This repository was archived by the owner on Apr 22, 2025. It is now read-only.
Commit 165256e
authored
feat(cli-integ): speed up image pulls by logging into ECR public (#84)
When running the suite on CodeBuild, we are seeing timeout failures in
the following tests:
- `CDK synth add the metadata properties expected by sam`
- `CDK synth bundled functions as expected`
- `test resource import with construct that requires bundling`
- `deploy new style synthesis to new style bootstrap (with docker
image)`
- `all calls from isolated container go through proxy`
These are all tests that either directly pull from `public.ecr.gallery`,
or do so as part of `cdk synth`. In both cases, the pull is
unauthenticated, and is therefore susceptible to stricter throttling
limits, which can significantly slow things down.
Running the tests individually succeeds, so it must be something to do
with the high concurrency introduced by atmosphere.
Lets login to ECR public before every test. I'm not convinced this will
solve the issue, but it can't hurt and worst case we eliminate that as a
cause and continue investigating.
**Why are we not seeing the same errors when running in GitHub?**
My theory right now is that CodeBuild IP addresses are being throttled
much more aggressively than Github ones.1 parent 7afef9c commit 165256e
File tree
3 files changed
+4
-0
lines changed- packages/@aws-cdk-testing/cli-integ/lib
3 files changed
+4
-0
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
46 | 46 | | |
47 | 47 | | |
48 | 48 | | |
| 49 | + | |
49 | 50 | | |
50 | 51 | | |
51 | 52 | | |
| |||
119 | 120 | | |
120 | 121 | | |
121 | 122 | | |
| 123 | + | |
122 | 124 | | |
123 | 125 | | |
124 | 126 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
27 | 27 | | |
28 | 28 | | |
29 | 29 | | |
| 30 | + | |
30 | 31 | | |
31 | 32 | | |
32 | 33 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
35 | 35 | | |
36 | 36 | | |
37 | 37 | | |
| 38 | + | |
38 | 39 | | |
39 | 40 | | |
40 | 41 | | |
| |||
0 commit comments