Implement parallel processing with retry logic for repository mirroring and syncing operations

This commit is contained in:
Arunavo Ray
2025-05-22 13:28:28 +05:30
parent aaf8dc6fe4
commit f4bc28e6c2
6 changed files with 503 additions and 156 deletions

View File

@@ -6,6 +6,8 @@ import { createGitHubClient } from "@/lib/github";
import { mirrorGitHubOrgToGitea } from "@/lib/gitea";
import { repoStatusEnum } from "@/types/Repository";
import { type MembershipRole } from "@/types/organizations";
import { processWithRetry } from "@/lib/utils/concurrency";
import { createMirrorJob } from "@/lib/helpers";
export const POST: APIRoute = async ({ request }) => {
try {
@@ -61,31 +63,73 @@ export const POST: APIRoute = async ({ request }) => {
);
}
// Fire async mirroring without blocking response
// Fire async mirroring without blocking response, using parallel processing
setTimeout(async () => {
for (const org of orgs) {
if (!config.githubConfig.token) {
throw new Error("GitHub token is missing in config.");
}
if (!config.githubConfig.token) {
throw new Error("GitHub token is missing in config.");
}
const octokit = createGitHubClient(config.githubConfig.token);
// Create a single Octokit instance to be reused
const octokit = createGitHubClient(config.githubConfig.token);
try {
// Define the concurrency limit - adjust based on API rate limits
// Using a lower concurrency for organizations since each org might contain many repos
const CONCURRENCY_LIMIT = 2;
// Process organizations in parallel with retry capability
await processWithRetry(
orgs,
async (org) => {
// Prepare organization data
const orgData = {
...org,
status: repoStatusEnum.parse("imported"),
membershipRole: org.membershipRole as MembershipRole,
lastMirrored: org.lastMirrored ?? undefined,
errorMessage: org.errorMessage ?? undefined,
};
// Log the start of mirroring
console.log(`Starting mirror for organization: ${org.name}`);
// Create a mirror job entry to track progress
await createMirrorJob({
userId: config.userId || "",
organizationId: org.id,
organizationName: org.name,
message: `Started mirroring organization: ${org.name}`,
details: `Organization ${org.name} is now in the mirroring queue.`,
status: "mirroring",
});
// Mirror the organization
await mirrorGitHubOrgToGitea({
config,
octokit,
organization: {
...org,
status: repoStatusEnum.parse("imported"),
membershipRole: org.membershipRole as MembershipRole,
lastMirrored: org.lastMirrored ?? undefined,
errorMessage: org.errorMessage ?? undefined,
},
organization: orgData,
});
} catch (error) {
console.error(`Mirror failed for organization ${org.name}:`, error);
return org;
},
{
concurrencyLimit: CONCURRENCY_LIMIT,
maxRetries: 2,
retryDelay: 3000,
onProgress: (completed, total, result) => {
const percentComplete = Math.round((completed / total) * 100);
console.log(`Organization mirroring progress: ${percentComplete}% (${completed}/${total})`);
if (result) {
console.log(`Successfully mirrored organization: ${result.name}`);
}
},
onRetry: (org, error, attempt) => {
console.log(`Retrying organization ${org.name} (attempt ${attempt}): ${error.message}`);
}
}
}
);
console.log("All organization mirroring tasks completed");
}, 0);
const responsePayload: MirrorOrgResponse = {

View File

@@ -8,6 +8,8 @@ import {
mirrorGitHubOrgRepoToGiteaOrg,
} from "@/lib/gitea";
import { createGitHubClient } from "@/lib/github";
import { processWithRetry } from "@/lib/utils/concurrency";
import { createMirrorJob } from "@/lib/helpers";
export const POST: APIRoute = async ({ request }) => {
try {
@@ -63,52 +65,84 @@ export const POST: APIRoute = async ({ request }) => {
);
}
// Start async mirroring in background
// Start async mirroring in background with parallel processing
setTimeout(async () => {
for (const repo of repos) {
if (!config.githubConfig.token) {
throw new Error("GitHub token is missing.");
}
if (!config.githubConfig.token) {
throw new Error("GitHub token is missing.");
}
const octokit = createGitHubClient(config.githubConfig.token);
// Create a single Octokit instance to be reused
const octokit = createGitHubClient(config.githubConfig.token);
try {
// Define the concurrency limit - adjust based on API rate limits
const CONCURRENCY_LIMIT = 3;
// Process repositories in parallel with retry capability
await processWithRetry(
repos,
async (repo) => {
// Prepare repository data
const repoData = {
...repo,
status: repoStatusEnum.parse("imported"),
organization: repo.organization ?? undefined,
lastMirrored: repo.lastMirrored ?? undefined,
errorMessage: repo.errorMessage ?? undefined,
forkedFrom: repo.forkedFrom ?? undefined,
visibility: repositoryVisibilityEnum.parse(repo.visibility),
mirroredLocation: repo.mirroredLocation || "",
};
// Log the start of mirroring
console.log(`Starting mirror for repository: ${repo.name}`);
// Create a mirror job entry to track progress
await createMirrorJob({
userId: config.userId || "",
repositoryId: repo.id,
repositoryName: repo.name,
message: `Started mirroring repository: ${repo.name}`,
details: `Repository ${repo.name} is now in the mirroring queue.`,
status: "mirroring",
});
// Mirror the repository based on whether it's in an organization
if (repo.organization && config.githubConfig.preserveOrgStructure) {
await mirrorGitHubOrgRepoToGiteaOrg({
config,
octokit,
orgName: repo.organization,
repository: {
...repo,
status: repoStatusEnum.parse("imported"),
organization: repo.organization ?? undefined,
lastMirrored: repo.lastMirrored ?? undefined,
errorMessage: repo.errorMessage ?? undefined,
forkedFrom: repo.forkedFrom ?? undefined,
visibility: repositoryVisibilityEnum.parse(repo.visibility),
mirroredLocation: repo.mirroredLocation || "",
},
repository: repoData,
});
} else {
await mirrorGithubRepoToGitea({
octokit,
repository: {
...repo,
status: repoStatusEnum.parse("imported"),
organization: repo.organization ?? undefined,
lastMirrored: repo.lastMirrored ?? undefined,
errorMessage: repo.errorMessage ?? undefined,
forkedFrom: repo.forkedFrom ?? undefined,
visibility: repositoryVisibilityEnum.parse(repo.visibility),
mirroredLocation: repo.mirroredLocation || "",
},
repository: repoData,
config,
});
}
} catch (error) {
console.error(`Mirror failed for repo ${repo.name}:`, error);
return repo;
},
{
concurrencyLimit: CONCURRENCY_LIMIT,
maxRetries: 2,
retryDelay: 2000,
onProgress: (completed, total, result) => {
const percentComplete = Math.round((completed / total) * 100);
console.log(`Mirroring progress: ${percentComplete}% (${completed}/${total})`);
if (result) {
console.log(`Successfully mirrored repository: ${result.name}`);
}
},
onRetry: (repo, error, attempt) => {
console.log(`Retrying repository ${repo.name} (attempt ${attempt}): ${error.message}`);
}
}
}
);
console.log("All repository mirroring tasks completed");
}, 0);
const responsePayload: MirrorRepoResponse = {

View File

@@ -10,6 +10,8 @@ import {
import { createGitHubClient } from "@/lib/github";
import { repoStatusEnum, repositoryVisibilityEnum } from "@/types/Repository";
import type { RetryRepoRequest, RetryRepoResponse } from "@/types/retry";
import { processWithRetry } from "@/lib/utils/concurrency";
import { createMirrorJob } from "@/lib/helpers";
export const POST: APIRoute = async ({ request }) => {
try {
@@ -65,10 +67,21 @@ export const POST: APIRoute = async ({ request }) => {
);
}
// Start background retry
// Start background retry with parallel processing
setTimeout(async () => {
for (const repo of repos) {
try {
// Create a single Octokit instance to be reused if needed
const octokit = config.githubConfig.token
? createGitHubClient(config.githubConfig.token)
: null;
// Define the concurrency limit - adjust based on API rate limits
const CONCURRENCY_LIMIT = 3;
// Process repositories in parallel with retry capability
await processWithRetry(
repos,
async (repo) => {
// Prepare repository data
const visibility = repositoryVisibilityEnum.parse(repo.visibility);
const status = repoStatusEnum.parse(repo.status);
const repoData = {
@@ -81,6 +94,20 @@ export const POST: APIRoute = async ({ request }) => {
forkedFrom: repo.forkedFrom ?? undefined,
};
// Log the start of retry operation
console.log(`Starting retry for repository: ${repo.name}`);
// Create a mirror job entry to track progress
await createMirrorJob({
userId: config.userId || "",
repositoryId: repo.id,
repositoryName: repo.name,
message: `Started retry operation for repository: ${repo.name}`,
details: `Repository ${repo.name} is now in the retry queue.`,
status: "imported",
});
// Determine if the repository exists in Gitea
let owner = getGiteaRepoOwner({
config,
repository: repoData,
@@ -93,16 +120,21 @@ export const POST: APIRoute = async ({ request }) => {
});
if (present) {
// If the repository exists, sync it
await syncGiteaRepo({ config, repository: repoData });
console.log(`Synced existing repo: ${repo.name}`);
} else {
// If the repository doesn't exist, mirror it
if (!config.githubConfig.token) {
throw new Error("GitHub token is missing.");
}
if (!octokit) {
throw new Error("Octokit client is not initialized.");
}
console.log(`Importing repo: ${repo.name} ${owner}`);
const octokit = createGitHubClient(config.githubConfig.token);
if (repo.organization && config.githubConfig.preserveOrgStructure) {
await mirrorGitHubOrgRepoToGiteaOrg({
config,
@@ -124,10 +156,28 @@ export const POST: APIRoute = async ({ request }) => {
});
}
}
} catch (err) {
console.error(`Failed to retry repo ${repo.name}:`, err);
return repo;
},
{
concurrencyLimit: CONCURRENCY_LIMIT,
maxRetries: 2,
retryDelay: 2000,
onProgress: (completed, total, result) => {
const percentComplete = Math.round((completed / total) * 100);
console.log(`Retry progress: ${percentComplete}% (${completed}/${total})`);
if (result) {
console.log(`Successfully processed repository: ${result.name}`);
}
},
onRetry: (repo, error, attempt) => {
console.log(`Retrying repository ${repo.name} (attempt ${attempt}): ${error.message}`);
}
}
}
);
console.log("All repository retry tasks completed");
}, 0);
const responsePayload: RetryRepoResponse = {

View File

@@ -5,6 +5,8 @@ import { eq, inArray } from "drizzle-orm";
import { repositoryVisibilityEnum, repoStatusEnum } from "@/types/Repository";
import { syncGiteaRepo } from "@/lib/gitea";
import type { SyncRepoResponse } from "@/types/sync";
import { processWithRetry } from "@/lib/utils/concurrency";
import { createMirrorJob } from "@/lib/helpers";
export const POST: APIRoute = async ({ request }) => {
try {
@@ -60,26 +62,66 @@ export const POST: APIRoute = async ({ request }) => {
);
}
// Start async mirroring in background
// Start async mirroring in background with parallel processing
setTimeout(async () => {
for (const repo of repos) {
try {
// Define the concurrency limit - adjust based on API rate limits
const CONCURRENCY_LIMIT = 5;
// Process repositories in parallel with retry capability
await processWithRetry(
repos,
async (repo) => {
// Prepare repository data
const repoData = {
...repo,
status: repoStatusEnum.parse(repo.status),
organization: repo.organization ?? undefined,
lastMirrored: repo.lastMirrored ?? undefined,
errorMessage: repo.errorMessage ?? undefined,
forkedFrom: repo.forkedFrom ?? undefined,
visibility: repositoryVisibilityEnum.parse(repo.visibility),
};
// Log the start of syncing
console.log(`Starting sync for repository: ${repo.name}`);
// Create a mirror job entry to track progress
await createMirrorJob({
userId: config.userId || "",
repositoryId: repo.id,
repositoryName: repo.name,
message: `Started syncing repository: ${repo.name}`,
details: `Repository ${repo.name} is now in the syncing queue.`,
status: "syncing",
});
// Sync the repository
await syncGiteaRepo({
config,
repository: {
...repo,
status: repoStatusEnum.parse(repo.status),
organization: repo.organization ?? undefined,
lastMirrored: repo.lastMirrored ?? undefined,
errorMessage: repo.errorMessage ?? undefined,
forkedFrom: repo.forkedFrom ?? undefined,
visibility: repositoryVisibilityEnum.parse(repo.visibility),
},
repository: repoData,
});
} catch (error) {
console.error(`Sync failed for repo ${repo.name}:`, error);
return repo;
},
{
concurrencyLimit: CONCURRENCY_LIMIT,
maxRetries: 2,
retryDelay: 2000,
onProgress: (completed, total, result) => {
const percentComplete = Math.round((completed / total) * 100);
console.log(`Syncing progress: ${percentComplete}% (${completed}/${total})`);
if (result) {
console.log(`Successfully synced repository: ${result.name}`);
}
},
onRetry: (repo, error, attempt) => {
console.log(`Retrying sync for repository ${repo.name} (attempt ${attempt}): ${error.message}`);
}
}
}
);
console.log("All repository syncing tasks completed");
}, 0);
const responsePayload: SyncRepoResponse = {