Fix permission issue in Beam pipeline deployment (#170)

This commit is contained in:
Shicong Huang
2019-07-15 16:13:42 -04:00
committed by GitHub
parent 650f1fdd52
commit 633dd887f4
5 changed files with 83 additions and 56 deletions
@@ -14,11 +14,14 @@
package google.registry.beam.invoicing;
import com.google.auth.oauth2.GoogleCredentials;
import google.registry.beam.invoicing.BillingEvent.InvoiceGroupingKey;
import google.registry.beam.invoicing.BillingEvent.InvoiceGroupingKey.InvoiceGroupingKeyCoder;
import google.registry.config.CredentialModule.LocalCredential;
import google.registry.config.RegistryConfig.Config;
import google.registry.reporting.billing.BillingModule;
import google.registry.reporting.billing.GenerateInvoicesAction;
import google.registry.util.GoogleCredentialsBundle;
import java.io.Serializable;
import javax.inject.Inject;
import org.apache.beam.runners.dataflow.DataflowRunner;
@@ -55,32 +58,31 @@ import org.apache.beam.sdk.values.TypeDescriptors;
*/
public class InvoicingPipeline implements Serializable {
@Inject
@Config("projectId")
String projectId;
private final String projectId;
private final String beamBucketUrl;
private final String invoiceTemplateUrl;
private final String beamStagingUrl;
private final String billingBucketUrl;
private final String invoiceFilePrefix;
private final GoogleCredentials googleCredentials;
@Inject
@Config("apacheBeamBucketUrl")
String beamBucketUrl;
@Inject
@Config("invoiceTemplateUrl")
String invoiceTemplateUrl;
@Inject
@Config("beamStagingUrl")
String beamStagingUrl;
@Inject
@Config("billingBucketUrl")
String billingBucketUrl;
@Inject
@Config("invoiceFilePrefix")
String invoiceFilePrefix;
@Inject
InvoicingPipeline() {}
public InvoicingPipeline(
@Config("projectId") String projectId,
@Config("apacheBeamBucketUrl") String beamBucketUrl,
@Config("invoiceTemplateUrl") String invoiceTemplateUrl,
@Config("beamStagingUrl") String beamStagingUrl,
@Config("billingBucketUrl") String billingBucketUrl,
@Config("invoiceFilePrefix") String invoiceFilePrefix,
@LocalCredential GoogleCredentialsBundle googleCredentialsBundle) {
this.projectId = projectId;
this.beamBucketUrl = beamBucketUrl;
this.invoiceTemplateUrl = invoiceTemplateUrl;
this.beamStagingUrl = beamStagingUrl;
this.billingBucketUrl = billingBucketUrl;
this.invoiceFilePrefix = invoiceFilePrefix;
this.googleCredentials = googleCredentialsBundle.getGoogleCredentials();
}
/** Custom options for running the invoicing pipeline. */
interface InvoicingPipelineOptions extends DataflowPipelineOptions {
@@ -105,6 +107,10 @@ public class InvoicingPipeline implements Serializable {
// This causes p.run() to stage the pipeline as a template on GCS, as opposed to running it.
options.setTemplateLocation(invoiceTemplateUrl);
options.setStagingLocation(beamStagingUrl);
// This credential is used when Dataflow deploys the template to GCS in target GCP project.
// So, make sure the credential has write permission to GCS in that project.
options.setGcpCredential(googleCredentials);
Pipeline p = Pipeline.create(options);
PCollection<BillingEvent> billingEvents =
@@ -17,9 +17,12 @@ package google.registry.beam.spec11;
import static com.google.common.base.Preconditions.checkArgument;
import static google.registry.beam.BeamUtils.getQueryFromFile;
import com.google.auth.oauth2.GoogleCredentials;
import com.google.auto.value.AutoValue;
import google.registry.beam.spec11.SafeBrowsingTransforms.EvaluateSafeBrowsingFn;
import google.registry.config.CredentialModule.LocalCredential;
import google.registry.config.RegistryConfig.Config;
import google.registry.util.GoogleCredentialsBundle;
import google.registry.util.Retrier;
import google.registry.util.SqlTemplate;
import java.io.Serializable;
@@ -77,26 +80,29 @@ public class Spec11Pipeline implements Serializable {
/** The JSON object field we put the threat match array for Spec11 reports. */
public static final String THREAT_MATCHES_FIELD = "threatMatches";
@Inject
@Config("projectId")
String projectId;
private final String projectId;
private final String beamStagingUrl;
private final String spec11TemplateUrl;
private final String reportingBucketUrl;
private final GoogleCredentials googleCredentials;
private final Retrier retrier;
@Inject
@Config("beamStagingUrl")
String beamStagingUrl;
@Inject
@Config("spec11TemplateUrl")
String spec11TemplateUrl;
@Inject
@Config("reportingBucketUrl")
String reportingBucketUrl;
@Inject Retrier retrier;
@Inject
Spec11Pipeline() {}
public Spec11Pipeline(
@Config("projectId") String projectId,
@Config("beamStagingUrl") String beamStagingUrl,
@Config("spec11TemplateUrl") String spec11TemplateUrl,
@Config("reportingBucketUrl") String reportingBucketUrl,
@LocalCredential GoogleCredentialsBundle googleCredentialsBundle,
Retrier retrier
) {
this.projectId = projectId;
this.beamStagingUrl = beamStagingUrl;
this.spec11TemplateUrl = spec11TemplateUrl;
this.reportingBucketUrl = reportingBucketUrl;
this.googleCredentials = googleCredentialsBundle.getGoogleCredentials();
this.retrier = retrier;
}
/** Custom options for running the spec11 pipeline. */
interface Spec11PipelineOptions extends DataflowPipelineOptions {
@@ -134,6 +140,9 @@ public class Spec11Pipeline implements Serializable {
// This causes p.run() to stage the pipeline as a template on GCS, as opposed to running it.
options.setTemplateLocation(spec11TemplateUrl);
options.setStagingLocation(beamStagingUrl);
// This credential is used when Dataflow deploys the template to GCS in target GCP project.
// So, make sure the credential has write permission to GCS in that project.
options.setGcpCredential(googleCredentials);
Pipeline p = Pipeline.create(options);
PCollection<Subdomain> domains =
@@ -100,7 +100,7 @@ public class AuthModule {
AbstractDataStoreFactory dataStoreFactory) {
try {
return new GoogleAuthorizationCodeFlow.Builder(
new NetHttpTransport(), jsonFactory, clientSecrets, requiredOauthScopes)
new NetHttpTransport(), jsonFactory, clientSecrets, requiredOauthScopes)
.setDataStoreFactory(dataStoreFactory)
.build();
} catch (IOException ex) {