spring-projects/spring-ai · error · CachedContentException

Cached content not found:

Error message

Cached content not found: 

What it means

extendTtl(name, additionalTtl) first loads the cached content via get(name); if it returns null (name not found or already expired/removed) it throws CachedContentException directly with the name in the message. This is a pre-condition check, not a wrapped remote error.

Source

Thrown at models/spring-ai-google-genai/src/main/java/org/springframework/ai/google/genai/cache/GoogleGenAiCachedContentService.java:387

			return CompletableFuture.completedFuture(false);
		}
	}

	// Utility methods

	/**
	 * Extends the TTL of cached content by the specified duration.
	 * @param name the cached content name
	 * @param additionalTtl the additional TTL to add
	 * @return the updated cached content
	 */
	public @Nullable GoogleGenAiCachedContent extendTtl(String name, Duration additionalTtl) {
		Assert.hasText(name, "Name must not be empty");
		Assert.notNull(additionalTtl, "Additional TTL must not be null");

		GoogleGenAiCachedContent existing = get(name);
		if (existing == null) {
			throw new CachedContentException("Cached content not found: " + name);
		}

		Instant newExpireTime = existing.getExpireTime() != null ? existing.getExpireTime().plus(additionalTtl)
				: Instant.now().plus(additionalTtl);

		CachedContentUpdateRequest updateRequest = CachedContentUpdateRequest.builder()
			.expireTime(newExpireTime)
			.build();

		return update(name, updateRequest);
	}

	/**
	 * Refreshes the expiration of cached content to the maximum TTL.
	 * @param name the cached content name
	 * @param maxTtl the maximum TTL to set
	 * @return the updated cached content
	 */

View on GitHub (pinned to 98a7beda4f)

Solutions

1. Call get(name) before extendTtl and recreate the cached content if null
2. Refresh TTLs at intervals shorter than the entry TTL
3. Verify the resource name and project/region
4. Catch CachedContentException and treat as recreation signal

Example fix

// before
service.extendTtl(name, Duration.ofHours(1));
// after
if (service.get(name) == null) {
    service.create(buildCreateRequest(name)); // recreate instead
} else {
    service.extendTtl(name, Duration.ofHours(1));
}

GoogleGenAiCachedContent existing = service.get(name);
if (existing == null) {
    // recreate or skip; do not call extendTtl
} else {
    service.extendTtl(name, additionalTtl);
}

try {
    service.extendTtl(name, additionalTtl);
} catch (CachedContentException e) {
    logger.warn("cache entry {} gone; recreating", name);
    recreateCache(name);
}

When it happens

Trigger: Calling extendTtl with a name that does not exist, was deleted, or whose TTL already lapsed so Google removed the resource.

Common situations: Background jobs refreshing TTL of cache entries that expired between runs, typos in fully-qualified resource names, or caches created in a different project/region.

Understand the failure class

Background: 'Could not be found', 'does not exist', 'not found in database': the resource-not-found family when an ID, slug, key, or URI lookup comes back empty — this error's family across 20 libraries.

Related errors

• Failed to update cached content:
• Failed to create cached content
• Failed to list cached content
• Vertex AI mode requires both 'project-id' and 'location' to
• Incomplete Google GenAI configuration: Provide 'api-key' for

AI-assisted analysis of spring-projects/spring-ai@98a7beda4f (2026-09-11). Data as JSON: /api/errors/3a9dcad868b241b1. Report an issue: GitHub.