New collector based tests #66
Conversation
avilevy18
force-pushed
the
avilevy-collector-e2e-poc
branch
from
b605c1b to
ed87d23
Compare
# Conflicts: # go.mod # go.sum
| @@ -0,0 +1,86 @@ | |||
| //go:build e2ecollector | |||
|
|
|||
| package e2etestrunner_collector | |||
There was a problem hiding this comment.
The folder should be named the same as the package, i.e. the folder should be renamed to e2etestrunner_collector
| ) | ||
|
|
||
| func TestMain(m *testing.M) { | ||
| rand.New(rand.NewSource(time.Now().UnixNano())) |
There was a problem hiding this comment.
Are lines 26..57 the same as in e2etestrunner/main_test.go? Maybe those could be factored out somehow rather than doing the hard duplication like this.
| }, | ||
| logger, | ||
| ) | ||
| if err != nil { |
There was a problem hiding this comment.
44..48 is equivalent to
return cleanupTf, errThis comment applies to all the setups here it looks like
| // SetupGceCollector Set up the collector to run in GCE container. Creates a new | ||
| // GCE VM resources, and runs the specified container image. The | ||
| // returned cleanup function tears down the VM. | ||
| func SetupGceCollector( |
There was a problem hiding this comment.
Do any of these setup functions differ in any way other than the terraform they are running?
| now := time.Now() | ||
| start := timestamppb.New(now.Add(-window)) | ||
| end := timestamppb.New(now) | ||
| fmt.Println(fmt.Sprintf("resource.type = %q", resourceType)) |
There was a problem hiding this comment.
Is this println left in intentionally?
|
|
||
| fmt.Println(fmt.Sprintf("Entry found: %v", entry)) | ||
| if err != nil { | ||
| t.Errorf(fmt.Sprintf("Could not any logs matching filter: %s", filter)) |
There was a problem hiding this comment.
- Should be fatal failure probably?
Could not *find any...- We might as well include the error in the failure message
| t.Errorf(fmt.Sprintf("Could not any logs matching filter: %s", filter)) | |
| t.Fatalf(fmt.Sprintf("Could not find any logs matching filter %s: %v", filter, err)) |
| time.Sleep(10 * time.Second) | ||
| } | ||
|
|
||
| fmt.Println(fmt.Sprintf("Entry found: %v", entry)) |
There was a problem hiding this comment.
This should be after the err check I think
| if tracesFound == 0 { | ||
| t.Errorf(fmt.Sprintf("Could not find traces with resource attribute otelcol_google_e2e:%s", args.TestRunID)) | ||
| } | ||
| fmt.Println(fmt.Sprintf("Found traces with otelcol_google_e2e:%s", args.TestRunID)) |
There was a problem hiding this comment.
| fmt.Println(fmt.Sprintf("Found traces with otelcol_google_e2e:%s", args.TestRunID)) | |
| fmt.Println(fmt.Sprintf("Found traces with resource attribute otelcol_google_e2e:%s", args.TestRunID)) |
| } | ||
|
|
||
| req := cloudService.Projects.Traces.List(args.ProjectID) | ||
| req.Filter(fmt.Sprintf("otelcol_google_e2e:%s", args.TestRunID)) |
There was a problem hiding this comment.
otelcol_google_e2e:%s appears multiple times, maybe the sprintf should be a var that is used in those spots instead.
| // limitations under the License. | ||
|
|
||
| package e2etestrunner | ||
| package util |
There was a problem hiding this comment.
We should give this package a clearer name than util, how about e2etesting?
open-telemetry/opentelemetry-collector#11205 describes the migration
…avilevy-collector-e2e-poc
3 participants
This PR creates a new set of tests that install the collector on an environment with the config passed in as an environment variable.
b/380100825
Testing on: