Every CVESearch/Get API key

DHI-5sov-dxuz-izqw-ojbm

`BytesStart::attributes()` returns an `Attributes` iterator which, by default

7.5 HighFix available

DHI-5sov-dxuz-izqw-ojbm is a high vulnerability (CVSS 7.5) in rust, published June 29, 2026. No source lists public exploit code. Fixed in quick-xml 0.41.0.

What does this mean?
Severity
7.5 HighCVSS 3.1, OSV
Exploit chance
UnscoredNo EPSS score yet
Exploitation
None knownNot on CISA’s or ENISA’s lists
Fixed in
0.41.0quick-xml +11 more

Description

BytesStart::attributes() returns an Attributes iterator which, by default (with_checks(true)), rejects a start tag that repeats an attribute name. For each attribute yielded, the iterator compared the new name against every name seen so far in the same tag using a linear scan, so a start tag with N distinct attribute names cost O(N²) byte comparisons. There was no bound on N other than the size of the buffered start tag.

Impact

Any code that parses untrusted XML and iterates a start tag's attributes with the default duplicate check enabled can be made to spend CPU time quadratic in the number of attributes on a single tag. Because the check is pure computation with no .await/I/O, an I/O-based timeout on the consumer (for example a read or request timeout) cannot interrupt it while it runs.

Measured cost of a single start tag, release build:

| Attributes on one tag | Time | |---|---| | 80,000 | ~6 s | | 800,000 | ~10 min |

The cost grows with the square of the attribute count, so a start tag of a few tens of megabytes can stall a parsing thread for hours. No memory is exhausted and the parser does not crash; the effect is CPU exhaustion on the thread doing the parsing: a single crafted start tag can pin a CPU core for minutes to hours, denying service to that worker. A deployment that places a wall-clock bound on parsing, or confines it to a non-critical thread, may consider the availability impact lower.

Affected code paths

  • BytesStart::attributes() / Attributes iterated with checks enabled (the default), and BytesStart::try_get_attribute.
  • NsReader, which resolves namespaces by iterating a tag's attributes and so reaches the same check internally.

Consumers that iterate attributes with .attributes().with_checks(false) and do not use NsReader are not affected.

This was reported as reachable by a remote, unauthenticated attacker in a real-world RPKI relying party (NLnet Labs Routinator) via a crafted RRDP snapshot.xml.

Remediation

Upgrade to quick-xml >= 0.41.0, where the duplicate check keeps the linear scan for start tags with a small number of attributes and switches to an O(1) hash pre-filter above a threshold, making the whole tag O(N). The reported AttrError::Duplicated positions are unchanged.

If upgrading is not possible and duplicate-name detection is not required, disable it with .attributes().with_checks(false) (this does not help NsReader consumers, which have no equivalent opt-out before 0.41.0).

Attack profile

How easy it is to exploit

Attack vector
Network
Complexity
Low
Privileges needed
None
User interaction
None

What an attacker can affect

Confidentiality
None
Integrity
None
Availability
High

Affected packages

Affected packages
PackageAffectedFixed
gitlab-rails-18
Docker Hardened Images:Alpine:3.24
Listed versions
18.11.11-r018.11.11-r118.11.11-r218.11.12-r018.11.12-r1
Show 15 moreShow less
18.11.12-r218.11.12-r318.11.12-r418.11.14-r018.11.14-r118.11.14-r1018.11.14-r1118.11.14-r218.11.14-r318.11.14-r418.11.14-r518.11.14-r618.11.14-r718.11.14-r818.11.14-r9
None known
gitlab-rails-19
Docker Hardened Images:Alpine:3.24
Listed versions
19.1.8-r019.1.8-r119.4.1-r019.4.1-r119.4.1-r2
Show 16 moreShow less
19.4.1-r319.4.2-r019.5.3-r019.5.3-r119.5.3-r219.5.3-r319.5.3-r419.5.3-r519.5.3-r619.5.3-r719.5.3-r819.5.4-r019.5.4-r119.5.5-r019.5.5-r119.5.5-r2
None known
rust
Docker Hardened Images:Alpine:3.23
Listed versions
1.91.1-r01.91.1-r21.91.1-r31.91.1-r41.93.1-r0
Show 13 moreShow less
1.93.1-r11.94.0-r01.94.0-r11.94.1-r01.94.1-r11.95.0-r01.96.0-r01.96.1-r01.97.0-r01.97.0-r11.98.1-r01.99.0-r21.99.1-r0
None known
gitlab-rails-18
Docker Hardened Images:Debian:13
Listed versions
18.11.11-018.11.11-118.11.11-218.11.12-018.11.12-1
Show 19 moreShow less
18.11.12-218.11.14-018.11.14-118.11.14-1018.11.14-1118.11.14-1218.11.14-1318.11.14-1418.11.14-1518.11.14-1618.11.14-1718.11.14-218.11.14-318.11.14-418.11.14-518.11.14-618.11.14-718.11.14-818.11.14-9
None known
gitlab-rails-19
Docker Hardened Images:Debian:13
Listed versions
19.1.8-019.1.8-119.4.1-019.4.1-119.4.1-2
Show 21 moreShow less
19.4.2-019.5.3-019.5.3-119.5.3-219.5.3-319.5.3-419.5.3-519.5.3-619.5.3-719.5.3-819.5.3-919.5.4-019.5.4-119.5.5-019.5.5-119.5.5-219.5.5-319.5.5-419.5.5-519.5.5-619.5.5-7
None known
milvus-3.0-fips-libs
Docker Hardened Images:Debian:13
Listed versions
3.0.1-13.0.2-03.0.2-13.0.2-23.0.2-3
Show 12 moreShow less
3.0.2-43.0.3-03.0.5-03.0.5-13.0.5-23.0.6-03.0.6-13.0.6-23.0.6-33.0.6-43.0.6-53.0.6-6
None known
milvus-3.0-fips
Docker Hardened Images:Debian:13
Listed versions
3.0.1-13.0.2-03.0.2-13.0.2-23.0.2-3
Show 12 moreShow less
3.0.2-43.0.3-03.0.5-03.0.5-13.0.5-23.0.6-03.0.6-13.0.6-23.0.6-33.0.6-43.0.6-53.0.6-6
None known
milvus-3.0-libs
Docker Hardened Images:Debian:13
Listed versions
3.0.1-13.0.2-03.0.2-13.0.2-23.0.2-3
Show 12 moreShow less
3.0.2-43.0.3-03.0.5-03.0.5-13.0.5-23.0.6-03.0.6-13.0.6-23.0.6-33.0.6-43.0.6-53.0.6-6
None known
milvus-3.0
Docker Hardened Images:Debian:13
Listed versions
3.0.1-13.0.2-03.0.2-13.0.2-23.0.2-3
Show 12 moreShow less
3.0.2-43.0.3-03.0.5-03.0.5-13.0.5-23.0.6-03.0.6-13.0.6-23.0.6-33.0.6-43.0.6-53.0.6-6
None known
miniconda3-26
Docker Hardened Images:Debian:13
Listed versions
26.7.1-1-0
None known
vector
Docker Hardened Images:Debian:13
Listed versions
0.58.0-00.59.0-00.60.0-00.61.0-0
None known
quick-xml
crates.io
0.0.0-0 up to 0.41.0
0.41.0

What each source reports

OSV Score used7.5

`BytesStart::attributes()` returns an `Attributes` iterator which, by default

BytesStart::attributes() returns an Attributes iterator which, by default (with_checks(true)), rejects a start tag that repeats an attribute name. For each attribute yielded, the iterator compared the new name against every name seen so far in the same tag using a linear scan, so a start tag with N distinct attribute names cost O(N²) byte comparisons. There was no bound on N other than the size of the buffered start tag.

Impact

Any code that parses untrusted XML and iterates a start tag's attributes with the default duplicate check enabled can be made to spend CPU time quadratic in the number of attributes on a single tag. Because the check is pure computation with no .await/I/O, an I/O-based timeout on the consumer (for example a read or request timeout) cannot interrupt it while it runs.

Measured cost of a single start tag, release build:

| Attributes on one tag | Time | |---|---| | 80,000 | ~6 s | | 800,000 | ~10 min |

The cost grows with the square of the attribute count, so a start tag of a few tens of megabytes can stall a parsing thread for hours. No memory is exhausted and the parser does not crash; the effect is CPU exhaustion on the thread doing the parsing: a single crafted start tag can pin a CPU core for minutes to hours, denying service to that worker. A deployment that places a wall-clock bound on parsing, or confines it to a non-critical thread, may consider the availability impact lower.

Affected code paths

  • BytesStart::attributes() / Attributes iterated with checks enabled (the default), and BytesStart::try_get_attribute.
  • NsReader, which resolves namespaces by iterating a tag's attributes and so reaches the same check internally.

Consumers that iterate attributes with .attributes().with_checks(false) and do not use NsReader are not affected.

This was reported as reachable by a remote, unauthenticated attacker in a real-world RPKI relying party (NLnet Labs Routinator) via a crafted RRDP snapshot.xml.

Remediation

Upgrade to quick-xml >= 0.41.0, where the duplicate check keeps the linear scan for start tags with a small number of attributes and switches to an O(1) hash pre-filter above a threshold, making the whole tag O(N). The reported AttrError::Duplicated positions are unchanged.

If upgrading is not possible and duplicate-name detection is not required, disable it with .attributes().with_checks(false) (this does not help NsReader consumers, which have no equivalent opt-out before 0.41.0).

Record
DHI-5sov-dxuz-izqw-ojbm
CVSS 3.1
7.5 HighCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Updated
2026-11-16
OSV Score usedRUSTSEC-2026-01947.5

Quadratic run time when checking a start tag for duplicate attribute names

BytesStart::attributes() returns an Attributes iterator which, by default (with_checks(true)), rejects a start tag that repeats an attribute name. For each attribute yielded, the iterator compared the new name against every name seen so far in the same tag using a linear scan, so a start tag with N distinct attribute names cost O(N²) byte comparisons. There was no bound on N other than the size of the buffered start tag.

Impact

Any code that parses untrusted XML and iterates a start tag's attributes with the default duplicate check enabled can be made to spend CPU time quadratic in the number of attributes on a single tag. Because the check is pure computation with no .await/I/O, an I/O-based timeout on the consumer (for example a read or request timeout) cannot interrupt it while it runs.

Measured cost of a single start tag, release build:

| Attributes on one tag | Time | |---|---| | 80,000 | ~6 s | | 800,000 | ~10 min |

The cost grows with the square of the attribute count, so a start tag of a few tens of megabytes can stall a parsing thread for hours. No memory is exhausted and the parser does not crash; the effect is CPU exhaustion on the thread doing the parsing: a single crafted start tag can pin a CPU core for minutes to hours, denying service to that worker. A deployment that places a wall-clock bound on parsing, or confines it to a non-critical thread, may consider the availability impact lower.

Affected code paths

  • BytesStart::attributes() / Attributes iterated with checks enabled (the default), and BytesStart::try_get_attribute.
  • NsReader, which resolves namespaces by iterating a tag's attributes and so reaches the same check internally.

Consumers that iterate attributes with .attributes().with_checks(false) and do not use NsReader are not affected.

This was reported as reachable by a remote, unauthenticated attacker in a real-world RPKI relying party (NLnet Labs Routinator) via a crafted RRDP snapshot.xml.

Remediation

Upgrade to quick-xml >= 0.41.0, where the duplicate check keeps the linear scan for start tags with a small number of attributes and switches to an O(1) hash pre-filter above a threshold, making the whole tag O(N). The reported AttrError::Duplicated positions are unchanged.

If upgrading is not possible and duplicate-name detection is not required, disable it with .attributes().with_checks(false) (this does not help NsReader consumers, which have no equivalent opt-out before 0.41.0).

Record
RUSTSEC-2026-0194
CVSS 3.1
7.5 HighCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Updated
2026-11-02

History

History
RecordedChangedSources
affected, aliases, facts, id, references, titleOSV
first indexedOSV

Dates are when Every CVE recorded each change. Upstream dates may differ.

References

MoveEnter Open