1787616 - Crash in [@ builtins_mdObject_GetAttributeSize]

Status: RESOLVED FIXED

(NSS :: Libraries, defect, P3)

Description

[Wayne Mery (:wsmwk)]

Occurs in 91, 102 and beta 103 and 104.

Crash report: https://crash-stats.mozilla.org/report/index/879f7a8d-041f-4b3c-a0fb-3713b0220813

Reason: EXCEPTION_ACCESS_VIOLATION_READ

Top 10 frames of crashing thread:

0 nssckbi.dll builtins_mdObject_GetAttributeSize security/nss/lib/ckfw/builtins/bobject.c:110
1 nssckbi.dll nssCKFWObject_GetAttributeSize security/nss/lib/ckfw/object.c:542
2 nssckbi.dll NSSCKFWC_GetAttributeValue security/nss/lib/ckfw/wrap.c:2244
3 nssckbi.dll builtinsC_GetAttributeValue security/nss/lib/ckfw/nssck.api:630
4 nss3.dll PK11_ReadAttribute security/nss/lib/pk11wrap/pk11obj.c:106
5 nss3.dll PK11_FindRawCertsWithSubject security/nss/lib/pk11wrap/pk11obj.c:1994
6 xul.dll mozilla::psm::NSSCertDBTrustDomain::FindIssuer security/certverifier/NSSCertDBTrustDomain.cpp:242
7 xul.dll mozilla::pkix::BuildForward security/nss/lib/mozpkix/lib/pkixbuild.cpp:345
8 xul.dll mozilla::pkix::PathBuildingStep::Check security/nss/lib/mozpkix/lib/pkixbuild.cpp:205
9 xul.dll mozilla::psm::CheckCandidates security/certverifier/NSSCertDBTrustDomain.cpp:181

bp-17359622-6e81-47aa-9095-0633a0220715 103.0 beta

Comment 1

[John Schanck [:jschanck]]

This is an NSS bug.

Comment 2

[John Schanck [:jschanck]]

This was fixed in 108 by Bug 1789520.

Comment 3

[Wayne Mery (:wsmwk)]

Thanks.
This is a top ~40 crash for Thunderbird 102.*, but I can see it is not a frequent crash for Firefox.
However, beta crash rate is only one per week. So I estimate two months or more before we can confirm this is gone for Thunderbird.

Also, FWIW, according to crash stats the average user only crashes once, so numbers are probably not inflated by multiple crashes per user.

Comment 4

[Wayne Mery (:wsmwk)]

It looks likely that both crashes signatures ended starting Thunderbird beta 108, so consistent with bug 1789520 being fixed in 108.
But I'm guessing esr102 won't get this nss version?

Comment 5

[Ryan VanderMeulen [:RyanVM]]

That'd be a question for the NSS folks, but that doesn't seem like a great backport candidate IMHO.

Comment 6

[John Schanck [:jschanck]]

We won't backport the fix for this.