Vulnerabilities

Not properly invalidated session vulnerability in Graylog Web Interface, version 2.2.3, due to incorrect management of session invalidation after new logins. The application generates a new 'sessionId' each time a user authenticates, but does not invalidate previously issued session identifiers, which remain valid even after multiple consecutive logins by the same user. As a result, a stolen or leaked 'sessionId' can continue to be used to authenticate valid requests. Exploiting this vulnerability would allow an attacker with access to the web service/API network (port 9000 or HTTP/S endpoint of the server) to reuse an old session token to gain unauthorized access to the application, interact with the API/web, and compromise the integrity of the affected account.

Graylog is a free and open log management platform. In versions 6.2.0 to before 6.2.4 and 6.3.0-alpha.1 to before 6.3.0-rc.2, Graylog users can gain elevated privileges by creating and using API tokens for the local Administrator or any other user for whom the malicious user knows the ID. For the attack to succeed, the attacker needs a user account in Graylog. They can then proceed to issue hand-crafted requests to the Graylog REST API and exploit a weak permission check for token creation. This issue has been patched in versions 6.2.4 and 6.3.0-rc.2. A workaround involves disabling the respective configuration found in System > Configuration > Users > "Allow users to create personal access tokens".

Graylog is a free and open log management platform. Prior to versions 6.0.14, 6.1.10, and 6.2.0, it is possible to obtain user session cookies by submitting an HTML form as part of an Event Definition Remediation Step field. For this attack to succeed, the attacker needs a user account with permissions to create event definitions, while the user must have permissions to view alerts. Additionally, an active Input must be present on the Graylog server that is capable of receiving form data (e.g. a HTTP input, TCP raw or syslog etc). Versions 6.0.14, 6.1.10, and 6.2.0 fix the issue. No known workarounds are available, as long as the relatively rare prerequisites are met.

Graylog is a free and open log management platform. Starting with 6.1, HTTP Inputs can be configured to check if a specified header is present and has a specified value to authenticate HTTP-based ingestion. Unfortunately, even though in cases of a missing header or a wrong value the correct HTTP response (401) is returned, the message will be ingested nonetheless. To mitigate the vulnerability, disable http-based inputs and allow only authenticated pull-based inputs. This vulnerability is fixed in 6.1.9.

Graylog is a free and open log management platform. The reporting functionality in Graylog allows the creation and scheduling of reports which contain dashboard widgets displaying individual log messages or metrics aggregated from fields of multiple log messages. This functionality, as included in Graylog 6.1.0 & 6.1.1, is vulnerable to information leakage triggered by multiple concurrent report rendering requests from authorized users. When multiple report renderings are requested at the same start time, the headless browser instance used to render the PDF will be reused. Depending on the timing, either a check for the browser instance "freshness" hits, resulting in an error instead of the report being returned, or one of the concurrent report rendering requests "wins" and this report is returned for all report rendering requests that do not return an error. This might lead to one user getting the report of a different user, potentially leaking indexed log messages or aggregated data that this user normally has no access to. This problem is fixed in Graylog 6.1.2. There is no known workaround besides disabling the reporting functionality.

Graylog is a free and open log management platform. Starting in version 2.0.0 and prior to versions 5.1.11 and 5.2.4, arbitrary classes can be loaded and instantiated using a HTTP PUT request to the `/api/system/cluster_config/` endpoint. Graylog's cluster config system uses fully qualified class names as config keys. To validate the existence of the requested class before using them, Graylog loads the class using the class loader. If a user with the appropriate permissions performs the request, arbitrary classes with 1-arg String constructors can be instantiated. This will execute arbitrary code that is run during class instantiation. In the specific use case of `java.io.File`, the behavior of the internal web-server stack will lead to information exposure by including the entire file content in the response to the REST request. Versions 5.1.11 and 5.2.4 contain a fix for this issue.

Graylog is a free and open log management platform. Graylog makes use of only one single source port for DNS queries. Graylog binds a single socket for outgoing DNS queries and while that socket is bound to a random port number it is never changed again. This goes against recommended practice since 2008, when Dan Kaminsky discovered how easy is to carry out DNS cache poisoning attacks. In order to prevent cache poisoning with spoofed DNS responses, it is necessary to maximise the uncertainty in the choice of a source port for a DNS query. Although unlikely in many setups, an external attacker could inject forged DNS responses into a Graylog's lookup table cache. In order to prevent this, it is at least recommendable to distribute the DNS queries through a pool of distinct sockets, each of them with a random source port and renew them periodically. This issue has been addressed in versions 5.0.9 and 5.1.3. Users are advised to upgrade. There are no known workarounds for this issue.

Graylog is a free and open log management platform. In a multi-node Graylog cluster, after a user has explicitly logged out, a user session may still be used for API requests until it has reached its original expiry time. Each node maintains an in-memory cache of user sessions. Upon a cache-miss, the session is loaded from the database. After that, the node operates solely on the cached session. Modifications to sessions will update the cached version as well as the session persisted in the database. However, each node maintains their isolated version of the session. When the user logs out, the session is removed from the node-local cache and deleted from the database. The other nodes will however still use the cached session. These nodes will only fail to accept the session id if they intent to update the session in the database. They will then notice that the session is gone. This is true for most API requests originating from user interaction with the Graylog UI because these will lead to an update of the session's "last access" timestamp. If the session update is however prevented by setting the `X-Graylog-No-Session-Extension:true` header in the request, the node will consider the (cached) session valid until the session is expired according to its timeout setting. No session identifiers are leaked. After a user has logged out, the UI shows the login screen again, which gives the user the impression that their session is not valid anymore. However, if the session becomes compromised later, it can still be used to perform API requests against the Graylog cluster. The time frame for this is limited to the configured session lifetime, starting from the time when the user logged out. This issue has been addressed in versions 5.0.9 and 5.1.3. Users are advised to upgrade.

A Session ID leak in the audit log in Graylog before 4.1.2 allows attackers to escalate privileges (to the access level of the leaked session ID).

Graylog before 3.3.3 lacks SSL Certificate Validation for LDAP servers. It allows use of an external user/group database stored in LDAP. The connection configuration allows the usage of unencrypted, SSL- or TLS-secured connections. Unfortunately, the Graylog client code (in all versions that support LDAP) does not implement proper certificate validation (regardless of whether the "Allow self-signed certificates" option is used). Therefore, any attacker with the ability to intercept network traffic between a Graylog server and an LDAP server is able to redirect traffic to a different LDAP server (unnoticed by the Graylog server due to the lack of certificate validation), effectively bypassing Graylog's authentication mechanism.

In Graylog before 2.4.6, XSS was possible in typeahead components, related to components/common/TypeAheadInput.jsx and components/search/QueryInput.ts.

Graylog before v2.4.4 has an XSS security issue with unescaped text in notifications, related to toastr and util/UserNotification.js.