NewApp Monitoring is liveSign up free
AgentbeamBeta
Open-source package vulnerability · PyPI

CVE-2026-128089: dnspython has excessive memory and CPU use

highCVSS 3.1 7.5GHSA-vvf4-9x63-8xvv

dnspython before 2.9.2 does not check the service number in WKS records. A crafted zone file with a huge number makes the from_text() parser allocate a giant bitmap, which burns memory and CPU and can kill the process. Upgrade to dnspython 2.9.2.

Published Nov 13, 2026 · Analysis Nov 16, 2026

What an attacker can do

The from_text() parser for WKS records does not check that the service number is between 0 and 65535 inclusive. It uses the number to resize the record's bitmap so it can hold a bit at that offset. A very large value forces an extremely large allocation. The array is also extended inefficiently, which wastes CPU.

The attacker must control the record type and rdata text in zone-format input that your program parses. Examples are a tool that loads user-supplied zone files, or a service that builds records from text. They send a WKS line with an enormous service number. The program can run out of memory and exit. The impact is availability only.

The advisory's example is dns.rdata.from_text("IN", "WKS", "10.0.0.1 udp 65536"), which shows the missing check without causing heavy load. Code that only queries DNS and reads wire-format answers does not go through this text parser.

Who is affected

Affected: dnspython on pip below 2.9.2, in programs that parse zone-file or rdata text from untrusted input.

Check the version:

pip show dnspython

Search your code for dns.zone.from_text, dns.zone.from_file and dns.rdata.from_text on input you do not control. Remember that other packages depend on dnspython, so check transitive installs too.

dnspython

Ecosystem
PyPI
Affected
< 2.9.2
Fixed in
2.9.2
Check
pip show dnspython

How to fix it

Upgrade to dnspython 2.9.2:

pip install -U dnspython==2.9.2

The advisory names no workaround. If you cannot upgrade at once, reject WKS records, or any untrusted zone-format text, before it reaches the parser.

How to tell if you were hit

Look for Python processes that parse zone files and show sudden memory growth, out-of-memory kills or long CPU spikes. Check your input logs for WKS records with a service number above 65535. The advisory gives no other indicator.

Original advisory

From the GitHub Advisory Database (CC BY 4.0).

Summary

A maliciously crafted zone file could use a WKS record to consume excessive memory and CPU, possibly causing the process to exit due to memory exhaustion.

Details

The from_text() parser does not check that a WKS record service number is between 0 and 65535 inclusive. It then uses this number to resize the bitmap to accommodate a bit with that offset. This can cause extremely large memory allocations, and also extends the array very inefficiently, wasting CPU.

Proof of concept

This demonstrates the failure to limit the service number, though with a value that does not cause excessive memory or CPU consumption.

rd = dns.rdata.from_text("IN", "WKS", "10.0.0.1 udp 65536")

Impact

If dnspython is used to read zonefile format where the attacker can control the type and rdata text, the program may consume excessive memory or CPU, denying service.

Credited to manus-use.

Questions

Does this affect normal DNS lookups with dnspython?

The advisory describes the zone-file text parser, from_text(). It affects programs that read zone-format text where an attacker controls the record type and rdata.

Which version fixes CVE-2026-128089?

dnspython 2.9.2. Versions below it are affected.

What does the attacker get?

Denial of service. A WKS service number out of range makes the parser allocate a huge bitmap, using memory and CPU, and the process may exit.

Is it exploited in the wild?

The advisory does not say so, and the entry has no CISA KEV data. Its EPSS score is 0.0053.

References

How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. The analysis is written by an AI model from the sources listed above and checked by automated rules before it is published. Check versions against the vendor's advisory before you act. Methodology

More PyPI vulnerabilities

All PyPI entries