Summary
This update for python-numpy fixes the following issue:
Security issue fixed:
- CVE-2019-6446: Set allow_pickle to false by default to restrict loading untrusted content (bsc#1122208).
With this update we decrease the possibility of allowing remote attackers to execute arbitrary code by
misusing numpy.load(). A warning during runtime will show-up when the allow_pickle is not explicitly set.
NOTE: By applying this update the behavior of python-numpy changes, which might break your application.
In order to get the old behaviour back, you have to explicitly set `allow_pickle` to True. Be aware
that this should only be done for trusted input, as loading untrusted input might lead to arbitrary code
execution.
Fixed packages 8
| Release | Package | Fixed in |
|---|---|---|
| SUSE · SLES 12 SP3 Linux Enterprise Desktop 12 SP3 | python-numpy | 1.8.0-5.8.1 |
| SUSE · SLES 12 SP4 Linux Enterprise Desktop 12 SP4 | python-numpy | 1.8.0-5.8.1 |
| SUSE · SLES 12 SP3 Linux Enterprise Server 12 SP3 | python-numpy | 1.8.0-5.8.1 |
| SUSE · SLES 12 SP4 Linux Enterprise Server 12 SP4 | python-numpy | 1.8.0-5.8.1 |
| SUSE · SLES for SAP 12 SP3 Linux Enterprise Server for SAP Applications 12 SP3 | python-numpy | 1.8.0-5.8.1 |
| SUSE · SLES for SAP 12 SP4 Linux Enterprise Server for SAP Applications 12 SP4 | python-numpy | 1.8.0-5.8.1 |
| SUSE · SLES 12 SP3 Linux Enterprise Software Development Kit 12 SP3 | python-numpy | 1.8.0-5.8.1 |
| SUSE · SLES 12 SP4 Linux Enterprise Software Development Kit 12 SP4 | python-numpy | 1.8.0-5.8.1 |
Vulnerabilities 1
-
Critical 9.8
Numpy Deserialization of Untrusted Data
An issue was discovered in NumPy before 1.16.3. It uses the pickle Python module unsafely, which allows remote attackers to execute arbitrary code via a crafted serialized object, as demonstrated by a numpy.load call. NOTE: third parties dispute this issue be…