| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
We only support Ruby 2.7+ and this takes less memory
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Gather CPU info from lscpu and create cpu/lscpu attribute namespace
This adds a cpu/lscpu node data name space containing information from the lscpu
command (if it exists). This augments and standardizes the information you can
gather from /proc/cpuinfo which can be an issue on non-x86 architectures such as
ppc64le and aarch64. Namely, when trying to determine whether or not a node is a
guest or not.
In addition, this properly fixes cpu counts on s390x nodes and uses lscpu data
for determining cpu/total, cpu/real and cpu/cores if lscpu exists.
* Fixes for detecting KVM guests on certain platforms and architectures
On some architectures (i.e. ppc64le), dmidecode does not exist nor does the
proper information in /proc/cpuinfo exist to detect KVM guests. To workaround
that, we can use lscpu data to easily determine this as a fallback.
In addition on CentOS 6, /sys/devices/virtual/misc/kvm does not exist so this
works around that problem as well.
Signed-off-by: Lance Albertson <lance@osuosl.org>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
| |
Closes #1653
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Fall back to v4-only getaddrinfo if we get socket errors
|
|/
|
|
|
|
|
|
|
|
|
| |
Handle the case where we have either v4- or v6-specific configuration
or network issues that causes gethostaddr() to fail
Note this was found due to a specific case: On OSX - and possibly others
(freebsd), if you have a linklocal address for your hostname in
/etc/hosts then `getaddrinfo` for your hostname will raise an exception.
Signed-off-by: Phil Dibowitz <phil@ipom.com>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
|
|
|
| |
Fixes some bugs
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
lpar_no and wpar_no in AIX Virtualizatin plugin should be Integers
|
|/
|
|
|
|
|
| |
These don't make sense as strings. We should make them ints. This is a
breaking change.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
|
|
|
|
|
|
| |
On newer Linux systems which use systemd, they should include /etc/os-release
[1] which includes a lot of platform information that can be useful downstream.
[1] https://www.freedesktop.org/software/systemd/man/os-release.html
Signed-off-by: Lance Albertson <lance@osuosl.org>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\ |
|
| |
| |
| |
| | |
Signed-off-by: Bhavesh Davda <bhavesh.davda@oracle.com>
|
| |
| |
| |
| |
| |
| | |
Replace invocation of libc.so.6 with "ldd --version" instead
Signed-off-by: Bhavesh Davda <bhavesh.davda@oracle.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
It's not safe to directly execute "libc.so.6" to determine glibc version.
This can lead to segfaults when other shared libraries are preloaded using
LD_PRELOAD or /etc/ld.so.preload due to circular relocation dependencies.
Replace this version detection with "ldd --version" as "ldd" is a core part
of glibc and reports the same version.
Signed-off-by: Bhavesh Davda <bhavesh.davda@oracle.com>
|
| |
| |
| |
| | |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\ \
| |/
|/| |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
|
|
|
| |
Populate node with offload features information queried via ethtool -k.
Signed-off-by: Zeeshan Lokhandwala <zeeshanl@fb.com>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
|
|
|
| |
Nothing super scientific here, but it's about 10% faster to match if the data is hit sooner. Stick the super obscure stuff last.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
- All supported releases have an /etc/os-release file so nuke the legacy detection
- The platform ID in the os-release file is XCP-ng in 7 and xenserver in 8 (this is a bug IMO) so update the XCP match
- Downcase platform IDs. It seems like everyone makes these lowercase, but XCP-ng. There may be others and ohai platforms need to be lowercase
Signed-off-by: Tim Smith <tsmith@chef.io>
|
| |
| |
| |
| | |
Obvious fix; these changes are the result of automation not creative thinking.
|
| |
| |
| |
| | |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\ \
| | |
| | | |
Remove support for Pidora which is discontinued
|
| | |
| | |
| | |
| | |
| | |
| | | |
Pidora is Fedora 18 and hasn't been updated since 2014.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|\ \ \
| |/ /
|/| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/ /
| |
| |
| |
| |
| |
| | |
This distro is now discontinued with the last release having shipped in
2019
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
| |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/ |
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\
| |
| | |
Update rubocop-performance to 1.10.1 and revert optimization
|
| |
| |
| |
| |
| |
| |
| | |
This has some nice bugfixes. Revert the previous optimization that broke
things
Signed-off-by: Tim Smith <tsmith@chef.io>
|
| |
| |
| |
| | |
Obvious fix; these changes are the result of automation not creative thinking.
|
|\ \
| |/
| | |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
|
|
|
| |
Looks like this comparison is 2x faster on average according to docs
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Obvious fix; these changes are the result of automation not creative thinking.
|