Service eligibility and regional restrictions
PuppyIP serves only compliant overseas businesses and their authorized personnel. Proxy services are not available in mainland China. The service may only be used for lawful business activities outside mainland China. Use of this service within mainland China is prohibited.
Hosting a proxy IP or server overseas does not change these restrictions. The service must not be provided to end users in mainland China through relaying, forwarding, sharing or resale. Before use, read the Terms of Service.
Key Takeaways
- Cloudflare announced the move to Observability > Logs on October 7, 2026. This guide follows the current documentation; it does not report a new launch.
- Existing datasets, SQL queries and saved searches continue to work. Dataset management is under Configure beside the selector.
- A shared query interface does not mean every dataset supports SQL. Current documentation limits SQL to Log Explorer datasets.
- Log Explorer is a paid add-on with no free version or trial. Ingestion and storage are billed by GB; query frequency itself does not add a query charge.
Where did Log Explorer go? Start at the new Logs entry
In the Cloudflare dashboard, open Observability > Logs, then use the dataset selector to find the relevant dataset under Log Explorer, such as HTTP requests. The October 7, 2026 announcement says the old Log Explorer menu was removed. This consolidates query navigation; a missing menu alone does not establish data loss.
Current product documentation labels Log Explorer as Legacy because it is being merged into Observability Logs, and explicitly says existing datasets, subscriptions and API access continue to work. The Legacy label alone is not a reason to buy the product again, delete datasets or migrate an existing API integration.
The Logs page also contains data from products such as Workers and Containers. Identify which logs you need first: sharing an interface does not make fields, activation requirements or retention identical. Whether a particular dataset is enabled still depends on the actual account settings.
Narrow the dataset, domain and time before investigating empty results
After choosing a dataset, select the domain you are investigating, or all applicable domains, if that dataset is scoped by zone. A zone generally represents a site added to Cloudflare. Account-level datasets and individual site datasets are not interchangeable just because their names look similar.
Use a preset, relative or calendar time range to focus on when the problem happened. Then add filters with the field browser by choosing a field, operator and value. For example, locate the relevant HTTP request before inspecting its method and response status.
If results are empty, check the dataset, site scope, time range and filters first. Then check whether logging was enabled in the source product and whether the records remain within retention. Cloudflare says some datasets must first be enabled in their product. A visible menu or a successful query does not establish that every historical request was captured.
Looking up a Ray ID? Use SQL only with a Log Explorer dataset
In the target Log Explorer dataset, select Open SQL editor, enter the query and choose Run query. A Ray ID is an identifier Cloudflare uses for a request. The following read-only SQL is shortened from the official example; replace the placeholder with an actual Ray ID you are authorized to investigate.
The example returns only the request scheme, hostname, method and edge response status, and changes no data. It targets http_requests rather than every dataset. Select the appropriate site and time range in the page first. An empty result still requires the collection, scope and retention checks above; it does not establish that the request never happened.
Current Log Search documentation explicitly limits SQL to Log Explorer datasets. Workers logs may share filtering and visualization controls, but that does not establish that this SQL can run against Workers data.
Find one HTTP request by Ray ID
textSELECT
clientRequestScheme,
clientRequestHost,
clientRequestMethod,
edgeResponseStatus
FROM http_requests
WHERE RayID = 'REPLACE_WITH_RAY_ID'
LIMIT 1
Where are old queries, Save search and Configure?
Look under Saved searches on the Logs page for existing queries, including those previously saved in Log Search. To save a new filter or SQL query, use Save search and give it a recognizable name and description. The original announcement says existing SQL and saved queries continue to work.
To manage datasets, find Log Explorer in the dataset selector and use Configure beside it. Reading logs, saving queries and enabling or disabling datasets are different operations. Finding the new entry does not require stopping ingestion or changing a dataset.
If you can query but cannot save or manage datasets, check permissions first. Current documentation allows roles such as Log Share Reader to read logs, while saving queries and managing datasets generally require Super Administrator or Administrator. It separately describes changes that Log Share can make to zone-level datasets and saved queries.
API token permissions such as Logs Read and Logs Edit are not member roles in the role picker. The API name Logs Write refers to the same permission as Logs Edit. Account-level datasets require account-level authorization; a role assigned to one zone covers only that zone. Datasets such as Zero Trust and audit logs can require additional member roles. Follow the current permission table instead of broadening unrelated access.
The new entry does not change billing, retention or log privacy
Log Explorer is currently a paid add-on for any Application Services or Zero Trust purchase, and its pricing page explicitly says there is no free version or trial. Charges are based on GB of logs ingested and stored. Searching or scanning data already ingested and stored does not itself add a query charge. A shared Logs entry does not make all log products free.
Fields and retention vary by dataset. The product overview separately says contract customers can buy storage for up to two years at an additional USD 0.10 per GB per month. That is a contract extension, not a default retention entitlement for every account, and it does not replace the ingestion, storage and tax terms in an actual contract. The reviewed sources give no universal country availability list; confirm specific eligibility and costs separately.
Finding the menu and query entry does not require capturing request headers, cookies or access tokens. Cloudflare supports separately configured custom fields, but a navigation issue is not a reason to collect more sensitive data. Before sharing logs, redact credentials, private links and unrelated personal information, retaining only the fields needed to explain the issue.
Sources
- Cloudflare changelog: Log Search moves into Observability Logs, October 7, 2026
- Cloudflare Logs: datasets, field filters and time ranges
- Cloudflare Log Search: SQL, saved searches and query optimization
- Cloudflare Log Explorer (Legacy): migration and permission table
- Cloudflare Log Explorer: availability, pricing and usage management
Frequently Asked Questions
Can I still use an old direct Log Search link after the dashboard move?
The October 7 announcement says the previous Log Search remains available at its direct URL. For routine navigation, follow the current documentation to Observability > Logs. This guide did not sign in to verify your account’s old bookmarks or their current redirects, so it cannot guarantee that every historical link works.
Does a log record prove a request was blocked or the problem is fixed?
No. Results are observations from the selected dataset and must be interpreted using the field definitions, rule actions and actual business outcome. This guide explains navigation and retrieval, performs no policy changes and makes no claim of a fix or blocking result.