RSS Atom Add a new post titled:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In July, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 25.25h for LTS (out of 30 max; all done) and 13.25h for ELTS (out of 20 max; all done).

We shifted suites: welcome Stretch LTS and Jessie ELTS. The LTS->ELTS switch happened at the start of the month, but the oldstable->LTS switch happened later (after finalizing and flushing proposed-updates to a last point release), causing some confusion but nothing major.

ELTS - Jessie

  • New local build setup
  • ELTS buildds: request timezone harmonization
  • Reclassify in-progress updates from jessie-LTS to jessie-ELTS
  • python3.4: finish preparing update, security upload ELA 239-1
  • net-snmp: global triage: bisect CVE-2019-20892 to identify affected version, jessie/stretch not-affected
  • nginx: global triage: clarify CVE-2013-0337 status; locate CVE-2020-11724 original patch and regression tests, update MITRE
  • nginx: security upload ELA-247-1 with 2 CVEs

LTS - Stretch

  • Reclassify in-progress/needed updates from stretch/oldstable to stretch-LTS
  • rails: upstream security: follow-up on CVE-2020-8163 (RCE) on upstream bug tracker and create pull request for 4.x (merged), hence getting some upstream review
  • rails: global security: continue coordinating upload in multiple Debian versions, prepare fixes for common stretch/buster vulnerabilities in buster
  • rails: security upload DLA-2282 fixing 3 CVEs
  • python3.5: security upload DLA-2280-1 fixing 13 pending non-critical vulnerabilities, and its test suite
  • nginx: security upload DLA-2283 (cf. common ELTS work)
  • net-snmp: global triage (cf. common ELTS work)
  • public IRC monthly team meeting
  • reach out to clarify the intro from last month's report, following unsettled feedback during meeting

Documentation/Scripts

  • ELTS/README.how-to-release-an-update: fix typo
  • ELTS buildd: attempt to diagnose slow perfs, provide comparison with Debian and local builds
  • LTS/Meetings: improve presentation
  • SourceOnlyUpload: clarify/de-dup pbuilder doc
  • LTS/Development: reference build logs URL, reference proposed-updates issue during dists switch, reference new-upstream-versioning discussion, multiple jessie->stretch fixes and clean-ups
  • LTS/Development/Asan: drop wheezy documentation
  • Warn about jruby mis-triage
  • Provide feedback for ksh/CVE-2019-14868
  • Provide feedback for condor update
  • LTS/TestsSuites/nginx: test with new request smuggling test cases
Posted Mon Aug 3 13:52:10 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In June, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 30h for LTS (out of 30 max; all done) and 5.25h for ELTS (out of 20 max; all done).

While LTS is part of the Debian project, fellow contributors sometimes surprise me: suggestion to vote for sponsors-funded projects with concorcet was only met with overhead concerns, and there were requests for executive / business owner decisions (we're currently heading towards consultative vote); I heard concerns about discussing non-technical issues publicly (IRC team meetings are public though); the private mail infrastructure was moved from self-hosting straight to Google; when some got an issue with Debian Social for our first video conference, there were immediate suggestions to move to Zoom...
Well, we do need some people to make those LTS firmware updates in non-free :)

Also this was the last month before shifting suites: goodbye to Jessie LTS and Wheezy ELTS, welcome Stretch LTS and Jessie ELTS.

ELTS - Wheezy

  • mysql-connector-java: improve testsuite setup; prepare wheezy/jessie/stretch triple builds; coordinate versioning scheme with security-team; security upload ELA 234-1
  • ntp: wheezy+jessie triage: 1 ignored (too intrusive to backport); 1 postponed (hard to exploit, no patch)
  • Clean-up (ditch) wheezy VMs :)

LTS - Jessie

  • mysql-connector-java: see common work in ELTS
  • mysql-connector-java: security uploads DLA 2245-1 (LTS) and DSA 4703 (oldstable)
  • ntp: wheezy+jessie triage (see ELTS)
  • rails: global triage, backport 2 patches, security upload DLA 2251-1
  • rails: global security: prepare stretch/oldstable update
  • rails: new important CVE on unmaintained 4.x, fixes introduce several regressions, propose new fix to upstream, update stretch proposed update [and jessie, but rails will turn out unsupported in ELTS]
  • python3.4: prepare update to fix all pending non-criticial issues, 5/6 ready
  • private video^W^Wpublic IRC team meeting

Documentation/Scripts

Posted Wed Jul 1 14:19:14 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In May, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 17.25h for LTS (out of 30 max; all done) and 9.25h for ELTS (out of 20 max; all done).

A survey will be published very shortly to gather feedback from all parties involved in LTS (users, other Debian teams...) -- let us know what you think, so we start the forthcoming new (Stretch) LTS cycle in the best conditions :)

Discussion is progressing on funding & governance of larger LTS-related projects. Who should decide: contributors, Freexian, sponsors? Do we fund with a percentage or by capping resources allocated on security updates? I voiced concerns over funding these at the expense of smaller, more organic, more recurrent tasks that are less easy to specify but greatly contribute to the overall quality nevertheless.

ELTS - Wheezy

  • mysql-connector-java: upgrade to 5.1.49, refresh patches, document/run test suite, prepare upload, prepare upgrade path (+ see LTS)
  • CVE-2020-3810/apt: triage (affected), enquire about failing test, run testsuite, security upload ELA 228-1

LTS - Jessie

  • ansible: global triage: finish last month's triage, fix affected versions, provide reproducer
  • ansible: backport patches to early version, security upload DLA 2202-1
  • mysql-connector-java: propose 5.1.49 update to all dists (+ see ELTS)
  • CVE-2019-20637/varnish: global triage: ping upstream, get PoC, determine status for all Debian dists, jessie not-affected
  • public IRC team meeting

Documentation/Scripts

Posted Tue Jun 2 11:29:02 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In April, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 28.75h for LTS (out of 30 max; all done) and 7.75h for ELTS (out of 20 max; I did 2.75).

Escalation procedures were (internally) documented with a focus on discussing issues with team coordinator(s) first.

Debian LTS had its first team meeting through IRC and lots of workflow question were discussed. This should help discuss questions that are a bit hard to bring up, and ensure everybody participates. There were lots of topics and it was a bit rushed, but this is something we want to repeat monthly now, possibly with audio/video in a couple months.

Remarks from last month's report were discussed, strengthening the Front-Desk role.

10% of the global funding is now reserved for infrastructure work. What kind of work, and who (LTS or external) will do the work, will be discussed further.

A fellow DD suggested (in a private conversation) that LTS may be taking time from the Debian Security team, due to additional commits to review. Conversely, this is another opportunity to mention all the global, non-LTS-specific work that LTS provides, which I usually highlight in my reports, and maybe I should be even more ;)

ELTS - Wheezy

  • CVE-2020-11612/netty: triage: ignored (deceptively hard to backport, OOM mitigation only)
  • mysql-connector-java: triage: in-progress (subscription-only update from Oracle, attempt to find more detail, waiting for public version)
  • CVE-2020-11868/ntp: global triage: identify and reference missing patch, coordinate with uploader

LTS - Jessie

  • netty, mysql-connector-java, ntp: common triage (see above)
  • CVE-2019-20637/varnish: global triage: attempt to reproduce, attempt to get PoC/vulnerable versions from upstream, update BTS
  • ansible: jessie triage: reset ignore->no-dsa old vulnerabilites after discussing with initial triager
  • ansible: global triage: identify more affected version ranges, locate more patches
  • ansible: prepare jessie upload (work-in-progress)
  • tiff: suites harmonization: offer to work on a tiff/stretch update, follow-up on maintainer's questions, who eventually did the update
  • dsa-needed.txt: identify stale entries from inactive LTS contributor, check for status
  • team meeting: see minutes

Documentation/Scripts

Posted Sat May 2 11:25:57 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In March, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 30h for LTS (out of 30 max; all done) and 20h for ELTS (out of 20 max; I did 0).

Most contributors claimed vulnerabilities by performing early CVE monitoring/triaging on their own, making me question the relevance of the Front-Desk role. It could be due to a transient combination of higher hours volume and lower open vulnerabilities.

Working as a collective of hourly paid freelancers makes it more likely to work in silos, resulting in little interaction when raising workflow topics on the mailing list. Maybe we're reaching a point where regular team meetings will be benefical.

As previously mentioned, I structure my work keeping the global Debian security in mind. It can be stressful though, and I believe current communication practices may deter such initiatives.

ELTS - Wheezy

  • No work. ELTS has few sponsors right now and few vulnerabilities to fix, hence why I could not work on it this month. I gave back my hours at the end of the month.

LTS - Jessie

  • lua-cgi: global triage: CVE-2014-10399,CVE-2014-10400/lua-cgi not-affected, CVE-2014-2875/lua-cgi referenced in BTS
  • libpcap: global triage: request CVE-2018-16301 rejection as upstream failed to; got MITRE to reject (not "dispute") a CVE for the first time!
  • nfs-utils: suites harmonization: CVE-2019-3689: ping upstream again, locate upstream'd commit, reference it at BTS and MITRE; close MR which had been ignored and now redone following said referencing
  • slurm-llnl: re-add; create CVE-2019-12838 reproducer, test abhijith's pending upload; reference patches; witness regression in CVE-2019-19728, get denied access to upstream bug, triage as ignored (minor issue + regression); security upload DLA 2143-1
  • xerces-c: global triage progress: investigate ABI-(in)compatibility of hle's patch direction; initiate discussion at upstream and RedHat; mark postponed
  • nethack: jessie triage fix: mark end-of-life
  • tor: global triage fix: CVE-2020-10592,CVE-2020-10593: fix upstream BTS links, fix DSA reference
  • php7.3: embedded copies: removed from unstable (replaced with php7.4); checked whether libonig is still bundled (no, now properly unbundled at upstream level); jessie still not-affected
  • okular: CVE-2020-9359: reference PoC, security upload DLA 2159-1

Documentation/Scripts

  • data/dla-needed.txt: tidy/refresh pending packages status
  • LTS/Development: DLA regression numbering when a past DLA affects a different package
  • LTS/FAQ: document past LTS releases archive location following a user request; trickier than expected, 3 contributors required to find the answer ;)
  • Question aggressive package claims; little feedback
  • embedded-copies: libvncserver: reference various state of embedded copies in italc/ssvnc/tightvnc/veyon/vncsnapshot; builds on initial research from sunweaver
  • Attempt to progress on libvncserver embedded copies triaging; technical topic not anwered, organizational topic ignored
  • phppgadmin: provide feedback on CVE-2019-10784
  • Answer general workflow question about vulnerability severity
  • Answer GPAC CVE information request from a PhD student at CEA, following my large security update
Posted Wed Apr 1 14:26:39 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In February, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 20h for LTS (out of 30 max; all done) and 8h for ELTS (out of 20 max; I did 7).

Security work is never completely isolated, typically my work on nodejs impacted jessie/stretch/buster, and my work on netty affected wheezy/jessie/stretch :)

ELTS - Wheezy

  • netty: refine prior triages, write minimal test server, adapt 3 fixes, security upload: ELA-214
  • Suggest redispatching hours from past month not given back in time, as team members only got 3.5h each; follow-up on the issue
  • Contribute to exchanges about supporting libgd2 (unsupported dependency of a supported package, an inconsistency we'll try to detect earlier)

LTS - Jessie

  • netty: refine prior triages, security upload DLA 2109-1
  • netty-3.9: identify duplicate package, fix prior vulnerabilities, security upload DLA 2110-1
  • nodejs: jessie/stretch/buster triage (3 CVEs), request access to not-yet-public hackerone reports
  • nodejs: clarify support status, reclassify open vulnerabilities on nodejs ecosystem as EOL (end-of-life) for jessie & stretch
  • http-parser: mark as affected by nodejs' CVE-2019-15605; jessie triage: ignored (invasive change with ABI breakage)
  • wordpress: precise my past triage (2 CVEs): postponed (serialization vulnerabilities related to PHP itself currently not addressed at application/wordpress level)
  • otrs2: security upload DLA 2118-1 (interestingly recent otrs2 is in non-free not due to licensing, but due to embedding specific versions of javascript dependencies)
  • CVE-2019-10784/phppgadmin: answer request for comment
  • xen: point out external support

Documentation/Scripts

  • TestSuites/netty: instruction on how to find, compile and adapt server examples
  • DLA-1993-1: update Debian website (was only published via mailing-list)
  • embedded-code-copies: reference http-parser embedded in nodejs
  • README.external-support: clean-up external support contact points
Posted Mon Mar 2 10:30:55 2020 Tags:

Escoria, the point-and-click system for the Godot game engine, is now working again with the latest Godot (3.2).

Godot is a general-purpose game engine. It comes with an extensive graphic editor with skeleton and animation support, can create all sorts of games and mini-games, making it an interesting choice for point-and-click's.

The Escoria point-and-click template provides notably a dialog system and the Esc language to write the story and interactions. It was developed for the Dog Mendonça and Pizzaboy crowdfunded game and later released as free software. A community is developing the next version, but the current version has been incompatible with the current Godot engine. So I upgraded the game template as well as the Escoria in Daïza tutorial game to Godot 3.2. Enjoy!

HTML5 support is still lacking, so I might get a compulsive need to fix it in the future ;)

Posted Sat Feb 8 16:24:43 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In January, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 23.75h for LTS (out of 30 max) and 20h for ELTS (max) of which I did 1.5h.

I couldn't work much on ELTS because there are very few sponsors left for oldoldoldstable (sic!), hence not many packages to support, hence not much possible work.

In a direct communication, one team member expressed that team workflow is to be discussed on a private mailing list because according to them these problems don't need to be discussed in public and only results count. I have an opposite approach -- anything that isn't strictly confidential / security-sensitive is to be discussed publicly. The Debian Social Contract says "We don't hide problems" so if we want to address problems in a Debian workflow, this is to be public. What do you think?

ELTS - Wheezy

  • request supported packages list update
  • sqlite3: re-triage: drop as it just reached end-of-life
  • nss: re-triage: suggest clarification since package just reached end-of-life, yet claimed; actually a static build dependency of openjdk
  • python-apt: re-triage: claimed, checked actual EOL status with triager, unclaimed
  • python2.7: re-triage: was marked end-of-life, checked !EOL status with triager, marked for update

LTS - Jessie

  • wordpress: jessie triage (7 CVEs), security upload
  • tomcat7: start working then cancel work since it was unclaimed since 9 days yet 2 LTS members were already working on it
  • gpac: jessie triage (17 CVEs), reported new crash, reported invalid fix, security upload

Documentation/Scripts

Posted Tue Feb 4 13:32:57 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In December, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 16.5h for LTS (out of 30 max) and 16.5h for ELTS (max).

This is less than usual, AFAICS due to having more team members requesting more hours (while I'm above average), and less unused hours given back (or given back too late).

ELTS - Wheezy

  • libonig: finish work started in November:
  • CVE-2019-19203/libonig: can't reproduce, backport non-trivial likely to introduce bugs,
  • CVE-2019-19012,CVE-2019-19204,CVE-2019-19246/libonig: security upload
  • libpcap: attempt to recap vulnerabilities mismatch (possibly affecting ELA-173-1/DLA-1967-1); no follow-up from upstream
  • CVE-2019-19317,CVE-2019-19603,CVE-2019-19645/sqlite3: triage: not-affected (development version only)
  • CVE-2019-1551/openssl: triage: not-affected; discuss LTS triage rationale
  • CVE-2019-14861,CVE-2019-14870/samba: triage: not-affected
  • CVE-2019-19725/sysstat: triage: not-affected (vulnerable code introduced in v11.7.1)
  • CVE-2019-15845,CVE-2019-16201,CVE-2019-16254,CVE-2019-16255/ruby1.9.1: security upload

LTS - Jessie

  • CVE-2019-19012,CVE-2019-19204,CVE-2019-19246/libonig: shared work with ELTS, security upload
  • libpcap: shared work with ELTS
  • libav: finish work started in November:
  • CVE-2018-18829/libav: triage: postponed (libav-specific issue, no patch)
  • CVE-2018-11224/libav: triage: postponed (libav-specific issue, no patch)
  • CVE-2017-18247/libav: triage: ignored (not reproducible, no targeted patch)
  • CVE-2017-18246/libav: triage: ignored (not reproducible)
  • CVE-2017-18245/libav: reproduce, track down fix in ffmpeg
  • CVE-2017-18244/libav: triage: ignored (not reproducible)
  • CVE-2017-18243/libav: triage: ignored (not reproducible)
  • CVE-2017-18242/libav: triage: ignored (not reproducible)
  • CVE-2017-17127/libav: reproduce, track down fix in ffmpeg
  • CVE-2016-9824/libav: triage: ignored: usan (undefined sanitized) warning only, no patch
  • CVE-2016-9823/libav: triage: ignored: usan (undefined sanitized) warning only, no patch
  • CVE-2016-5115/libav: triage: postpone due different (indirect mplayer) vulnerability and lack of time
  • CVE-2017-17127,CVE-2017-18245,CVE-2018-19128,CVE-2018-19130,CVE-2019-14443,CVE-2019-17542/libav: security upload

Documentation/Scripts

Posted Thu Jan 2 10:18:46 2020 Tags:

Debian LTS Logo

Here is my transparent report for my work on the Debian Long Term Support (LTS) and Debian Extended Long Term Support (ELTS), which extend the security support for past Debian releases, as a paid contributor.

In November, the monthly sponsored hours were split evenly among contributors depending on their max availability - I was assigned 24.5h for LTS (out of 30 max) and 20h for ELTS (max).

Multiple vulnerabilities come from in-process fuzzing (library fuzzing with compiler instrumentation, as opposed to fuzzing a user executable). This is an interesting technique, though those are harder to reproduce, especially with older versions or (even worse) forks. A significant portion of such vulnerabilities comes from google's OSS-117Fuzz infrastructure.

data/CVE/list from the debian security-tracker repository reached 20M. With multiple changes per hour, git blame is consequently near-unusable: several minutes for a targetted, single-line look-up, if the entry is not too old. Despite this, the git commit messages are often used for triage justification or even as a substitute for personal communication, a practice I wouldn't recommend. #908678 looks stalled.

MITRE is still reactive when reporting issues on various free software project, and still very shy about changing the status of vulnerabilities. This is understandable when dealing with hard-to-reproduce issues, less understandable with legit-looking bogus vulnerabilities, which some people still like to throw at us so we have more work to do and get paid (seriously: please don't).

ELTS - Wheezy

  • Second part of my Front-Desk week, though only auto-triaged unsupported packages
  • CVE-2019-14866/cpio: help opal investigate reproducibility issue, contact cpio maintainer and security@gnu.org to get official patch/review
  • CVE-2019-18684/sudo: deconstruct bogus vulnerability; MITRE now marks it as DISPUTED
  • CVE-2019-5068/mesa: attempt to reproduce the issue, BTS update, testing, security upload
  • CVE-2019-3466/postgresql-common: triage: not-affected
  • libonig: start work on multiple vulnerabilities with non-trivial backports; to be completed in December
  • CVE-2019-19012/libonig: backport for 5.9, get maintainer review
  • CVE-2019-19246/libonig: register CVE for untracked vulnerability (discovered through upstream fuzzing, re-discovered through php-mbstring)
  • libonig: find embedded copy in php7.0 (Stretch) and php7.3 (Buster); LTS/ELTS not-affected

LTS - Jessie

  • CVE-2019-3689/nfs-util: ping upstream and debian sid, no pong
  • CVE-2019-14866/cpio: shared work with ELTS
  • CVE-2019-18684/sudo: shared work with ELTS
  • CVE-2019-5068/mesa: shared work with ELTS, security upload
  • CVE-2019-3466/postgresql-common: confirmed fix: jessie already fixed but I didn't notice due to late DLA
  • CVE-2019-11027/ruby-openid: provide requested second opinion
  • libav: start processing pending issues, package is a ffmpeg fork, was removed from newer dists and is unresponsive to security issues, requiring more work; to be completed in December
  • CVE-2019-17542/libav: heap-based buffer overflow: apply fix though libfuzzer-based reproducer not reproducible
  • CVE-2019-17539/libav: triage: not-affected (vulnerable code introduced later)
  • CVE-2019-14443/libav: reproduce, track down fix in ffmpeg, update libav bug
  • CVE-2019-14441/libav: mitre request: duplicate CVE-2018-19129 (got DISPUTED); fix attempt, update libav bug
  • CVE-2019-14371/libav: triage: already fixed through CVE-2018-11102
  • CVE-2019-9720/libav: triage: unimportant (stretching the definition of DoS)
  • CVE-2019-9719/libav: mitre request: rejection (got DISPUTED): generic warning, no vulnerability
  • CVE-2019-9717/libav: triage: unimportant (stretching the definition of DoS)
  • CVE-2018-20001/libav: jessie triage: postponed (not reproducible)
  • CVE-2018-19130/libav: mitre request: duplicate CVE-2017-17127 (got DISPUTED)
  • CVE-2018-19128/libav: reproduce, track down fix in ffmpeg
  • Welcome new trainee

Documentation/Scripts

Posted Sat Nov 30 19:59:55 2019 Tags:

This blog is powered by ikiwiki.