ghsa

A vulnerability in Apache CXF before versions 3.5.5 and 3.4.10 allows an attacker to perform a remote directory listing or code exfiltration. The vulnerability only applies when the CXFServlet is configured with both the static-resources-list and redirect-query-check attributes. These attributes are not supposed to be used together, and so the vulnerability can only arise if the CXF service is misconfigured.

A stack overflow in the org.json.JSONTokener.nextValue::JSONTokener.java component of hutool-json v5.8.10 allows attackers to cause a Denial of Service (DoS) via crafted JSON or XML data.

A stack overflow in the XML.toJSONObject component of hutool-json v5.8.10 allows attackers to cause a Denial of Service (DoS) via crafted JSON or XML data.

hutool-json v5.8.10 was discovered to contain an out of memory error. This issue is similar to CVE-2022-45690.

### Impact The linux-loader crate used the offsets and sizes provided in the ELF headers to determine the offsets to read from. If those offsets pointed beyond the end of the file this could lead to an infinite loop. Virtual Machine Monitors using the `linux-loader` crate could enter an infinite loop if the ELF header of the kernel they are loading was modified in a malicious manner. ### Patches The issue has been addressed in 0.8.1 ### Workarounds The issue can be mitigated by ensuring that only trusted kernel images are loaded or by verifying that the headers to not point beyond the end of the file. ### References See: https://github.com/rust-vmm/linux-loader/pull/125

A problem with Candy Machine V2 allow minting NFTs to an arbitrary collection due to a missing check. Here is a description of the exploit: Details: Here is the tx/ix to exploit: Transaction: Ix 1: candy_machine v2, mint_nft, passing in empty metadata -1 Ix 2: custom handler, 0 cpi A --> token_metadata create_metadata_account, creates NFT cpi B --> candy_machine v2, set_collection_during_mint Ix 1 passes our first check for empty metadata, but eventually will hit a bot tax and return Ok. We do have a CPI check in this function but even if we hit that or moved it to the top, it returns Ok as a bot tax and still enables the issue. Ix 2, cpi A is Ok and mints an arbitrary NFT. Ix 2, cpi B checks the previous instruction using index_relative_to_current-1. This turns out to be Ix 1 which was Ok, so then your newly minted arbitrary NFT is successfully added to the collection. Conclusion: Candy machine could be out of NFTs and it still works. If the CM is closed, (we think?) it doesn'...

This was an error found by @metamania01 of the Audit Company Solshield. It allowed one to verify a creator that did not sign by making use of a provision in Token Metadata that allows Creators who have signed compressed nfts to allow them to decompress with verified creators. The issue is now patched. For more info see. https://twitter.com/thehasheddude/status/1601642138143375360

### Impact All authenticated Cube clients could bypass row-level security and run arbitrary SQL via the newly introduced /v1/sql-runner endpoint. ### Patches The change has been reverted in 0.31.24 ### Workarounds Upgrade to >=0.31.24 or downgrade to <=0.31.22 ### Post mortem As part of implementing the Cube Cloud SQL runner functionality, we’ve added a new endpoint to the Cube Core so that we could add arbitrary queries directly to the queue, bypassing the modeling layer. The endpoint was added in this commit: https://github.com/cube-js/cube.js/commit/f1e25bb50323c0b99f3891d349467e7b637baeea It went through the code review; however, it slipped everyone’s attention that this endpoint completely bypasses any row-level security logic implemented in the modeling layer. Now anyone with a valid Cube JWT token could fetch any data, even if they were not allowed to do so by their security context. The issue was noticed by the Core team on Dec 12 and immediately reverted. The just-relea...

With a known valid invite link (i.e. not already accepted or expired) an unauthenticated attacker can manipulate the cookie to allow the same invite link to be reused on multiple accounts when joining an organization. ### Impact An attacker with a valid invite link can create multiple users and join the organization from which the invite link was generated. ### Patches This issue was patched in version 22.11.0. ### Workarounds Sentry SaaS customers do not need to take action. Self-hosted Sentry installs can disable the invite functionality until they are ready to deploy the patched version by editing their `sentry.conf.py` file (usually located at `~/.sentry/`). 1. Add the following line into `sentry.conf.py`: ```python SENTRY_FEATURES["organizations:invite-members"] = False ``` 2. Restart the Sentry web service. ``` docker compose restart web ``` ### For more information If you have any questions or comments about this advisory: * [Visit our FAQs on th...

### Impact When calling `DefaultHttpHeaders.set` with an _iterator_ of values (as opposed to a single given value), header value validation was not performed, allowing malicious header values in the iterator to perform [HTTP Response Splitting](https://owasp.org/www-community/attacks/HTTP_Response_Splitting). ### Patches The necessary validation was added in Netty 4.1.86.Final. ### Workarounds Integrators can work around the issue by changing the `DefaultHttpHeaders.set(CharSequence, Iterator<?>)` call, into a `remove()` call, and call `add()` in a loop over the iterator of values. ### References [HTTP Response Splitting](https://owasp.org/www-community/attacks/HTTP_Response_Splitting) [CWE-113: Improper Neutralization of CRLF Sequences in HTTP Headers](https://cwe.mitre.org/data/definitions/113.html) ### For more information If you have any questions or comments about this advisory: * Open an issue in [[example link to repo](https://github.com/netty/netty)](https://github.com/nett...