oss-sec mailing list archives
Re: Re: CVE request: Jenkins remote code execution vulnerability due to unsafe deserialization
From: Daniel Beck <ml () beckweb net>
Date: Wed, 18 Nov 2015 11:00:13 +0100
On 18.11.2015, at 01:54, cve-assign () mitre org wrote:
As far as we know, "the Groovy variant in 'ysoserial'" means: https://github.com/frohoff/ysoserial/blob/master/src/main/java/ysoserial/payloads/Groovy1.java
Exactly. My apologies for the vague description.
Also, we are guessing that Groovy is relevant because of: https://wiki.jenkins-ci.org/display/JENKINS/Groovy+plugin
Groovy Plugin and its version is unrelated, as Groovy is included in Jenkins core. Jenkins was vulnerable even without Groovy Plugin.
If it were necessary or recommended to change any component unique to Jenkins, then you can have an additional CVE ID for the ysoserial Groovy aspect of SECURITY-218. (Our expectation is that separate CVE IDs are needed because the Groovy plugin has own version numbering -- such as version 1.27 -- that's separate from the version numbering of Jenkins core.)
We updated neither commons-collections nor Groovy, the fix for both is specific to Jenkins, in the same component, and was part of the same release of Jenkins. Does this mean the one CVE ID covers both? -- Daniel Beck
Current thread:
- CVE request: Jenkins remote code execution vulnerability due to unsafe deserialization Daniel Beck (Nov 09)
- Re: CVE request: Jenkins remote code execution vulnerability due to unsafe deserialization cve-assign (Nov 17)
- Re: Re: CVE request: Jenkins remote code execution vulnerability due to unsafe deserialization Daniel Beck (Nov 18)
- Re: CVE request: Jenkins remote code execution vulnerability due to unsafe deserialization cve-assign (Nov 17)
