CVE-2013-4393

aka Journald Logging Clogging

CVE: CVE-2013-4393
CWE: 399
ipc:
  note: "This subsystem is responsible for parsing and creating logging data \nwhile
    systemd is running. This involves reading/writing information to \nrelevant files.\n"
  answer: true
  question: |
    Did the feature that this vulnerability affected use inter-process
    communication? IPC includes OS signals, pipes, stdin/stdout, message
    passing, and clipboard. Writing to files that another program in this
    software system reads is another form of IPC.

    Answer must be true or false.
    Write a note about how you came to the conclusions you did, regardless of
    what your answer was.
bugs:
- https://bugzilla.redhat.com/show_bug.cgi?id=859104
i18n:
  note: "This vulnerability happened entirely on the local level- no \nnetworking
    was involved. It can be safely assumed that this issue is independent \nof the
    Linux localization.\n"
  answer: false
  question: |
    Was the feature impacted by this vulnerability about internationalization
    (i18n)?

    An internationalization feature is one that enables people from all
    over the world to use the system. This includes translations, locales,
    typography, unicode, or various other features.

    Answer should be true or false
    Write a note about how you came to the conclusions you did, regardless of
    what your answer was.
vccs:
- note: "Discovered automatically by archeogit. Most likely not relevant- \nrefactors
    native journal functionality out of original journald file and \ninto its own.\n"
  commit: '0153028ae379eb7c9a463c548ef73ea392c6cdb0'
- note: "Discovered automatically by archeogit. This commit improves \nfunctionality
    for searching journal entries for matching field-value pairs.\n"
  commit: cbdca8525b4f36297cb9e5cb090a9648763ed1bf
- note: "Discovered automatically by archeogit. This commit sets up \nfunctionality
    for reading user log files into the system.\n"
  commit: b070e7f3c9ed680c821bd89d42506695f2438506
- note: "Discovered automatically by archeogit. This commit adds error handling \nfor
    problems caused by user-given log files.\n"
  commit: 18c7ed186be28800a2eeb37ad31c9c44480d3d9c
fixes:
- note: "Based on patch mentioned in https://cgit.freedesktop.org/systemd/systemd/commit/?id=1dfa7e79a60de680086b1d93fcc3629b463f58bd\nThis
    fix makes it so predefined log data can only come from trusted logging \ndirectories
    in the Linux OS (such as /tmp/). This also means only people with \nthe proper
    permissions to access files in those directories (who would also\nbe familiar
    with using them for the proper purposes) would be able to make use \nof the native
    logging protocol.\n"
  commit: 1dfa7e79a60de680086b1d93fcc3629b463f58bd
vouch:
  note: "The commit history of the file in between the vulnerability being \ndiscovered
    and fixed only showed work from one person, due to the history \nof the repository
    starting after the discovery and fix of the vulnerability. \nA search of the commit
    comments between the release version tags showed no \ndiscussion on the matter.\n"
  answer: false
  question: "Was there any part of the fix that involved one person vouching for \nanother's
    work?\n\nThis can include:\n  * signing off on a commit message\n  * mentioning
    a discussion with a colleague checking the work\n  * upvoting a solution on a
    pull request\n\nAnswer must be true or false.\nWrite a note about how you came
    to the conclusions you did, regardless of what your answer was.\n"
bounty:
  amt: 
  url: 
  announced: 
lessons:
  yagni:
    note: 
    applies: false
  question: |
    Are there any common lessons we have learned from class that apply to this
    vulnerability? In other words, could this vulnerability serve as an example
    of one of those lessons?

    Leave "applies" blank or put false if you did not see that lesson (you do
    not need to put a reason). Put "true" if you feel the lesson applies and put
    a quick explanation of how it applies.

    Don't feel the need to claim that ALL of these apply, but it's pretty likely
    that one or two of them apply.

    If you think of another lesson we covered in class that applies here, feel
    free to give it a small name and add one in the same format as these.
  serial_killer:
    note: "Serialization was a necessary risk for the sake of being able to \nproperly
      test functionality in other parts of systemd. In an ideal world, \nthe user
      would not need to manually specify and/or move around what log \nfiles are needed
      for testing.\n"
    applies: true
  complex_inputs:
    note: 
    applies: false
  distrust_input:
    note: "Similarly to the methods for achieving defense in depth, the \nassumption
      that user log files would be correct all the time was unsafe.\n"
    applies: true
  least_privilege:
    note: 
    applies: false
  native_wrappers:
    note: 
    applies: false
  defense_in_depth:
    note: "Even though the vulnerability was not severe enough to be actively \nabused,
      it was still a scenario that the developer of the native journal \ncode did
      not account for, even while testing. They definitely should have \nadded tests
      with invalid/malformed input files to make sure the error \nhandling was graceful.\n"
    applies: true
  secure_by_default:
    note: 
    applies: false
  environment_variables:
    note: 
    applies: false
  security_by_obscurity:
    note: 
    applies: false
  frameworks_are_optional:
    note: 
    applies: false
reviews: []
sandbox:
  note: "Though part of the vulnerability was related to poor path management, \nthere
    was no place for the user to change the path of the input log files to \nsomething
    unexpected. To exploit the vulnerability, they would not need to \nchange the
    source code's functionality.\n"
  answer: false
  question: "Did this vulnerability violate a sandboxing feature that the system\nprovides?
    \n\nA sandboxing feature is one that allows files, users, or other features \nlimited
    access. Vulnerabilities that violate sandboxes are usually based on \naccess control,
    checking privileges incorrectly, path traversal, and the \nlike.\n\nAnswer should
    be true or false\nWrite a note about how you came to the conclusions you did,
    regardless of\nwhat your answer was.\n"
upvotes: 16
CWE_note: "Resource management error- there was a lack of handling for \ngracefully
  continuing logging after error due to user input. In direct \nrelation to the fix,
  improper resource name/file restrictions also applies.\n"
mistakes:
  answer: "This vulnerability most likely came from a mistake in planning, which \nis
    totally valid for a project that grew as much as systemd. It seems that \nallowing
    the user to have some control over the logging subsystem was an \nafterthought,
    meaning that the original logging functionality was not \nadequately prepared
    to handle the extra complexity and error that comes \nwith any user input. \nAn
    ideal fix to the solution would have been eliminating \nthe assumption in the
    preexisting logging system that all log messages would \nbe formatted correctly.
    Despite that, restricting where logging input files \nare allowed to come from
    also works as a mitigation in this scenario, \nbecause it is unlikely that the
    average user would need to use the feature \nin the first place.\n"
  question: "In your opinion, after all of this research, what mistakes were made
    that\nled to this vulnerability? Coding mistakes? Design mistakes?\nMaintainability?
    Requirements? Miscommunications?\n\nThere can, and usually are, many mistakes
    behind a vulnerability.\n\nRemember that mistakes can come in many forms:\n* slip:
    failing to complete a properly planned step due to inattention\n          e.g.
    wrong key in the ignition\n          e.g. using < instead of <=\n* lapse: failing
    to complete a properly planned step due to memory failure\n          e.g. forgetting
    to put car in reverse before backing up\n          e.g. forgetting to check null\n*
    planning error: error that occurs when the plan is inadequate\n          e.g.
    getting stuck in traffic because you didn’t consider the \n               impact
    of the bridge closing\n          e.g. calling the wrong method\n          e.g.
    using a poor design\n\nThese are grey areas, of course. But do your best to analyze
    the mistakes \naccording to this framework.\n\nLook at the CWE entry for this
    vulnerability and examine the mitigations\nthey have written there. Are they doing
    those? Does the fix look proper?\n\nWrite a thoughtful entry here that people
    in the software engineering\nindustry would find interesting.\n"
nickname: Journald Logging Clogging
subsystem:
  name: journald
  note: "Fairly straightforward- all the reports and included files reference \nthis
    name accordingly.\n"
  question: "What subsystems was the mistake in? These are subsystems WITHIN systemd\n\nTwo
    areas to look:\n  - Bug labels\n  - Directory names\n\nLook at the path of the
    source code files code that were fixed to get\ndirectory names. Look at comments
    in the code. Look at the bug reports how\nthe bug report was tagged. \n\nExample
    systemd subsystems are:\n  * fsck\n  * apparmor\n  * fstab\n  * sd-daemon\n\nName
    should be:\n  * all lowercase English letters\n  * NOT a specific file\n  * can
    have digits, and _-@/\n\nCan be multiple subsystems involved, in which case you
    can make it an array\ne.g.\n    name: [\"subsystemA\", \"subsystemB\"] # ok\n
    \   name: subsystemA # also ok\n"
discovered:
  answer: "This vulnerability was discovered long enough ago that the problem \nreport
    (if it was done on Git) is no longer on the repository. I found \ndetails for
    this vulnerability on the Red Hat bug reporting forums. It \nwas originally reported
    by Jan Lieskovsky, but discovered by Florian \nWeimer from the Red Hat Product
    Security Team.\n"
  contest: false
  question: |
    How was this vulnerability discovered?

    Go to the bug report and read the conversation to find out how this was
    originally found. Answer in longform below in "answer", fill in the date in
    YYYY-MM-DD, and then determine if the vulnerability was found by a Google
    employee (you can tell from their email address). If it's clear that the
    vulenrability was discovered by a contest, fill in the name there.

    The automated, contest, and developer flags can be true, false, or nil.

    If there is no evidence as to how this vulnerability was found, then please
    explain where you looked.
  automated: false
  developer: false
discussion:
  note: "This vulnerability was patched relatively quickly. The only active \ndicussion
    thread I could find about the matter was the Red Hat bug report. \nThe post got
    a reply after the bug had already been resolved. The discussion \nwas a short
    one about which versions of systemd were affected.\n"
  question: "Was there any discussion surrounding this?\n\nA discussion can include
    debates, disputes, or polite talk about how to \nresolve uncertainty.\n\nExample
    include:\n  * Is this out of our scope?\n  * Is this a security?\n  * How should
    we fix this?\n\nJust because you see multiple comments doesn't mean it's a discussion.
    \nFor example:\n  * \"Fix line 10\". \"Ok\" is not what we call a discussion\n
    \ * \"Ping\" (reminding people)\n\nCheck the bugs reports, pull requests, and
    mailing lists archives.\n\nThese answers should be boolean.\n  discussed_as_security:
    true or false\n  any_discussion: true or false\n\nPut any links to disagreements
    you found in the notes section, or any other\ncomment you want to make.\n"
  any_discussion: true
  discussed_as_security: false
stacktrace:
  note: "No traces were provided in either the bug report or in a comment \nabout
    the corresponding fix.\n"
  question: "Are there any stacktraces in the bug reports? \n\nSecondly, if there
    is a stacktrace, is the fix in the same file that the \nstacktrace points to?
    \n\nIf there are no stacktraces, then both of these are false - but be sure to\nmention
    where you checked in the note.\n\nAnswer must be true or false.\nWrite a note
    about how you came to the conclusions you did, regardless of\nwhat your answer
    was.\n"
  any_stacktraces: false
  stacktrace_with_fix: false
description: The native journal protocol in systemd is responsible for parsing predefined
  log data into the program for testing purposes. Its original implementation introduced
  a vulnerability when combined with the previously existing logging functionality.
  Local users could make a maliciously-crafted file to force the logging module to
  block, causing later logging messages from systemd to not be recorded.
unit_tested:
  fix: true
  code: true
  question: |
    Were automated unit tests involved in this vulnerability?
    Was the original code unit tested, or not unit tested? Did the fix involve
    improving the automated tests?

    For code: and fix: - your answer should be boolean.

    For the code_answer below, look not only at the fix but the surrounding
    code near the fix in related directories and determine if and was there were
    unit tests involved for this subsystem.

    For the fix_answer below, check if the fix for the vulnerability involves
    adding or improving an automated test to ensure this doesn't happen again.
  fix_answer: "It looks like a unit test was modified with the fix for the \nvulnerability.
    It ensures that the journald subsystem does not enter an \ninfinite blocking state
    when processing a very large piece of data.\n"
  code_answer: "There were unit tests for this subsystem, but they did not seem \nto
    account for the scenario surrounding the vulnerability.\n"
reported_date: '2012-09-20'
specification:
  note: "None of the bug reports nor relevant commits mention any violation of \nspecifications.
    This logging subsystem looks to be unique to systemd.\n"
  answer: false
  instructions: |
    Is there mention of a violation of a specification? For example, the POSIX
    spec, an RFC spec, a network protocol spec, or some other requirements
    specification.

    Be sure to check the following artifacts for this:
      * bug reports
      * security advisories
      * commit message
      * pull request
      * mailing lists
      * anything else

    The answer field should be boolean. In answer_note, please explain
    why you come to that conclusion.
announced_date: '2013-06-12'
curation_level: 1
published_date: '2013-10-28'
forgotten_check:
  note: "The main fix for the vulnerability involved adding a check when \nprocessing
    a native logging file (whether from user or systemd) to make \nsure that its absolute
    path was in a safe Linux OS location.\n"
  answer: true
  question: "Does the fix for the vulnerability involve adding a forgotten check?\n\nA
    \"forgotten check\" can mean many things. It often manifests as the fix \ninserting
    an entire if-statement or a conditional to an existing \nif-statement. Or a call
    to a method that checks something.\n\nExample of checks can include:\n  * null
    pointer checks\n  * check the current role, e.g. root\n  * boundary checks for
    a number\n  * consult file permissions\n  * check a return value\n\nAnswer must
    be true or false.\nWrite a note about how you came to the conclusions you did,
    regardless of\nwhat your answer was.\n"
CWE_instructions: |
  Please go to http://cwe.mitre.org and find the most specific, appropriate CWE
  entry that describes your vulnerability. We recommend going to
  https://cwe.mitre.org/data/definitions/699.html for the Software Development
  view of the vulnerabilities. We also recommend the tool
  http://www.cwevis.org/viz to help see how the classifications work.

  If you have anything to note about why you classified it this way, write
  something in CWE_note. This field is optional.

  Just the number here is fine. No need for name or CWE prefix. If more than one
  apply here, then place them in an array like this
    CWE: ["123", "456"] # this is ok
    CWE: [123, 456]     # also ok
    CWE: 123            # also ok
autodiscoverable:
  note: "The native journal functionality requires log data files to be \nformatted
    in a specific manner, and then further purposely formatted to \nabuse the vulnerability
    in question. This is not automatable without \nspecific subsystem knowledge.\n"
  answer: false
  instructions: |
    Is it plausible that a fully automated tool could have discovered
    this? These are tools that require little knowledge of the domain,
     e.g. automatic static analysis, compiler warnings, fuzzers.

    Examples for true answers: SQL injection, XSS, buffer overflow

    In systemd, the actually use OZZ Fuzz. If there's a link to it, add it here.

    Examples for false: RFC violations, permissions issues, anything
    that requires the tool to be "aware" of the project's
    domain-specific requirements.

    The answer field should be boolean. In answer_note, please explain
    why you come to that conclusion.
vcc_instructions: "The vulnerability-contributing commits.\n\nThese are found by our
  tools by traversing the Git Blame history, where we \ndetermine which commit(s)
  introduced the functionality. \n\nLook up these VCC commits and verify that they
  are not simple refactorings, \nand that they are, in fact introducing the vulnerability
  into the system. \nOften, introducing the file or function is where the VCC is,
  but VCCs can be \nanything. \n\nPlace any notes you would like to make in the notes
  field.\n"
bugs_instructions: "What bugs and/or pull requests are involved in this vulnerability?
  \n\nFor systemd, this is typically their GitHub issues, but could also include \nbugs
  from other databases. Put a URL instead of a single number.\n"
yaml_instructions: |
  =================
  ===YAML Primer===
  =================
  This is a dictionary data structure, akin to JSON.
  Everything before a colon is a key, and the values here are usually strings
  For one-line strings, you can just use quotes after the colon
  For multi-line strings, as we do for our instructions, you put a | and then
  indent by two spaces

  For readability, we hard-wrap multi-line strings at 80 characters. This is
  not absolutely required, but appreciated.
fixes_instructions: "Please put the commit hash in \"commit\" below.\n\nThis must
  be a git commit hash from the systemd source repo, a  40-character \nhexademical
  string/\n\nPlace any notes you would like to make in the notes field.\n"
bounty_instructions: |
  If you came across any indications that a bounty was paid out for this
  vulnerability, fill it out here. Or correct it if the information already here
  was wrong. Otherwise, leave it blank.
interesting_commits:
  commits:
  - note: "This commit shows that there was some consideration for what could \ngo
      wrong with the logging, when left up to the local user. This particular \ncommit
      protects against potential log overflow exploits.\n"
    commit: 505b6a61c22d5565e9308045c7b9bf79f7d0517e
  question: |
    Are there any interesting commits between your VCC(s) and fix(es)?

    Use this to specify any commits you think are notable in some way, and
    explain why in the note.
order_of_operations:
  note: "The method calls in the log file processing chain (after being kicked \noff
    by the fixed portion of code) remained entirely in the same order.\n"
  answer: false
  question: "Does the fix for the vulnerability involve correcting an order of \noperations?\n\nThis
    means the fix involves moving code around or changing the order of \nhow things
    are done. \n\nAnswer must be true or false.\nWrite a note about how you came to
    the conclusions you did, regardless of\nwhat your answer was.\n"
curated_instructions: "If you are manually editing this file, then you are \"curating\"
  it. \n\nSet the version number that you were given in the instructions. \n \nThis
  will enable additional editorial checks on this file to make sure you \nfill everything
  out properly. If you are a student, we cannot accept your work\nas finished unless
  curated is properly updated. \n"
upvotes_instructions: |
  For the first round, ignore this upvotes number.

  For the second round of reviewing, you will be giving a certain amount of
  upvotes to each vulnerability you see. Your peers will tell you how
  interesting they think this vulnerability is, and you'll add that to the
  upvotes score on your branch.
nickname_instructions: |
  A catchy name for this vulnerability that would draw attention it.
  If the report mentions a nickname, use that.
  Must be under 30 characters. Optional.
reported_instructions: |
  What date was the vulnerability reported to the security team? Look at the
  security bulletins and bug reports. It is not necessarily the same day that
  the CVE was created.  Leave blank if no date is given.

  Please enter your date in YYYY-MM-DD format.
announced_instructions: |
  Was there a date that this vulnerability was announced to the world? You can
  find this in changelogs, blogs, bug reports, or perhaps the CVE date.

  This is not the same as published date in the NVD - that is below.

  Please enter your date in YYYY-MM-DD format.
published_instructions: |
  Is there a published fix or patch date for this vulnerability?
  Please enter your date in YYYY-MM-DD format.
description_instructions: |
  You can get an initial description from the CVE entry on cve.mitre.org. These
  descriptions are a fine start, but they can be kind of jargony.

  Rewrite this description IN YOUR OWN WORDS. Make it interesting and easy to
  read to anyone with some programming experience. We can always pull up the NVD
  description later to get more technical.

  Try to still be specific in your description, but remove project-specific
  stuff. Remove references to versions, specific filenames, and other jargon
  that outsiders to this project would not understand. Technology like "regular
  expressions" is fine, and security phrases like "invalid write" are fine to
  keep too.

  Your target audience is people just like you before you took any course in
  security