New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
libxml2: 2.9.9 -> 2.9.10, libxslt: 1.1.33 -> 1.1.34, addressing CVE-2019-18197 #73178
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
some nitpicks, apart from that, LGTM.
# disable test that's problematic with newer pythons: see | ||
# https://mail.gnome.org/archives/xml/2017-August/msg00014.html | ||
preCheck = if pythonSupport && !(python?pythonOlder && python.pythonOlder "3.5") then '' | ||
echo "" > python/tests/tstLastError.py |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can't we just
echo "" > python/tests/tstLastError.py | |
rm python/tests/tstLastError.py |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The makefile isn't clever enough to cope with a missing file and still tries to call it in this case.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why would this failure suddenly occur for older Python versions while it did not in our case before?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Because, as I said, the failure did indeed happen (look at hydra's logs), the makefile just never paid attention to its exit code. This changed in GNOME/libxml2@d188eb9 so it suddenly became a problem for us. I did look into it for a while, but to no avail - looks like it's deep in c module code.
disable python test which was previously failing anyway, but in previous versions it was being ignored
e4f85e5
to
4fec9d5
Compare
@@ -53,6 +53,12 @@ stdenv.mkDerivation rec { | |||
|
|||
enableParallelBuilding = true; | |||
|
|||
# disable test that's problematic with newer pythons: see | |||
# https://mail.gnome.org/archives/xml/2017-August/msg00014.html | |||
preCheck = lib.optionalString (pythonSupport && !(python?pythonOlder && python.pythonOlder "3.5")) '' |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
python?pythonOlder
did you notice an interpreter that did not have this? A common passthru
is used for all Python interpreters.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Meh I was just fitting in with lower down where python?isPy3
is used.
@GrahamcOfBorg build libxml2 libxslt xmlstarlet python37Packages.lxml |
Motivation for this change
https://nvd.nist.gov/vuln/detail/CVE-2019-18197
libxslt
1.1.34 requireslibxml2
2.9.10. To getlibxml2
2.9.10 building successfully I had to disable a couple of the python tests. These tests had actually always been failing on previous versions with python >= 3.5 - the results had just been previously ignored by the makefiles. We're also not the only ones to have had this problem: https://mail.gnome.org/archives/xml/2017-August/msg00014.htmlI also checked the new
libxslt
does indeed include the fixes for the patches I removed in the bump.Anyway, I clearly have not done a full rebuild - I OOM'd my laptop just trying to compute the rebuilds.
Supersedes #72549
Things done
sandbox
innix.conf
on non-NixOS linux)nix-shell -p nix-review --run "nix-review wip"
./result/bin/
)nix path-info -S
before and after)Notify maintainers
cc @