Fedora has issued an advisory on July 31: https://lists.fedoraproject.org/archives/list/package-announce@lists.fedoraproject.org/thread/UFJ4HNVK37HLZUQTQNVGRX53R37JIFL2/ The issue is fixed upstream in 0.67.0. Mageia 5 and Mageia 6 are also affected.
Whiteboard: (none) => MGA6TOO
Assigning to all packagers collectively, since there is no registered maintainer for this package. CC'ing some committers.
CC: (none) => geiger.david68210, marja11, nicolas.salguero, smelrorAssignee: bugsquad => pkg-bugs
Suggested advisory: ======================== The updated packages fix a security vulnerability: Poppler through 0.62 contains an out of bounds read vulnerability due to an incorrect memory access that is not mapped in its memory space, as demonstrated by pdfunite. This can result in memory corruption and denial of service. This may be exploitable when a victim opens a specially crafted PDF file. (CVE-2018-13988) References: http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2018-13988 ======================== Updated packages in core/updates_testing: ======================== poppler-0.52.0-3.8.mga6 lib(64)poppler66-0.52.0-3.8.mga6 lib(64)poppler-devel-0.52.0-3.8.mga6 lib(64)poppler-cpp0-0.52.0-3.8.mga6 lib(64)poppler-qt4-devel-0.52.0-3.8.mga6 lib(64)poppler-qt5-devel-0.52.0-3.8.mga6 lib(64)poppler-qt4_4-0.52.0-3.8.mga6 lib(64)poppler-qt5_1-0.52.0-3.8.mga6 lib(64)poppler-glib8-0.52.0-3.8.mga6 lib(64)poppler-gir0.18-0.52.0-3.8.mga6 lib(64)poppler-glib-devel-0.52.0-3.8.mga6 lib(64)poppler-cpp-devel-0.52.0-3.8.mga6 from SRPMS: poppler-0.52.0-3.8.mga6.src.rpm
Whiteboard: MGA6TOO => (none)Source RPM: poppler-0.63.0-3.mga7.src.rpm => poppler-0.52.0-3.7.mga6.src.rpmCVE: (none) => CVE-2018-13988Version: Cauldron => 6Status: NEW => ASSIGNEDAssignee: pkg-bugs => qa-bugs
Mageia 6, x86_64 CVE-2018-13988 The vulnerability is awaiting analysis, therefore no reproducer available. Updated the 12 packages. Followed leads in earlier bug reports like https://bugs.mageia.org/show_bug.cgi?id=23183 Ran some of the utilities on PDF files in my docs/books directory. $ pdffonts PythonCrashCourse.pdf Syntax Warning: Invalid Font Weight name type encoding emb sub uni object ID ------------------------------------ ----------------- ---------------- --- --- --- --------- MFSDXQ+Futura-Book CID Type 0C Identity-H yes yes yes 5031 0 MMKCRQ+DogmaBold Type 1C WinAnsi yes yes no 5024 0 [...] $ pdfimages -png pragpub-2013-04.pdf test $ ls test* test-000.png test-007.png test-014.png test-021.png test-028.png test-001.png test-008.png test-015.png test-022.png test-029.png test-002.png test-009.png test-016.png test-023.png test-030.png test-003.png test-010.png test-017.png test-024.png test-004.png test-011.png test-018.png test-025.png test-005.png test-012.png test-019.png test-026.png test-006.png test-013.png test-020.png test-027.png $ eom test*.png All images looked correct. This generated a large number of image files pr<number>.{png|jpg} and two HTML files. $ ll pr*.html -rw-r--r-- 1 lcl lcl 7790514 Aug 27 19:10 pr-html.html -rw-r--r-- 1 lcl lcl 20218 Aug 27 19:10 prs.html $ firefox file:///home/lcl/docs/books/prs.html This opened a web page with the document outline which looked like a hyperlink list of contents but clicking the links showed an index number in the address bar. However $ firefox file:///home/lcl/docs/books/pr-html.html Brought up an okular style web page which allowed scrolling through the pages. Clicking on hyperlinks in the text raised the document outline page which did not help much. Looks like the hyperlink functionality has not been implemented very well. External site links work perfectly though. Be that as it may, the web document contains all the information from the PDF and the images display inline. $ pdfseparate -f 8 -l 15 StatisticsDoneWrong.pdf stats_%d $ ll stats* -rw-r--r-- 1 lcl lcl 3660564 Aug 27 19:54 stats_10 -rw-r--r-- 1 lcl lcl 3660722 Aug 27 19:54 stats_11 -rw-r--r-- 1 lcl lcl 3660562 Aug 27 19:54 stats_12 -rw-r--r-- 1 lcl lcl 3660838 Aug 27 19:54 stats_13 -rw-r--r-- 1 lcl lcl 3660869 Aug 27 19:54 stats_14 -rw-r--r-- 1 lcl lcl 3660829 Aug 27 19:54 stats_15 -rw-r--r-- 1 lcl lcl 3660562 Aug 27 19:54 stats_8 -rw-r--r-- 1 lcl lcl 3660560 Aug 27 19:54 stats_9 All display as separate pages of the original, pages 8 to 15. $ pdftops stats_11 stats11.ps $ gs stats11.ps This displayed the Brief Contents page from the original pdf file. $ pdftoppm stats_11 abc $ ls abc* abc-1.ppm $ display abc-1.ppm That displayed Brief Contents also. $ pdftocairo -jpeg stats_14 stats14 $ ls *.jpg stats14-1.jpg The image displayed fine. As noted before, several more utilities are missing, like pdfunite. Good for 64-bits.
CC: (none) => tarazed25Whiteboard: (none) => MGA6-64-OK
Len's tests look sufficient to me. Validating. Suggested advisory in Comment 2.
Keywords: (none) => validated_updateCC: (none) => andrewsfarm, sysadmin-bugs
CC: (none) => tmbKeywords: (none) => advisory
An update for this issue has been pushed to the Mageia Updates repository. https://advisories.mageia.org/MGASA-2018-0358.html
Status: ASSIGNED => RESOLVEDResolution: (none) => FIXED