-
Notifications
You must be signed in to change notification settings - Fork 207
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
node-bootstrapper #4891
node-bootstrapper #4891
Conversation
Pull Request Test Coverage Report for Build 11284326238Details
💛 - Coveralls |
6752a69
to
deef18d
Compare
@@ -32,6 +32,9 @@ type Configuration struct { | |||
IgnoreScenariosWithMissingVHD bool `env:"IGNORE_SCENARIOS_WITH_MISSING_VHD"` | |||
SkipTestsWithSKUCapacityIssue bool `env:"SKIP_TESTS_WITH_SKU_CAPACITY_ISSUE"` | |||
KeepVMSS bool `env:"KEEP_VMSS"` | |||
BlobStorageAccount string `env:"BLOB_STORAGE_ACCOUNT" envDefault:"https://abe2e.blob.core.windows.net"` |
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.
is this a new blob storage ? how was this created ? do we have IaC for it? if this is purely to host the binary for the test, I'm wondering if we could avoid that, knowing that the more azure resources we have to more SFI work we have to do.
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.
It's a hack to speed e2e feedback loop.
Without it we need to rebuild the VHD before running any tests. It adds about an hour to the loop. (It takes minutes with the hack)
This code put binary on a node and run a similar CSE command to what we're going to use.
I had difficulty finding a way to upload a binary to a VM. It's not available from a public network. So, I couldn't SSH into it. And we need to execute it before we join the k8s cluster (other e2e test helpers utilizes k8s api to connect to the VM)
I think I created it manually (it's been a while ago).
I don't really like it and happy to change if you have any other ideas.
@@ -148,7 +148,7 @@ linters-settings: | |||
shadow: | |||
# Whether to be strict about shadowing; can be noisy. | |||
# Default: false | |||
strict: true | |||
strict: false |
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.
why was this changed ?
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.
The linter is being really noisy for a typical go error handling.
I think shadowing errors is fine, we shouldn't come up with err2 and err3 variables.
The current implementation is a simple wrapper around CSE scripts and a custom data generator for Linux (previously known as "self-contained" efforts).
This approach moves the script generation logic from the API layer into the VHD itself, making virtual machines more consistent and allowing for significant simplification.
By having a Go application as the main entry point to provision the VHD, we unlock a lot of new possibilities.
Switching to the new generation should be as simple as replacing
GetNodeBootstrapping
withGetNodeBootstrappingV2
.No changes are introduced to the current VHD provisioning.