optimize process-user-data startup time #2160
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
mkulke
changed the title
mkulke
changed the title
mkulke
force-pushed
the
mkulke/optimize-process-user-data-startup
branch
6 times, most recently
from
65149b6
to
46be024
Compare
mkulke
changed the title
mkulke
force-pushed
the
mkulke/optimize-process-user-data-startup
branch
2 times, most recently
from
8253fc8
to
d889a65
Compare
mkulke
force-pushed
the
mkulke/optimize-process-user-data-startup
branch
from
d889a65
to
4469e2a
Compare
stevenhorsman
approved these changes
Nov 21, 2024
bpradipt
reviewed
Nov 21, 2024
| logger.Printf("provider: AWS, userDataUrl: %s\n", url) | ||
| return aws.GetUserData(ctx, url) | ||
| return imdsGet(ctx, url, false, nil) |
There was a problem hiding this comment.
shouldn't the bool be true here ?
There was a problem hiding this comment.
aws doesn't encode the userdata with base64, does it? at least it wasn't in the original code
There was a problem hiding this comment.
Correct.. I forgot about it.. Do you mind adding a comment above the code ?
the binary is currently importing a lot of libraries which have costly init() procedures (around 8-10s at startup) this is due to sharing some consts with the CAA modules. We can avoid that by inlining code in process-user-data and moving shared consts into dedicated modules. We can use a cpuid field to narrow down a hypervisor, so we don't have to probe an IMDS endpoint while circling through the cloud provider, delaying startup. Signed-off-by: Magnus Kulke <magnuskulke@microsoft.com>
mkulke
force-pushed
the
mkulke/optimize-process-user-data-startup
branch
from
4469e2a
to
0b5f9f4
Compare
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.
the binary is currently importing a lot of libraries which have costly init() procedures (around 8-10s at startup) this is due to sharing some consts with the CAA modules. We can avoid that by inlining code in process-user-data and moving shared consts into dedicated modules.
We can use a cpuid field to narrow down a hypervisor, so we don't have to probe an IMDS endpoint while circling through the cloud provider, delaying startup.
this is the result w/ debug enabled, the crypto libraries that are used to hash initdata are still consuming a few, but it's far better now.