself doesn't have CAP_NET_ADMIN, which it doesn't.
The cni-calico-deep integration test exercises CNI mode with this proxy.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok this is in line with what I was also imagining the case to be. Thanks for verifying that there is a test that exercises this logic.