| 1 |
micah |
2985 |
A Narrative Introduction to the Testing Security Tracker |
| 2 |
|
|
|
| 3 |
|
|
|
| 4 |
|
|
About |
| 5 |
|
|
----- |
| 6 |
|
|
|
| 7 |
nion |
6527 |
Everything that Testing Security does is publicly available, as in |
| 8 |
micah |
2985 |
"Debian doesn't hide problems" available. |
| 9 |
|
|
|
| 10 |
|
|
The best thing about our tracking 'system' is that it is very basic. |
| 11 |
|
|
There is no horrible overhead of web-based ticket/issue trackers, its |
| 12 |
|
|
just a subversion repository and some text files that we |
| 13 |
|
|
collaboratively edit and then some scripts to parse these files and |
| 14 |
|
|
generate useful reports available online. Everything is designed to be |
| 15 |
tedp-guest |
6022 |
very simple to use, transparent and easy to see what other people are |
| 16 |
micah |
2985 |
working on so you can work on other things. |
| 17 |
|
|
|
| 18 |
|
|
Why are these issues disclosed to the public? |
| 19 |
|
|
|
| 20 |
|
|
The way we look at it is that 99% of all vulnerabilities are already |
| 21 |
|
|
public, and 1% are vendor-sec/embargoed issues. |
| 22 |
|
|
|
| 23 |
|
|
Stable security deals with embargoed/vendor-sec issues, we don't, we |
| 24 |
|
|
deal with issues that have already been assigned CVE numbers (although |
| 25 |
|
|
we often times request these assignments), have been posted to common |
| 26 |
|
|
security mailing lists, or are seen in commit logs of software that is |
| 27 |
|
|
tracked (such as the Linux Kernel). |
| 28 |
|
|
|
| 29 |
|
|
It is our philosophy that if the Internet knows that there is a |
| 30 |
|
|
vulnerability in something, then we better know about it and the |
| 31 |
|
|
package maintainer needs to know about it and it needs to be fixed as |
| 32 |
|
|
soon as possible. It doesn't make sense to hide issues that everyone |
| 33 |
|
|
knows about already, in fact users have told us that they prefer to |
| 34 |
|
|
know not only when a package they have installed is vulnerable (so |
| 35 |
|
|
they can disable it or firewall it off, or patch it or whatever), but |
| 36 |
tedp-guest |
6022 |
to also know that Debian is working on a fix. Transparency is what our |
| 37 |
nion |
6527 |
users expect, and what they deserve. Tracking publicly known issues |
| 38 |
micah |
2985 |
openly (and the occasional unfortunate embargoed issue privately) is |
| 39 |
|
|
good for the project as a whole, especially the public's perception of |
| 40 |
|
|
the project. |
| 41 |
|
|
|
| 42 |
micah |
3520 |
Gentle Introduction |
| 43 |
micah |
2985 |
-------------------- |
| 44 |
|
|
|
| 45 |
|
|
This following will give you a basic walk-through of how the files are |
| 46 |
|
|
structured and how we do our work tracking issues. There is much more |
| 47 |
|
|
that can be documented, but it is difficult to get all the issues |
| 48 |
|
|
notated and updated. It is easier to get a basic idea and then |
| 49 |
|
|
extrapolate from there how to do the rest. Ok, thats a bad excuse, so |
| 50 |
|
|
the full information should be filled in. |
| 51 |
|
|
|
| 52 |
|
|
The best way to understand is to check out our repository from |
| 53 |
|
|
subversion so you have the files on your computer and can follow along |
| 54 |
|
|
at home. To do this, you need an Alioth account, and then you just |
| 55 |
|
|
need to do the following: |
| 56 |
|
|
|
| 57 |
|
|
svn co svn+ssh://svn.debian.org/svn/secure-testing |
| 58 |
|
|
|
| 59 |
|
|
This will check out our working repository into a directory called |
| 60 |
|
|
secure-testing. Inside this directory are a number of subdirectories. |
| 61 |
|
|
The data directory is where we do most of our work. |
| 62 |
|
|
|
| 63 |
jmm-guest |
2991 |
If you don't need write access, you can of course check out our files |
| 64 |
|
|
without an Alioth account as well: |
| 65 |
|
|
|
| 66 |
|
|
svn co svn://svn.debian.org/svn/secure-testing |
| 67 |
|
|
|
| 68 |
micah |
2985 |
Automatic Issue Updates |
| 69 |
|
|
----------------------- |
| 70 |
|
|
Twice a day a cronjob runs that pulls down the latest full CVE lists |
| 71 |
thijs |
8409 |
from Mitre, this automatically gets checked into data/CVE/list, and |
| 72 |
|
|
also syncs that file with other lists like data/DSA/list and |
| 73 |
|
|
data/DTSA/list. |
| 74 |
|
|
|
| 75 |
|
|
We get notified via either email |
| 76 |
micah |
2985 |
(http://lists.alioth.debian.org/mailman/listinfo/secure-testing-commits) |
| 77 |
|
|
of every SVN commit, by RSS feed |
| 78 |
|
|
(http://svn.debian.org/wsvn/secure-testing/?op=rss&rev=0&sc=0&isdir=1) |
| 79 |
|
|
or via the CIA bot on #debian-security on OFTC. For example, the bot |
| 80 |
|
|
will say in the channel: |
| 81 |
|
|
|
| 82 |
thijs |
8409 |
17:14 < CIA-1> joeyh * r2314 /data/CVE/list: automatic update |
| 83 |
micah |
2985 |
|
| 84 |
|
|
Most of our work is taking the new issues that Mitre releases and |
| 85 |
|
|
processing them so that the tracking data is correct. Read on for how we |
| 86 |
|
|
do this. |
| 87 |
|
|
|
| 88 |
neilm |
5083 |
Processing TODO entries |
| 89 |
micah |
2985 |
----------------------- |
| 90 |
|
|
The Mitre update typically manifests in new CVE entries. So what we do |
| 91 |
|
|
is to update our svn repository and then edit data/CVE/list and look |
| 92 |
|
|
for new TODO entries. These will often be in blocks of 10-50 or so, |
| 93 |
|
|
depending on how many new issues they have assigned. Depending on how |
| 94 |
|
|
you feel you will "claim" a block of say 10 new entries by |
| 95 |
|
|
putting your name in the file at the beginning and the end of the new |
| 96 |
|
|
TODO entries and then commit the repository. This looks like this: |
| 97 |
|
|
|
| 98 |
|
|
begin claimed by jmm |
| 99 |
|
|
CVE-2005-4066 (Total Commander 6.53 uses weak encryption to store FTP |
| 100 |
|
|
usernams and ...) |
| 101 |
|
|
TODO: check |
| 102 |
|
|
CVE-2005-4065 (SQL injection vulnerability in the search module in |
| 103 |
|
|
Edgewall Trac ...) |
| 104 |
|
|
TODO: check |
| 105 |
|
|
CVE-2005-4030 (SQL injection vulnerability in Quicksilver Forums |
| 106 |
|
|
before 1.5.1 allows ...) |
| 107 |
|
|
TODO: check |
| 108 |
|
|
end claimed by jmm |
| 109 |
|
|
|
| 110 |
|
|
Once these are checked-in, then others will not do work on these TODO |
| 111 |
|
|
issues. |
| 112 |
|
|
|
| 113 |
nion |
11854 |
IMPORTANT: make sure to read: http://lists.alioth.debian.org/pipermail/secure-testing-team/2009-May/002394.html |
| 114 |
|
|
|
| 115 |
micah |
2985 |
Issues Not-For-Us (NFU) |
| 116 |
|
|
----------------------- |
| 117 |
thijs |
6422 |
Processing your claimed entries is done by first seeing if the issue |
| 118 |
micah |
2985 |
is related to any software packaged in Debian, if it isn't a package |
| 119 |
nion |
10534 |
in Debian and has no ITP then you note that in the file. Another case |
| 120 |
|
|
are meta packages that only provide a downloader (e.g. flashplugin-nonfree). |
| 121 |
|
|
There is no way to mark such packages as we have no influence on the version |
| 122 |
|
|
and technically the code is not present in Debian. |
| 123 |
micah |
2985 |
|
| 124 |
nion |
10534 |
Example: |
| 125 |
|
|
|
| 126 |
micah |
2985 |
CVE-2005-3018 (Apple Safari allows remote attackers to cause a denial of |
| 127 |
|
|
service ...) |
| 128 |
|
|
NOT-FOR-US: Safari |
| 129 |
|
|
|
| 130 |
stef-guest |
8325 |
There is a tool that helps with sorting out all the NOT-FOR-US issues: |
| 131 |
|
|
See "bin/check-new-issues -h". For the search functions in |
| 132 |
|
|
check-new-issues to work, you need to have unstable in your |
| 133 |
|
|
sources.list and have done "apt-get update" and "apt-file update". |
| 134 |
|
|
Having libterm-readline-gnu-perl installed helps, too. Unfortunately, |
| 135 |
|
|
check-new-issues does not yet support the "claimed by" tags mentioned above. |
| 136 |
|
|
|
| 137 |
nion |
8327 |
Please also make sure to check the wnpp list for possible <itp> items and |
| 138 |
|
|
the ftp-master removal list to see if the issue way maybe present in the past |
| 139 |
|
|
but the package was removed |
| 140 |
|
|
|
| 141 |
jmm-guest |
3029 |
Reserved entries |
| 142 |
|
|
---------------- |
| 143 |
|
|
Several security problems have coordinated dates of public disclosure, |
| 144 |
|
|
i.e. a CVE identifier has been assigned to a problem, but it's not |
| 145 |
|
|
public yet. Also, several vendors have a pool of CVE ids they can |
| 146 |
|
|
assign to problems that are detected in their products. Such entries |
| 147 |
|
|
are marked as RESERVED in the tracker: |
| 148 |
micah |
2985 |
|
| 149 |
jmm-guest |
3029 |
CVE-2005-1432 |
| 150 |
|
|
RESERVED |
| 151 |
|
|
|
| 152 |
|
|
Rejected entries |
| 153 |
|
|
---------------- |
| 154 |
|
|
Sometimes there are CVE assignments that later turn out to be duplicates, |
| 155 |
|
|
mistakes or non-issues. These items are reverted and turned into REJECTED |
| 156 |
|
|
entries: |
| 157 |
|
|
|
| 158 |
|
|
CVE-2005-4129 |
| 159 |
|
|
REJECTED |
| 160 |
|
|
|
| 161 |
micah |
2985 |
ITP packages |
| 162 |
|
|
------------ |
| 163 |
|
|
If it is a package that someone has filed an RFP or ITP for, then that |
| 164 |
|
|
is also noted, so it can be tracked to make sure that the issue is |
| 165 |
|
|
resolved before the package enters the archive: |
| 166 |
|
|
|
| 167 |
|
|
CVE-2004-2525 (Cross-site scripting (XSS) vulnerability in compat.php |
| 168 |
|
|
in Serendipity ...) |
| 169 |
|
|
- serendipity <itp> (bug #312413) |
| 170 |
|
|
|
| 171 |
|
|
|
| 172 |
|
|
Packages in the archive |
| 173 |
|
|
----------------------- |
| 174 |
neilm |
5083 |
If it is a package in Debian, look to see if the package is affected or |
| 175 |
|
|
not (sometimes newer versions that have the fixes have already been |
| 176 |
|
|
uploaded). |
| 177 |
micah |
2985 |
|
| 178 |
|
|
If the version has been fixed already, note the package name and the |
| 179 |
|
|
Debian version that fixes it and assign a severity level to it, for |
| 180 |
|
|
example: |
| 181 |
|
|
|
| 182 |
|
|
CVE-2005-2596 (User.php in Gallery, as used in Postnuke, allows users |
| 183 |
|
|
with any Admin ...) |
| 184 |
|
|
- gallery 1.5-2 (medium) |
| 185 |
|
|
|
| 186 |
neilm |
5083 |
If it hasn't been fixed, we determine if there has been a bug filed |
| 187 |
|
|
about the issue, and if not, file one and then note it in the list |
| 188 |
|
|
(again with a severity level): |
| 189 |
micah |
2985 |
|
| 190 |
|
|
CVE-2005-3054 (fopen_wrappers.c in PHP 4.4.0, and possibly other |
| 191 |
|
|
versions, does not ...) |
| 192 |
|
|
- php4 <unfixed> (bug #353585; medium) |
| 193 |
|
|
- php5 <unfixed> (bug #353585; medium) |
| 194 |
|
|
|
| 195 |
stef-guest |
4001 |
Bug numbers can be added as in the example above. To avoid duplicate bugs, |
| 196 |
|
|
"bug filed" can be added instead of "bug #123456" when the bug report has |
| 197 |
gilbert-guest |
11654 |
been sent but the bug number is not yet known (however, it is more |
| 198 |
|
|
desirable to file the bug, wait for the BTS to assign a number, then update |
| 199 |
|
|
the entry in the CVE list so that complete information is always available |
| 200 |
|
|
in the tracker). The bug number is important because it makes it clear |
| 201 |
|
|
that the maintainer has been contacted about the problem, and that they are |
| 202 |
gilbert-guest |
12570 |
aware of their responsibility to work swiftly toward a fix. |
| 203 |
jmm-guest |
3039 |
|
| 204 |
gilbert-guest |
11654 |
Since CVEs often drop in bulk, submission of multiple CVEs in a single bug |
| 205 |
|
|
report is permissable and encouraged. However, some maintainers have |
| 206 |
|
|
indicated a preference for only one issue per bug report. The following |
| 207 |
|
|
is a list of packages for which each CVE should be reported separately: |
| 208 |
|
|
- php5 |
| 209 |
|
|
|
| 210 |
thijs |
8409 |
A special exception is made for kernel related issues. The kernel-sec |
| 211 |
|
|
group will take care of them and file bugs if needed. |
| 212 |
nion |
6896 |
|
| 213 |
nion |
8437 |
If you wan't to report a bug, bin/report-vuln might be helpful in creating |
| 214 |
|
|
the bug report. |
| 215 |
|
|
|
| 216 |
jmm-guest |
3029 |
If a vulnerability does not affect Debian, e.g. because the vulnerable |
| 217 |
|
|
code is not contained, it is marked as <not-affected>: |
| 218 |
|
|
|
| 219 |
|
|
CVE-2004-2628 (Multiple directory traversal vulnerabilities in thttpd 2.07 beta 0.4, ...) |
| 220 |
|
|
- thttpd <not-affected> (Windows-specific vulnerabilities) |
| 221 |
|
|
|
| 222 |
|
|
<not-affected> is also used if a vulnerability was fixed before a |
| 223 |
|
|
package was uploaded into the Debian archive. |
| 224 |
|
|
|
| 225 |
|
|
Sometimes there are cases, where a vulnerability hasn't been fixed with |
| 226 |
|
|
a code change, but simply by deciding that a package is that broken that |
| 227 |
|
|
it needs to be removed from the archive entirely. This is tracked with |
| 228 |
|
|
the <removed> tag: |
| 229 |
|
|
|
| 230 |
|
|
CVE-2005-1435 (Open WebMail (OWM) before 2.51 20050430 allows remote authenticated ...) |
| 231 |
|
|
- openwebmail <removed> |
| 232 |
|
|
|
| 233 |
fw |
5669 |
After a new Debian release, some packages vanish from the database, |
| 234 |
|
|
and consistency checks might fail. In this case, a single <removed> |
| 235 |
|
|
entry needs to be added to an input file, or the package name should |
| 236 |
|
|
be included in the data/packages/removed-packages file. |
| 237 |
jmm-guest |
3029 |
|
| 238 |
nion |
12563 |
Also note that it is sufficient to mark a package as removed in unstable. |
| 239 |
gilbert-guest |
12564 |
The tracker is aware of which package is present in which distribution |
| 240 |
nion |
12563 |
and marks other distributions that still contain the package automagically |
| 241 |
gilbert-guest |
12565 |
as unfixed. For example, if libxml is in oldstable, but not stable or |
| 242 |
|
|
unstable, then: |
| 243 |
nion |
12563 |
|
| 244 |
gilbert-guest |
12564 |
- libxml <removed> |
| 245 |
|
|
|
| 246 |
|
|
will track oldstable as affected, but stable and unstable as not-affected. |
| 247 |
|
|
|
| 248 |
jmm-guest |
2991 |
Severity levels |
| 249 |
|
|
--------------- |
| 250 |
|
|
These levels are mostly used to prioritize the order in which security |
| 251 |
|
|
problems are resolved. Anyway, we have a rough overview on how you should |
| 252 |
djoume-guest |
5280 |
assess these levels. |
| 253 |
jmm-guest |
2991 |
|
| 254 |
|
|
unimportant: This problem does not affect the Debian binary package, e.g. |
| 255 |
djoume-guest |
5280 |
a vulnerable source file, which is not built, a vulnerable file |
| 256 |
|
|
in doc/foo/examples/, PHP Safe mode bugs, path disclosure (doesn't |
| 257 |
|
|
matter on Debian). |
| 258 |
|
|
All "non-issues in practice" fall also into this category, like |
| 259 |
|
|
issues only "exploitable" if the code in question is setuid root, |
| 260 |
|
|
exploits which only work if someone already has administrative |
| 261 |
|
|
privileges or similar. |
| 262 |
|
|
|
| 263 |
jmm-guest |
2991 |
low : A security problem, which has only mild security implications |
| 264 |
nion |
7786 |
(local DoS, /tmp file races and so on). |
| 265 |
djoume-guest |
5280 |
|
| 266 |
|
|
medium : For anything which permits code execution after user interaction. |
| 267 |
|
|
Local privilege escalation vulnerabilities are in this category as |
| 268 |
|
|
well, or remote privilege escalation if it's constrained to the |
| 269 |
|
|
application (i.e. no shell access to the underlying system, such |
| 270 |
|
|
as simple cross-site scripting). Most remote DoS vulnerabilities |
| 271 |
|
|
fall into this category, too. |
| 272 |
|
|
|
| 273 |
jmm-guest |
2991 |
high : A typical, exploitable security problem, which you'll really |
| 274 |
jmm-guest |
3029 |
like to fix or at least implement a workaround. This could |
| 275 |
jmm-guest |
2991 |
be because the vulnerable code is very broadly used, because |
| 276 |
|
|
an exploit is in the wild or because the attack vector is |
| 277 |
djoume-guest |
5280 |
very wide. |
| 278 |
|
|
Should be put into that category anything that permits an attacker |
| 279 |
|
|
to execute arbitrary code on the vulnerable system (with or |
| 280 |
|
|
without root privileges) and high-impact denial-of-service bugs |
| 281 |
|
|
(for instance, an IPv4 forwarding path vulnerability which |
| 282 |
|
|
requires only very few packets to exploit). |
| 283 |
|
|
Significant defects in security software can be rated "high" as |
| 284 |
|
|
well (for instance, a vulnerability in a piece of cryptographic |
| 285 |
|
|
software which flags forged digital signatures as genuine). |
| 286 |
jmm-guest |
2991 |
|
| 287 |
djoume-guest |
5280 |
|
| 288 |
|
|
Certain packages may get higher or lower rating than usual, based on |
| 289 |
|
|
their importance. |
| 290 |
|
|
|
| 291 |
|
|
|
| 292 |
micah |
2985 |
NOTE and TODO entries |
| 293 |
|
|
--------------------- |
| 294 |
|
|
There are many instances where more work has to be done to determine |
| 295 |
|
|
if something is affected, and you might not be able to do this at the |
| 296 |
|
|
time. These entries can have their TODO line changed to something |
| 297 |
|
|
descriptive so that it is clear what remains to be done. For example: |
| 298 |
|
|
|
| 299 |
|
|
CVE-2005-3990 (Directory traversal vulnerability in FastJar 0.93 |
| 300 |
|
|
allows remote ...) |
| 301 |
|
|
TODO: check, whether fastjar from the gcc source packages is affected |
| 302 |
|
|
|
| 303 |
|
|
It is also useful to add information to issues as you find it, so that |
| 304 |
|
|
when others go to look at an issue and want to know why you marked it |
| 305 |
|
|
as you did, or need a reference, it will be there. The more |
| 306 |
|
|
information left, the better. For example, the following entry lets |
| 307 |
|
|
you know that CVE-2005-3258 doesn't affect the squid that we have |
| 308 |
|
|
because the issue was introduced in a patch that was never applied to |
| 309 |
|
|
the Debian package: |
| 310 |
|
|
|
| 311 |
|
|
CVE-2005-3258 (The rfc1738_do_escape function in ftp.c for Squid 2.5 |
| 312 |
|
|
STABLE11 and ...) |
| 313 |
|
|
- squid <not-affected> (bug #334882; medium) |
| 314 |
|
|
NOTE: Bug was introduced in a patch to squid-2.5.STABLE10, |
| 315 |
|
|
NOTE: this patch was never applied to the Debian package. |
| 316 |
|
|
|
| 317 |
fw |
11279 |
CVE assignments |
| 318 |
|
|
--------------- |
| 319 |
|
|
Debian can only assign CVE names from its own pool for issues which |
| 320 |
|
|
are not public. To request a CVE from the Debian pool, write to |
| 321 |
|
|
<security@debian.org> and include a description which follows CVE |
| 322 |
fw |
11280 |
conventions. To request a CVE for public issues, write to Mitre and |
| 323 |
fw |
11279 |
possibly to the moderated oss-security list. In the meantime, you can |
| 324 |
|
|
add an entry of the form |
| 325 |
|
|
|
| 326 |
|
|
CVE-2009-XXXX [optipng array overflow] |
| 327 |
|
|
- optipng 0.6.2.1-1 (low) |
| 328 |
|
|
NOTE: http://secunia.com/advisories/34035/ |
| 329 |
|
|
|
| 330 |
|
|
in the data/CVE/list file. It is desirable to include references |
| 331 |
|
|
which uniquely identify the issue, such as a permanent link to an |
| 332 |
|
|
entry in the upstream bug tracker, or a bug in the Debian BTS. If the |
| 333 |
|
|
issue is likely present in unstable, a bug should be filed to help the |
| 334 |
|
|
maintainer to track it. |
| 335 |
|
|
|
| 336 |
|
|
Lack of CVE entries should not block advisory publication which are |
| 337 |
|
|
otherwise ready, but we should strieve to release fully |
| 338 |
|
|
cross-referenced advisories nevertheless. |
| 339 |
|
|
|
| 340 |
jmm-guest |
3027 |
Distribution tags |
| 341 |
|
|
----------------- |
| 342 |
|
|
Our data is primarily targeted at sid, as we track the version that |
| 343 |
|
|
a certain issue was fixed in sid. The Security Tracker web site (see |
| 344 |
|
|
below) derives information about the applicability of a vulnerability |
| 345 |
|
|
to stable and oldstable from the list of DSAs issued by the security |
| 346 |
|
|
team and the fact that a source package is part of a release. |
| 347 |
|
|
Distribution tags can be used to denote information about a vulnerability |
| 348 |
|
|
for the version of a package in a specific release. An example: |
| 349 |
micah |
2985 |
|
| 350 |
jmm-guest |
3027 |
CVE-2005-3974 (Drupal 4.5.0 through 4.5.5 and 4.6.0 through 4.6.3, when running on ...) |
| 351 |
|
|
- drupal 4.5.6-1 (low) |
| 352 |
|
|
[sarge] - drupal <not-affected> (Only vulnerable if running PHP 5) |
| 353 |
|
|
|
| 354 |
|
|
Drupal has been fixed since 4.5.6, however Drupal from Sarge still isn't |
| 355 |
|
|
vulnerable as the vulnerability is only effective when run under PHP 5, |
| 356 |
|
|
which isn't part of Sarge. |
| 357 |
|
|
|
| 358 |
micah |
2985 |
Generated Reports |
| 359 |
|
|
----------------- |
| 360 |
|
|
All of this tracking information gets automatically parsed and |
| 361 |
|
|
compared against madison to determine what has been fixed and what is |
| 362 |
joeyh |
4589 |
still waiting, this results in this website: |
| 363 |
micah |
2985 |
|
| 364 |
joeyh |
4589 |
http://security-tracker.debian.net/ |
| 365 |
micah |
2985 |
|
| 366 |
joeyh |
4589 |
It incorporates package lists and parses distribution lists and can |
| 367 |
|
|
thus be used to |
| 368 |
|
|
- Present the security history of a package |
| 369 |
|
|
- Provide overviews of vulnerable packages in stable, testing, sid and |
| 370 |
|
|
oldstable (it still has some false positives, wrt packages in |
| 371 |
|
|
stable that are present in stable, but not vulnerable, but these |
| 372 |
|
|
will be ironed out soon). The oldstable data is likely inaccurate. |
| 373 |
|
|
- Generate a list of packages that are subject to security problems, but |
| 374 |
|
|
stuck in testing migration due to problems with the dependency chain |
| 375 |
|
|
and thus candidates for a DTSA |
| 376 |
tedp-guest |
6022 |
- Generate a list of TODO issues that need to be addressed |
| 377 |
joeyh |
4589 |
- Generate a list of packages that will enter Debian soon and need to |
| 378 |
|
|
be checked for security problems |
| 379 |
|
|
- Generate a list of provisional IDs that need to be turned into proper |
| 380 |
|
|
CVE entries |
| 381 |
|
|
- Show some potential problems in the data pool (e.g. misspelled package |
| 382 |
|
|
names not found in the packages list, or potentially missing epochs) |
| 383 |
micah |
2985 |
|
| 384 |
joeyh |
4589 |
For every security problem it displays |
| 385 |
|
|
- The CVE information |
| 386 |
|
|
- A severity assessment by NVD |
| 387 |
|
|
- Cross references to DTSAs, DSAs and bugs in the BTS |
| 388 |
|
|
- The status of a security problem in stable, oldstable, testing and sid |
| 389 |
|
|
- Additional notes from our tracker |
| 390 |
micah |
2985 |
|
| 391 |
jmm-guest |
3030 |
The DSA list |
| 392 |
|
|
------------ |
| 393 |
|
|
We maintain a list of all DSA advisories issued by the stable security |
| 394 |
|
|
team. This information is used to derive information about the state |
| 395 |
|
|
of security problems for the stable and oldstable distribution. An |
| 396 |
|
|
entry for a DSA looks like this: |
| 397 |
|
|
|
| 398 |
|
|
[21 Nov 2005] DSA-903-1 unzip - race condition |
| 399 |
|
|
{CVE-2005-2475} |
| 400 |
|
|
[woody] - unzip 5.50-1woody4 |
| 401 |
|
|
[sarge] - unzip 5.52-1sarge2 |
| 402 |
|
|
NOTE: fixed in testing at time of DSA |
| 403 |
|
|
|
| 404 |
micah |
3614 |
The first line tracks the date, when a DSA was issued, the DSA |
| 405 |
|
|
identifier, the affected source package and the type of vulnerability. |
| 406 |
|
|
The second line performs a cross-reference to the entry in CVE/list |
| 407 |
|
|
that maintains the state of the vulnerability in sid. Every entry that |
| 408 |
|
|
is added like this to DSA/list is parsed by a script and automatically |
| 409 |
|
|
added to CVE/list. The next lines contain the fixes for stable and |
| 410 |
|
|
optionally oldstable, addressed with distribution tags. You may add |
| 411 |
|
|
NOTE: entries freely, we use a NOTE entry for statistical purposes |
| 412 |
|
|
that tracks, when a fix has reached testing relative to the time when |
| 413 |
|
|
it hit stable. |
| 414 |
jmm-guest |
3030 |
|
| 415 |
micah |
3615 |
There is no need to add anything to CVE/list for a DSA, the DSA |
| 416 |
|
|
cross-reference will be added automatically by the cron job. However, |
| 417 |
|
|
you do need to add [sarge] or [woody] entries to CVE/list when there |
| 418 |
micah |
3614 |
is a 'no-dsa' or 'not-affected' condition. |
| 419 |
|
|
|
| 420 |
fw |
3107 |
The bin/dsa2list script can be used to generate a template for a new |
| 421 |
thijs |
8409 |
DSA entry once the official DSA is published on debian-security-announce. |
| 422 |
|
|
You should not blindly trust the script output and double-check it, though. |
| 423 |
fw |
3107 |
|
| 424 |
nion |
6899 |
Checking your changes |
| 425 |
|
|
--------------------- |
| 426 |
|
|
Commits are checked for syntax errors before they are actually committed, |
| 427 |
|
|
and you'll receive an error and your commit is aborted if it is in error. |
| 428 |
|
|
To check your changes yourself beforehand, use "make check-syntax" from |
| 429 |
|
|
the root of the svn directory. |
| 430 |
|
|
|
| 431 |
micah |
2985 |
Following up on security issues |
| 432 |
|
|
------------------------------- |
| 433 |
|
|
By simply loading this page and doing a little gardening of the |
| 434 |
|
|
different issues many things can be done. One thing is that you can |
| 435 |
|
|
read all the bug reports of each issue and see if new information has |
| 436 |
|
|
been added to the end that might provide updated or changed |
| 437 |
|
|
information (such as if an issue has been closed, or a version of the |
| 438 |
|
|
package has been uploaded that contains the fix). It is also useful to |
| 439 |
|
|
follow-up on the issues to prod the maintainer to deal with the issue, |
| 440 |
|
|
which they may have forgotten about. |
| 441 |
|
|
|
| 442 |
|
|
|
| 443 |
jmm-guest |
12004 |
Tracking of security bugs in the BTS and linking them to a user tag by CVE |
| 444 |
|
|
-------------------------------------------------------------------------- |
| 445 |
|
|
|
| 446 |
|
|
There's an automated tagging of security-related bugs to CVE IDs through |
| 447 |
|
|
the user tag security for the user debian-security@lists.debian.org |
| 448 |
|
|
|
| 449 |
|
|
All bugs added to the tracker are automatically tagged. You can use |
| 450 |
|
|
the search |
| 451 |
|
|
http://bugs.debian.org/cgi-bin/pkgreport.cgi?tag=security;users=debian-security@lists.debian.org;exclude=tracked |
| 452 |
|
|
to find all bugs not yet present in the tracker. |
| 453 |
|
|
|
| 454 |
|
|
All bug numbers added to the tracked are automatically associated |
| 455 |
|
|
to the relevant user tag. |
| 456 |
|
|
|
| 457 |
|
|
If you checked an issue which doesn't need to be added to the tracked |
| 458 |
|
|
(e.g. because it's not security-relevant or otherwise bogus you can either |
| 459 |
|
|
remove the security tag from the bugs or sent a mail to control@bugs.debian.org |
| 460 |
|
|
with the following content: |
| 461 |
|
|
|
| 462 |
derevko-guest |
12337 |
user debian-security@lists.debian.org |
| 463 |
jmm-guest |
12004 |
usertag $BUGNUM + tracked |
| 464 |
|
|
|
| 465 |
micah |
2985 |
IRC Channel |
| 466 |
|
|
----------- |
| 467 |
|
|
We hang-out on #debian-security on OFTC, stop by the IRC channel if |
| 468 |
|
|
you'd like, also we can add you to the alioth project so you have svn |
| 469 |
|
|
write permission and you can test drive it on the testing issues for |
| 470 |
|
|
however long you like to get an idea or feel comfortable (and hey it |
| 471 |
|
|
helps!) |
| 472 |
|
|
|
| 473 |
|
|
|
| 474 |
|
|
TODO: |
| 475 |
|
|
document DTSAs |
| 476 |
|
|
document tsck |
| 477 |
jmm-guest |
3032 |
document tracked tag |