[MIR] libgssglue
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| gsasl (Ubuntu) |
Fix Released
|
Undecided
|
Lukas Märdian | ||
| libgssglue (Ubuntu) |
Fix Released
|
Undecided
|
Unassigned | ||
Bug Description
Hi
I am upstream author/maintainer of libgssglue and gsasl and I am also
Debian package maintainer of libgssglue, gsasl, and some related
packages like gssproxy and gss/shishi. This is a request that
'libgssglue' is included in Ubuntu main too.
Upstream project page:
https:/
Debian packaging page:
https:/
Around 18 months ago 'gsasl' was added to main, but its
build-dependency on 'libgssglue' was dropped in a manual patch for
Ubuntu because dependency was new in 'gsasl', had not been used in a
Debian stable release, and the rationale wasn't articulated. See
https:/
See MIR template below.
/Simon
[Availability]
TODO: The package TBDSRC is already in Ubuntu universe.
OK: The package libgssglue is already in Ubuntu universe.
TODO: The package TBDSRC build for the architectures it is designed to work on.
OK: The package libgssglue build for the architectures it is designed to work on.
TODO: It currently builds and works for architectures: TBD
OK: It currently builds and works for architectures:
[FULLYBUILT] amd64
[FULLYBUILT] arm64
[FULLYBUILT] armhf
[FULLYBUILT] ppc64el
[FULLYBUILT] riscv64
[FULLYBUILT] s390x
TODO: Link to package https:/
OK: Link to package https:/
The package libgssglue is also built inside Debian, and included in a
stable release.
[Rationale]
RULE: There must be a certain level of demand for the package
TODO: - The package TBDSRC is required in Ubuntu main for TBD
TODO-A: - The package TBDSRC will generally be useful for a large part of
TODO-A: our user base
TODO-B: - The package TBDSRC will not generally be useful for a large part of
TODO-B: our user base, but is important/helpful still because TBD
TODO: - Additional reasons TBD
TODO: - Additionally new use-cases enabled by this are TBD
TODO: - Package TBDSRC covers the same use case as TBD, but is better
TODO: because TBD, thereby we want to replace it.
TODO: - The package TBDSRC is a new runtime dependency of package TBD that
TODO: we already support
RULE: Sometimes there are other/better ways, often are achieved by using a
RULE: library with similar functionality that is more commonly used and
RULE: thereby already in main or a better candidate to promote.
RULE: Reducing the set of supported software in Ubuntu helps to focus on the
RULE: right things, otherwise Ubuntu developers will be consumed by updating
RULE: many variations of the same - wasting valuable time that could be better
RULE: spent elsewhere.
RULE: If there are other packages in the archive that are close, but unable to
RULE: address the problem you might spend some time explaining what exists and
RULE: why it isn't a sufficient alternative.
TODO: - There is no other/better way to solve this that is already in main or
TODO: should go universe->main instead of this.
The package libgssglue is useful in Ubuntu main for gsasl which is in
main. Including it would avoid having to patch 'gsasl' and
'libgssglue' packages.
I can't honestly claim that libgssglue is directly useful for a large
part of the user base, because it is a low-level library. Most users
would never interact with it. One advantage for non-Kerberos users is
that availability of libgssglue would allow more packages in 'main' to
migrate away from direct dependencies on libkrb5, which may be
something that is useful for a large part of the user base.
Another purpose is for those users who care about Kerberos (arguably a
small part of the entire user base) is to dynamically be able to
switch GSS-API mechanism during runtime.
Libgssglue may be a way for some of OpenSSH, Curl, Fetchmail, Cyrus
SASL, Bind9, etc to NOT have to directly link with libkrb5. We
already demonstrated that this approach is practical for 'gsasl' in
Debian 12 stable. It is a matter of porting more packages to link
with libgssglue to spread this ability.
Some arguments covered in my blog here:
https:/
Re-iterating:
1) Reduce reverse dependencies on MIT Kerberos for packages where
Kerberos is only rarely used by a small portion of all users.
2) Allow sysadmin to change the Kerberos implementation to use on the
system, switching between MIT Kerberos and Heimdal (or any future
compatible GSS-API implementation). GNU SASL does GitLab CI/CD interop
checking between GNU SASL and Dovecot switching between different
implementations to make sure this works in practice not just theory.
3) Reduce packaging complexity caused by having packages built for both
MIT and Heimdal, which is inflexible. Compare why libsasl2-modules-
gssapi-heimdal and libsasl2-
4) Pave the wave for a 'gss-agent' style interface, similar to how ssh-
agent works, to avoid storing GSS-API credentials in the same process
memory as a potentially vulnerable network application. For OpenSSH
this would be nice.
Expanding on 1): For Debian 12 only 'gsasl' is using 'libgssglue', but
you can see below that 'gsasl' is operational without dragging in
libkrb5. But on Ubuntu 23.10 libkrb5 is dragged in too. See outputs
below. There is potential to achieve the same for all other packages
that uses GSS-API: OpenSSH, Curl, Fetchmail, Cyrus SASL, Bind9, GNOME.
etc. Some of them may be (probably incorrectly) using libkrb5 API's
directly instead of pure GSS-API interfaces, and can never be rescued
without more patching. But several of them should be possible to
improve.
jas@kaka:~$ podman run -it --rm ubuntu:23.10
root@aceae90e92
Reading package lists...
root@aceae90e92
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
The following additional packages will be installed:
gsasl-common krb5-locales libgsasl18 libgssapi-krb5-2 libidn12
libk5crypto3 libkeyutils1 libkrb5-3 libkrb5support0 libntlm0 libssl3
Suggested packages:
krb5-doc krb5-user
The following NEW packages will be installed:
gsasl gsasl-common krb5-locales libgsasl18 libgssapi-krb5-2 libidn12
libk5crypto3 libkeyutils1 libkrb5-3 libkrb5support0 libntlm0 libssl3
0 upgraded, 12 newly installed, 0 to remove and 5 not upgraded.
Need to get 2789 kB of archives.
After this operation, 9348 kB of additional disk space will be used.
Do you want to continue? [Y/n] ^C
root@aceae90e92
exit
jas@kaka:~$ podman run -it --rm debian:12-slim
root@897456681c
Reading package lists...
root@897456681c
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
The following additional packages will be installed:
gsasl-common libgsasl18 libgssglue1 libidn12 libntlm0
The following NEW packages will be installed:
gsasl gsasl-common libgsasl18 libgssglue1 libidn12 libntlm0
0 upgraded, 6 newly installed, 0 to remove and 0 not upgraded.
Need to get 747 kB of archives.
After this operation, 1512 kB of additional disk space will be used.
Do you want to continue? [Y/n] ^C
root@897456681c
RULE: Reviews will take some time. Also the potential extra work out of review
RULE: feedback from either MIR-team and/or security-team will take time.
RULE: For better prioritization it is quite helpful to clearly state the
RULE: target release and set a milestone to the bug task.
RULE: When doing so do not describe what you "wish" or "would like to have".
RULE: Only milestones that are sufficiently well-founded and related to
RULE: major releases will be considered
TODO-A: - The package TBDSRC is required in Ubuntu main no later than TBD
TODO-A: due to TBD
TODO-B: - It would be great and useful to community/processes to have the
TODO-B: package TBD in Ubuntu main, but there is no definitive deadline.
It would be nice to get libgssglue into 'main' before Ubuntu 24.04.
[Security]
RULE: The security history and the current state of security issues in the
RULE: package must allow us to support the package for at least 9 months (120
RULE: for LTS+ESM support) without exposing its users to an inappropriate level
RULE: of security risks. This requires checking of several things:
RULE: - Search in the National Vulnerability Database using the PKG as keyword
RULE: https:/
RULE: - check OSS security mailing list (feed into search engine
RULE: 'site:www.
RULE: - Ubuntu CVE Tracker
RULE: https:/
RULE: - Debian Security Tracker
RULE: https:/
TODO-A: - Had #TBD security issues in the past
TODO-A: - TBD links to such security issues in trackers
TODO-A: - TBD to any context that shows how these issues got handled in
TODO-A: the past
TODO-B: - No CVEs/security issues in this software in the past
Libgssglue had one CVE in 2011, see CVE-2011-2709:
https:/
The library is fairly minimal and stable: the vulnerable aspects are
probably the ones related to dlopen()'ing of the mechanism and the
configuration file handling (including the environment variable).
RULE: - Check for security relevant binaries, services and behavior.
RULE: If any are present, this requires a more in-depth security review.
RULE: Demonstrating that common isolation/
RULE: will help to raise confidence. For example a service running as root
RULE: open to the network will need to be considered very carefully. The same
RULE: service dropping the root permissions after initial initialization,
RULE: using various systemd isolation features and having a default active
RULE: apparmor profile is much less concerning and can speed up acceptance.
RULE: This helps Ubuntu, but you are encouraged to consider working with
RULE: Debian and upstream to get those security features used at wide scale.
TODO: - no `suid` or `sgid` binaries
TODO-A: - no executables in `/sbin` and `/usr/sbin`
TODO-B: - Binary TBD in sbin is no problem because TBD
TODO-A: - Package does not install services, timers or recurring jobs
TODO-B: - Package does install services, timers or recurring jobs
TODO-B: TBD (list services, timers, jobs)
TODO: - Security has been kept in mind and common isolation/
TODO: patterns are in place utilizing the following features:
TODO: TBD (add details and links/examples about things like dropping
TODO: permissions, using temporary environments, restricted users/groups,
TODO: seccomp, systemd isolation features, apparmor, ...)
TODO-A: - Packages does not open privileged ports (ports < 1024).
TODO-B: - Packages open privileged ports (ports < 1024), but they have
TODO-B: a reason to do so (TBD)
TODO-A: - Package does not expose any external endpoints
TODO-B: - Package does not expose an external endpoint, it is
TODO-B: TBD endpoint + TBD purpose
TODO: - Packages does not contain extensions to security-sensitive software
TODO: (filters, scanners, plugins, UI skins, ...)
The package only ship a shared library and a global configuration
file. The *-dev package ships C header files, static library and
pkg-config helper.
No executable binaries.
No installed services, timers or recurring jobs.
No opening of ports or external endpoint.
[Quality assurance - function/usage]
RULE: - After installing the package it must be possible to make it working with
RULE: a reasonable effort of configuration and documentation reading.
TODO-A: - The package works well right after install
TODO-B: - The package needs post install configuration or reading of
TODO-B: documentation, there isn't a safe default because TBD
The package by default dlopen libkrb5 so it works as a drop-in, as
shown in 'gsasl' that behaves the same under libgssglue and direct
linking with libkrb5. No configuration necessary.
[Quality assurance - maintenance]
RULE: - To support a package, we must be reasonably convinced that upstream
RULE: supports and cares for the package.
RULE: - The status of important bugs in Debian, Ubuntu and upstream's bug
RULE: tracking systems must be evaluated. Important bugs must be pointed out
RULE: and discussed in the MIR report.
TODO: - The package is maintained well in Debian/
TODO: not have too many, long-term & critical, open bugs
TODO: - Ubuntu https:/
TODO: - Debian https:/
TODO: - Upstream's bug tracker, e.g., GitHub Issues
TODO: - The package has important open bugs, listing them: TBD
TODO-A: - The package does not deal with exotic hardware we cannot support
TODO-B: - The package does deal with exotic hardware, it is present at TBD
TODO-B: to be able to test, fix and verify bugs
I'm upstream maintainer and committed to maintain it in the long-run,
I've already maintained gsasl for 20+ years.
There are some open bugs, but nothing critical preventing it from
working as intended that I'm aware of.
[Quality assurance - testing]
RULE: - The package must include a non-trivial test suite
RULE: - it should run at package build and fail the build if broken
TODO-A: - The package runs a test suite on build time, if it fails
TODO-A: it makes the build fail, link to build log TBD
TODO-B: - The package does not run a test at build time because TBD
The upstream project has a somewhat limited tests/ self-test suite but
the 'gsasl' project has extensive self-check testing of libgssglue on
Ubuntu with MIT Kerberos and Heimdal against Dovecot and GNU Mailutils
servers, see amount of testing here:
https:/
https:/
https:/
RULE: - The package should, but is not required to, also contain
RULE: non-trivial autopkgtest(s).
TODO-A: - The package runs an autopkgtest, and is currently passing on
TODO-A: this TBD list of architectures, link to test logs TBD
TODO-B: - The package does not run an autopkgtest because TBD
RULE: - existing but failing tests that shall be handled as "ok to fail"
RULE: need to be explained along the test logs below
TODO-A: - The package does have not failing autopkgtests right now
TODO-B: - The package does have failing autopkgtests tests right now, but since
TODO-B: they always failed they are handled as "ignored failure", this is
TODO-B: ok because TBD
The Debian libgssglue package has extensive autopkgtest, see:
https:/
It sets up a Kerberos KDC and does interop testing between GNU SASL
(client) and Dovecot server, using MIT Kerberos and Heimdal libraries
in the client, both via libgssglue.
There is a minor spurious failure in the self-tests sometimes, but it
is a reported MIT Kerberos problem:
https:/
RULE: - If no build tests nor autopkgtests are included, and/or if the package
RULE: requires specific hardware to perform testing, the subscribed team
RULE: must provide a written test plan in a comment to the MIR bug, and
RULE: commit to running that test either at each upload of the package or
RULE: at least once each release cycle. In the comment to the MIR bug,
RULE: please link to the codebase of these tests (scripts or doc of manual
RULE: steps) and attach a full log of these test runs. This is meant to
RULE: assess their validity (e.g. not just superficial).
RULE: If possible such things should stay in universe. Sometimes that is
RULE: impossible due to the way how features/
RULE: but if you are going to ask for promotion of something untestable
RULE: please outline why it couldn't provide its value (e.g. by splitting
RULE: binaries) to users from universe.
RULE: This is a balance that is hard to strike well, the request is that all
RULE: options have been exploited before giving up. Look for more details
RULE: and backgrounds https:/
RULE: Just like in the SRU process it is worth to understand what the
RULE: consequences a regression (due to a test miss) would be. Therefore
RULE: if being untestable we ask to outline what consequences this would
RULE: have for the given package. And let us be honest, even if you can
RULE: test you are never sure you will be able to catch all potential
RULE: regressions. So this is mostly to force self-awareness of the owning
RULE: team than to make a decision on.
TODO: - The package can not be well tested at build or autopkgtest time
TODO: because TBD. To make up for that:
TODO-A: - We have access to such hardware in the team
TODO-B: - We have allocated budget to get this hardware, but it is not here
TODO-B: yet
TODO-C: - We have checked with solutions-qa and will use their hardware
TODO-C: through testflinger
TODO-D: - We have checked with other team TBD and will use their hardware
TODO-D: through TBD (eg. MAAS)
TODO-E: - We have checked and found a simulator which covers this case
TODO-E: sufficiently for testing, our plan to use it is TBD
TODO-F: - We have engaged with the upstream community and due to that
TODO-F: can tests new package builds via TBD
TODO-G: - We have engaged with our user community and due to that
TODO-G: can tests new package builds via TBD
TODO-H: - We have engaged with the hardware manufacturer and made an
TODO-H: agreement to test new builds via TBD
TODO-A-H: - Based on that access outlined above, here are the details of the
TODO-A-H: test plan/automation TBD (e.g. script or repo) and (if already
TODO-A-H: possible) example output of a test run: TBD (logs).
TODO-A-H: We will execute that test plan
TODO-A-H1: on-uploads
TODO-A-H2: regularly (TBD details like frequency: monthly, infra: jira-url)
TODO-X: - We have exhausted all options, there really is no feasible way
TODO-X: to test or recreate this. We are aware of the extra implications
TODO-X: and duties this has for our team (= help SEG and security on
TODO-X: servicing this package, but also more effort on any of your own
TODO-X: bug triage and fixes).
TODO-X: Due to TBD there also is no way to provide this to users from
TODO-X: universe.
TODO-X: Due to the nature, integration and use cases of the package the
TODO-X: consequences of a regression that might slip through most likely
TODO-X: would include
TODO-X: - TBD
TODO-X: - TBD
TODO-X: - TBD
RULE: - In some cases a solution that is about to be promoted consists of
RULE: several very small libraries and one actual application uniting them
RULE: to achieve something useful. This is rather common in the go/rust space.
RULE: In that case often these micro-libs on their own can and should only
RULE: provide low level unit-tests. But more complex autopkgtests make no
RULE: sense on that level. Therefore in those cases one might want to test on
RULE: the solution level.
RULE: - Process wise MIR-requesting teams can ask (on the bug) for this
RULE: special case to apply for a given case, which reduces the test
RULE: constraints on the micro libraries but in return increases the
RULE: requirements for the test of the actual app/solution.
RULE: - Since this might promote micro-lib packages to main with less than
RULE: the common level of QA any further MIRed program using them will have
RULE: to provide the same amount of increased testing.
TODO: - This package is minimal and will be tested in a more wide reaching
TODO: solution context TBD, details about this testing are here TBD
[Quality assurance - packaging]
RULE: - The package uses a debian/watch file whenever possible. In cases where
RULE: this is not possible (e.g. native packages), the package should either
RULE: provide a debian/
RULE: comments only) providing clear instructions on how to generate the
RULE: source tar file.
TODO-A: - debian/watch is present and works
TODO-B: - debian/watch is not present, instead it has TBD
TODO-C: - debian/watch is not present because it is a native package
The package has a debian/watch and passes many other packaging QA
checks, see pipeline here:
https:/
RULE: - The package should define the correct "Maintainer:" field in
RULE: debian/control. This needs to be updated, using `update-maintainer`
RULE: whenever any Ubuntu delta is applied to the package, as suggested by
RULE: dpkg (LP: #1951988)
TODO: - debian/control defines a correct Maintainer field
OK
RULE: - It is often useful to run `lintian --pedantic` on the package to spot
RULE: the most common packaging issues in advance
RULE: - Non-obvious or non-properly commented lintian overrides should be
RULE: explained
TODO: - This package does not yield massive lintian Warnings, Errors
TODO: - Please link to a recent build log of the package <TBD>
TODO: - Please attach the full output you have got from
TODO: `lintian --pedantic` as an extra post to this bug.
TODO-A: - Lintian overrides are not present
TODO-B: - Lintian overrides are present, but ok because TBD
See recent build log here:
https:/
Lintian --pedantic doesn't output anything:
https:/
RULE: - The package should not rely on obsolete or about to be demoted packages.
RULE: That currently includes package dependencies on Python2 (without
RULE: providing Python3 packages), and packages depending on GTK2.
TODO: - This package does not rely on obsolete or about to be demoted packages.
TODO: - This package has no python2 or GTK2 dependencies
OK
RULE: - Debconf questions should not bother the default user too much
TODO-A: - The package will be installed by default, but does not ask debconf
TODO-A: questions higher than medium
TODO-B: - The package will not be installed by default
No debconf usage.
RULE: - The source packaging (in debian/) should be reasonably easy to
RULE: understand and maintain.
TODO-A: - Packaging and build is easy, link to debian/rules TBD
TODO-B: - Packaging is complex, but that is ok because TBD
I'd say debian/ is simple and easy to maintain.
[UI standards]
TODO-A: - Application is not end-user facing (does not need translation)
TODO-B: - Application is end-user facing, Translation is present, via standard
TODO-B: intltool/gettext or similar build and runtime internationaliz
TODO-B: system see TBD
TODO-A: - End-user applications that ships a standard conformant desktop file,
TODO-A: see TBD
TODO-B: - End-user applications without desktop file, not needed because TBD
It is a low-level library with no direct user-facing interactions.
There are some internal error messages that could leak through on the
terminal in some situations, but they are not translated.
[Dependencies]
RULE: - In case of alternative the preferred alternative must be in main.
RULE: - Build(-only) dependencies can be in universe
RULE: - If there are further dependencies they need a separate MIR discussion
RULE: (this can be a separate bug or another task on the main MIR bug)
TODO-A: - No further depends or recommends dependencies that are not yet in main
TODO-B: - There are further dependencies that are not yet in main, MIR for them
TODO-B: is at TBD
TODO-C: - There are further dependencies that are not yet in main, the MIR
TODO-C: process for them is handled as part of this bug here.
No dependencies except for 'libc6'.
[Standards compliance]
RULE: - Major violations should be documented and justified.
RULE: - FHS: https:/
RULE: - Debian Policy: https:/
TODO-A: - This package correctly follows FHS and Debian Policy
TODO-B: - This package violates FHS or Debian Policy, reasons for that are TBD
OK, follows policies.
[Maintenance/Owner]
RULE: The package must have an acceptable level of maintenance corresponding
RULE: to its complexity:
RULE: - All packages must have a designated "owning" team, regardless of
RULE: complexity.
RULE: This requirement of an owning-team comes in two aspects:
RULE: - A case needs to have a team essentially saying "yes we will own that"
RULE: to enter the MIR process. Usually that is implied by team members
RULE: filing MIR requests having the backup by their management for the
RULE: long term commitment this implies.
RULE: - A community driven MIR request might be filed to show the use case,
RULE: but then, as a first step, needs to get a team agreeing to own
RULE: it before the case can be processed further.
RULE: If unsure which teams to consider have a look at the current mapping
RULE: http://
RULE: - The package needs a bug subscriber before it can be promoted to main.
RULE: Strictly speaking that subscription can therefore wait until the
RULE: moment of the actual promotion by an archive admin. But it is
RULE: strongly recommended to subscribe early, as the owning team will get
RULE a preview of the to-be-expected incoming bugs later on.
RULE: - Simple packages (e.g. language bindings, simple Perl modules, small
RULE: command-line programs, etc.) might not need very much maintenance
RULE: effort, and if they are maintained well in Debian we can just keep them
RULE: synced. They still need a subscribing team to handle bugs, FTBFS and
RULE: tests
RULE: - More complex packages will usually need a developer or team of
RULE: developers paying attention to their bugs, whether that be in Ubuntu
RULE: or elsewhere (often Debian). Packages that deliver major new headline
RULE: features in Ubuntu need to have commitment from Ubuntu developers
RULE: willing to spend substantial time on them.
TODO-A: - The owning team will be TBD and I have their acknowledgement for
TODO-A: that commitment
TODO-B: - I Suggest the owning team to be TBD
TODO-A: - The future owning team is already subscribed to the package
TODO-B: - The future owning team is not yet subscribed, but will subscribe to
TODO-B: the package before promotion
I'm not fully literate about all the Ubuntu MIR process, so I'd
appreciate guidance here, but if relevant, I can offer to be part of
the owning team but also happy to give it up to someone else who is
interested.
RULE: - Responsibilities implied by static builds promoted to main, which is
RULE: not a recommended but a common case with golang and rust packages.
RULE: - the security team will track CVEs for all vendored/embedded sources in main
RULE: - the security team will provide updates to main for all `golang-*-dev`
RULE: packages
RULE: - the security team will provide updates to main for non-vendored
RULE: dependencies as per normal procedures (including e.g.,
RULE: sponsoring/
RULE: - the security team will perform no-change-rebuilds for all packages
RULE: listing an CVE-fixed package as Built-Using and coordinate testing
RULE: with the owning teams responsible for the rebuilt packages
RULE: - for packages that build using any `golang-*-dev` packages:
RULE: - the owning team must state their commitment to test
RULE: no-change-rebuilds triggered by a dependent library/compiler and to
RULE: fix any issues found for the lifetime of the release (including ESM
RULE: when included)
RULE: - the owning team must provide timely testing of no-change-rebuilds
RULE: from the security team, fixing the rebuilt package as necessary
RULE: - for packages that build with approved vendored code:
RULE: - the owning team must state their commitment to provide updates to
RULE: the security team for any affected vendored code for the lifetime of
RULE: the release (including ESM when included)
RULE: - the security team will alert the owning team of issues that may
RULE: affect their vendored code
RULE: - the owning team will provide timely, high quality updates for the
RULE: security team to sponsor to fix issues in the affected vendored code
RULE: - if subsequent uploads add new vendored components or dependencies
RULE: these have to be reviewed and agreed by the security team.
RULE: - Such updates in the project might be trivial, but imply that a
RULE: dependency for e.g. a CVE fix will be moved to a new major version.
RULE: Being vendored that does gladly at least not imply incompatibility
RULE: issues with other packages or the SRU policy. But it might happen
RULE: that this triggers either:
RULE: a) The need to adapt the current version of the main package and/or
RULE: other vendored dependencies to work with the new dependency
RULE: b) The need to backport the fix in the dependency as the main
RULE: package will functionally only work well with the older version
RULE: c) The need to backport the fix in the dependency, as it would imply
RULE: requiring a newer toolchain to be buildable that isn't available
RULE: in the target release.
RULE: - The rust ecosystem currently isn't yet considered stable enough for
RULE: classic lib dependencies and transitions in main; therefore the
RULE: expectation for those packages is to vendor (and own/test) all
RULE: dependencies (except those provided by the rust runtime itself).
RULE: This implies that all the rules for vendored builds always
RULE: apply to them. In addition:
RULE: - The rules and checks for rust based packages are preliminary and might
RULE: change over time as the ecosystem matures and while
RULE: processing the first few rust based packages.
RULE: - It is expected rust builds will use dh-cargo so that a later switch
RULE: to non vendored dependencies isn't too complex (e.g. it is likely
RULE: that over time more common libs shall become stable and then archive
RULE: packages will be used to build).
RULE: - Right now that tooling to get a Cargo.lock that will include internal
RULE: vendored dependencies isn't in place yet (expect a dh-cargo change
RULE: later). Until it is available, as a fallback one can scan the
RULE: directory at build time and let it be generated in debian/rules.
RULE: An example might look like:
RULE: debian/rules:
RULE: override_
RULE: CARGO_HOME=debian /usr/share/
RULE: debian/
RULE: Cargo.lock /usr/share/
RULE: debian/config.toml
RULE: # Use the vendorized sources to produce the Cargo.lock file. This
RULE: # can be performed by pointing $CARGO_HOME to the path containing
RULE: # this file.
RULE: [source]
RULE: [source.
RULE: directory = "vendor"
RULE: [source.crates-io]
RULE: replace-with = "my-vendor-source"
OK
RULE: - All vendored dependencies (no matter what language) shall have a
RULE: way to be refreshed
TODO-A: - This does not use static builds
TODO-B: - The team TBD is aware of the implications by a static build and
TODO-B: commits to test no-change-rebuilds and to fix any issues found for the
TODO-B: lifetime of the release (including ESM)
TODO-A: - This does not use vendored code
TODO-B: - The team TBD is aware of the implications of vendored code and (as
TODO-B: alerted by the security team) commits to provide updates and backports
TODO-B: to the security team for any affected vendored code for the lifetime
TODO-B: of the release (including ESM).
TODO-A: - This does not use vendored code
TODO-B: - This package uses vendored go code tracked in go.sum as shipped in the
TODO-B: package, refreshing that code is outlined in debian/
TODO-C: - This package uses vendored rust code tracked in Cargo.lock as shipped,
TODO-C: in the package (at /usr/share/
TODO-C: compressed), refreshing that code is outlined in debian/
TODO-D: - This package uses vendored code, refreshing that code is outlined
TODO-D: in debian/
TODO-A: - This package is not rust based
TODO-B: - This package is rust based and vendors all non language-runtime
TODO-B: dependencies
Libgssglue is a C library with no dependencies, so I don't think this
is relevant.
RULE: - if there has been an archive test rebuild that has occurred more recently
RULE: than the last upload, the package must have rebuilt successfully
TODO-A: - The package has been built in the archive more recently than the last
TODO-A: test rebuild
TODO-B: - The package successfully built during the most recent test rebuild
TODO-C: - The package was test rebuilt in PPA or sbuild recently (provide link/logs)
Ubuntu ships libgssglue in universe, so it should build correctly.
[Background information]
RULE: - The package descriptions should explain the general purpose and context
RULE: of the package. Additional explanations/
RULE: the MIR report.
RULE: - If the package was renamed recently, or has a different upstream name,
RULE: this needs to be explained in the MIR report.
TODO: The Package description explains the package well
TODO: Upstream Name is TBD
TODO: Link to upstream project TBD
TODO: TBD (any further background that might be helpful
See at the top of this report!
CVE References
| Lukas Märdian (slyon) wrote | #1 |
| summary: |
- [MIR] libssglue + [MIR] libgssglue |
| Adrien Nader (adrien) wrote | #2 |
| Lukas Märdian (slyon) wrote | #3 |
| Simon Josefsson (simon-josefsson) wrote | #4 |
| Changed in libgssglue (Ubuntu): | |
| assignee: | nobody → Christian Ehrhardt (paelzer) |
| Simon Josefsson (simon-josefsson) wrote | #5 |
| Christian Ehrhardt (paelzer) wrote | #6 |
| Changed in libgssglue (Ubuntu): | |
| assignee: | Christian Ehrhardt (paelzer) → Ubuntu Security Team (ubuntu-security) |
| Simon Josefsson (simon-josefsson) wrote | #7 |
| tags: | added: sec-3198 |
| Mark Esler (eslerm) wrote | #8 |
| Christian Ehrhardt (paelzer) wrote | #9 |
| Simon Josefsson (simon-josefsson) wrote | #10 |
| Christian Ehrhardt (paelzer) wrote | #11 |
| Changed in gsasl (Ubuntu): | |
| status: | New → Won't Fix |
| Lukas Märdian (slyon) wrote | #12 |
| Changed in gsasl (Ubuntu): | |
| status: | Won't Fix → Invalid |
| João Gomes (gomesjoao) wrote | #13 |
| Changed in libgssglue (Ubuntu): | |
| status: | New → In Progress |
| assignee: | Ubuntu Security Team (ubuntu-security) → nobody |
| tags: | added: update-excuse |
| Lukas Märdian (slyon) wrote | #14 |
| Changed in libgssglue (Ubuntu): | |
| status: | In Progress → Fix Committed |
| Changed in gsasl (Ubuntu): | |
| status: | Invalid → New |
| assignee: | nobody → Lukas Märdian (slyon) |
| Gianfranco Costamagna (costamagnagianfranco) wrote | #15 |
| Simon Josefsson (simon-josefsson) wrote | #16 |
| Lukas Märdian (slyon) wrote | #17 |
| Steve Langasek (vorlon) wrote | #18 |
| Changed in libgssglue (Ubuntu): | |
| status: | Fix Committed → Fix Released |
| Lukas Märdian (slyon) wrote | #19 |
| Changed in gsasl (Ubuntu): | |
| status: | New → Fix Released |
Foundations is a potential candidate for owning this package, I'll check with the team!