|
Bugzilla – Full Text Bug Listing |
| Summary: | VUL-0: CVE-2021-28878: rust: memory safety violation in the Zip implementation related to next_back() and next() | ||
|---|---|---|---|
| Product: | [Novell Products] SUSE Security Incidents | Reporter: | Robert Frohl <rfrohl> |
| Component: | Incidents | Assignee: | William Brown <william.brown> |
| Status: | RESOLVED FIXED | QA Contact: | Security Team bot <security-team> |
| Severity: | Minor | ||
| Priority: | P4 - Low | CC: | federico, i.gnatenko.brain, smash_bz, william.brown |
| Version: | unspecified | ||
| Target Milestone: | --- | ||
| Hardware: | Other | ||
| OS: | Other | ||
| URL: | https://smash.suse.de/issue/281615/ | ||
| Whiteboard: | CVSSv3.1:SUSE:CVE-2021-28878:7.5:(AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H) | ||
| Found By: | Security Response Team | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
| Attachments: | QA reproducer | ||
|
Description
Robert Frohl
2021-04-12 09:37:08 UTC
tracking as affected: - SUSE:SLE-15:Update/rust - SUSE:SLE-15-SP1:Update/rust also still a problem in openSUSE:Factory Created attachment 848244 [details] QA reproducer > rustc CVE-2021-28878.rs && ./CVE-2021-28878 strange, changing output like 'z4' or no text at all. Fixed rust should return 'aaaa', see [0]. [0] https://play.rust-lang.org/?version=nightly&mode=release&edition=2018&gist=196385a61f316746f71e9a59aa68d6e7 openSUSE:Factory should be resolved now, as we have updated to 1.52.0. For SLE-15/SLE-15-SP1 I'm not sure of the best approach here. This issue in zip is a really really really niche case, that requires someone to have used zip in a really weird way. We can't update to 1.52 in SLE-15 right now because of the requirements of firefox to be on 1.43 (I think it is). in SP4 we are aiming to move to parallel rust versions https://en.opensuse.org/Parallel_Rust_Versions_Roadmap which will mean we can have > 1.52 for almost everything except firefox. I think that this is a really low risk issue. I think I'd want to see evidence of the incorrect usage of zip in the calling application, and then we can decide to backport to rust or if we just patch the affected libraries instead. i.e. it would be a lot of work to resolve this, for an issue that is extremely unlikely to be hit outside of pathological cases. Does not affect any supported version anymore. Updated tracking. Closing |