| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
This commit marks the release of sos-4.5.6.
Signed-off-by: Jake Hunsaker <jacob.r.hunsaker@gmail.com>
|
|
|
|
|
|
|
|
| |
This prevents mb_groups from being collected as it is large, and not
useful for debug.
This also collects more state of the filesystem.
Signed-off-by: Nathaniel Clark <nclark@whamcloud.com>
|
|
|
|
| |
Signed-off-by: Nikhil Kshirsagar <nikhil.kshirsagar@canonical.com>
|
|
|
|
|
|
|
|
| |
This lets files triggers use * and also updates the
ceph plugins to trigger only if folders are non empty
Closes: #3311
Signed-off-by: Nikhil Kshirsagar <nikhil.kshirsagar@canonical.com>
|
|
|
|
|
|
|
|
|
| |
Addresses ConfigParser.readfp removal in 3.12.
Fixes #3308
Bug-Url: https://bugzilla.redhat.com/show_bug.cgi?id=2223526
Signed-off-by: Sandro Bonazzola <sbonazzo@redhat.com>
|
|
|
|
|
|
|
|
| |
Add a common regex for re.MULTILINE so that doesn't need to be defined
and remove the definitions where `(?m)` is being defined
Closes: #3301
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
| |
Collect the microceph data in the ceph plugins.
Signed-off-by: Nikhil Kshirsagar <nikhil.kshirsagar@canonical.com>
|
|
|
|
|
|
|
| |
MAAS no longer uses apache2 for it web hosting, so no longer required
for this plugin
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Adding workdlow to publish the released sos to the latest/candidate channel
when the a new release is created via the tag. This ensures less mannual
intervention for future releases.
Update the snap creation for anyone building manually, and need python3-venv
as a build dependancy.
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some of mac addresses was not properly obfuscated because
some collected data contains mac addresses in format
01:<mac> and parser was not ready for that.
Also added mapper which will obfuscate mac address in case
it is in format with _ instead of : as for example
00_50_56_87_5d_01
instead of
00:50:56:87:5d:01
Format with _ is used for example by vmware plugin.
Resolves: #3302
Signed-off-by: Jan Jansky <jjansky@redhat.com>
|
|
|
|
| |
Signed-off-by: Pablo Acevedo Montserrat <pacevedo@redhat.com>
|
|
|
|
|
|
|
|
| |
Resolves issue SUPDEV-137.
Adds a plugin option for the ssh module, defining whether it will or will not collect .ssh config files per user
Default for new option is True
Signed-off-by: Daniel Zhou <dzhou@redhat.com>
|
|
|
|
|
|
|
|
| |
Add a test that cleaner keeps permissions of a sanitised file
Relevant to: #3292
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
| |
When copying files we applied a substitution in, we must replace just
original file content (shutil.copyfile) and not also its stat data
(shutil.copy).
Resolves: #3292
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Update the maintainer field to be more generic and direct readers to the
upstream repo.
Closes: #3290
Signed-off-by: Jake Hunsaker <jacob.r.hunsaker@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
sos report does not gather the symlink
/etc/grub2-efi.cfg when present.
If a system is legacy, but has this file,
grubby default to using this one first in
some cases. Therefore we need to know if
it's present or not.
Closes: RHBZ#2218563
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch helps simplify the capture of data
from ceph plugins. After the original code was
divided in 8 different plugins, it may be a bit
confusing which plugins to use when running
sos and attempting to capture only ceph data.
By selecting the profile, sos will activate only
these ceph_* plugins that apply to a specific
ceph node.
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
options_tests.py tests -n option which means tens of plugins are
redundantly collected. Testing -o option is effectively the same while
the sosreport run needs a half time only.
The functionality of -n option is tested elsewhere, thus the change does
not shrink test coverage.
Resolves: #3288
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
| |
This commit marks the release of sos-4.5.5.
Signed-off-by: Jake Hunsaker <jacob.r.hunsaker@gmail.com>
|
|
|
|
|
|
|
|
| |
Tested by:
- manually patching file on Debian 11 and 12 (both on sos version 4.0.2)
- sudo ./bin/sos report from this commit (and on python 3.11)
Signed-off-by: Johan Guldmyr <johan@meetiqm.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When calling get_cmd_output_path() to create a directory
via the plugin hpssm, sos threw the following exception:
Traceback (most recent call last):
File "/root/sos/sos/report/__init__.py", line 1224, in setup
plug.setup()
File "/root/sos/sos/report/plugins/hpssm.py", line 67, in setup
logpath = self.get_cmd_output_path()
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/sos/sos/report/plugins/__init__.py", line 2168,
in get_cmd_output_path
os.makedirs(cmd_output_path)
File "<frozen os>", line 225, in makedirs
FileExistsError: [Errno 17] File exists:
'/var/tmp/sos.1gdy83zb/sosreport-localhost-vbwfnpn/sos_commands/hpssm'
This was happening because the directory 'hpssm' was already created.
With this change we avoid any race where we call os.makedirs() after
we have already created the plugin directory.
Closes: RHBZ #2216608
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Two conntrack commands require several kmods - guard collection of the
commands by relevant SoSPredicate.
Resolves: #3279
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Adds a new Prepper to handle keyword preparation. This is slightly
inefficient since we will only realistically need this once, but baking
it into the archive loop does not pose any other problems, and it would
be more fragile to break out a special flow just for keywords.
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Adds a new Prepper for usernames, and removes the bits from the parser
and mapping that otherwise handled the initial preparation of the
mapping. The prepper will source from the same initial files, as well as
from the `--usernames` command line option.
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
| |
Adds a new Prepper to handle feeding relevant files to the mac parser
for initial preparation of the mappings.
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Adds a new Prepper for handling hostname determination for preparing the
mapping and parser.
As part of this new prepper, pass the CLI options to each prepper for use.
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
| |
Adds a new prepper for IP network address sourcing.
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Up until now, the archive abstractions have defined what files `sos
clean` will use to prepare the mappings for obfuscation before entering
the normal obfuscation loop over every file in every archive.
While this is straight forward enough, it is not particularly flexible,
and prevents us from easily using other approaches for preparing the
mappings beyond what is directly obtained via the parsers (which in some
cases need special handling to be prepared at all).
Change this by introducing `SoSPrepper`s which will be used to determine
now only what files to pass to which parsers on an archive-by-archive
basis, but will also allow for manually retrieving items from disaparate
sources within the archive(s) and handing those directly to the
mappings, without the need for those items to first pass the parser
check.
Related: RH: SUPDEV-135
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
| |
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Adding the below paths to collection as they are useful to
troubleshoot some issues.
"/var/lib/maas/http/*.conf"
"/var/lib/maas/*.conf"
"/var/lib/maas/rsyslog"
Resolves #3280
Signed-off-by: Alejandro Santoyo <alejandro.santoyo@canonical.com>
|
|
|
|
|
|
|
|
| |
Originally, it was working by coincidence
Reference: https://github.com/canonical/rockcraft/issues/225#issuecomment-1597025315
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
| |
RedHat distributions can deliver kernel in RPM named either 'kernel' or
'kernel-redhat', in which case we must verify both package names.
Resolves: #3274
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Since https://www.freeipa.org/page/Releases/4.8.7, IPA includes a
tool to send email notifications to users when their passwords are
about to expire.
This logs into: /var/log/ipaepn.log
More information: https://pagure.io/freeipa/issue/3687
Resolves: #3272
Signed-off-by: François Cami <fcami@redhat.com>
|
|
|
|
|
|
|
|
|
|
| |
Widen the capturing of strings for the `ldap_default_authtok`
substitution performed on sssd configuration files to also capture
strings with spaces.
Closes: #3266
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For now neither "/boot/grub2/custom.cfg" nor
"/boot/grub2/user.cfg" are embedded in the sosreport,
which may contain critical information when troubleshooting
boot issues.
Indeed it happens customers create the files badly,
e.g. create a copy of "boot/grub2/grub.cfg" itself,
which ends up having a dead loop in Grub and preventing
the system to boot.
With such information, we would troubleshoot such issues
more rapidly.
Resolves: RHBZ#2213951
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Since https://www.freeipa.org/page/Releases/4.9.8, SID configuration
is instegrated in the base installer. This logs into:
/var/log/ipaserver-enable-sid.log
More information: https://pagure.io/freeipa/issue/8995
Resolves: #3268
Signed-off-by: François Cami <fcami@redhat.com>
|
|
|
|
|
|
|
|
| |
This change simply makes `SosNode` report execution errors using the
standardized UI formatting, instead of printing the log format to
console (but the logged message will still be written to the log file).
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In scenarios where the cluster is unable to delete the temporary project
due to a failure on one of the cluster nodes during the collection, we
would previously fail to delete the temporary project with a nasty
traceback after waiting for 5 minutes.
Be better at this by reducing the wait time (by adding a new `timeout`
parameter to `exec_primary_cmd()`) and more gracefully reporting the
error, and the fact that the user must make manual intervention to
remove the temporary project.
Related: RH SUPDEV-74
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Now that we ae using flags in re.subn, it doesn't like the full
pattern that is being passed through via do_file_private_sub.
So, we separete pattern and the flags, and pass these through to
re.subn
Closes: #3261
Signed-off-by: Arif Ali <arif.ali@canonical.com>
Signed-off-by: David Negreira <david.negreira@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Compute Express Link (CXL) is an industry-supported
Cache-Coherent Interconnect for Processors, Memory
Expansion, and Accelerators.
The CXL Consortium (https://www.computeexpresslink.org/)
is an open industry standard group formed to develop
technical specifications that facilitate breakthrough
performance for emerging usage models while supporting
an open ecosystem for data center accelerators and other
high-speed enhancements.
There are currently three specifications (v1.1, 2.0, and
v3.0) publicly available. CPUs are available from Intel
and AMD that support CXL v1.1. Hardware is emerging that
supports up to CXL v3.0.
This plugin uses open-source utilities from
https://github.com/pmem/ndctl that introduce new and updated
commands to support CXL devices in the Linux Kernel.
Signed-off-by: Steve Scargall <37674041+sscargal@users.noreply.github.com>
|
|
|
|
|
|
|
|
|
|
| |
Remove `check_enabled()` for the maas plugin, as that is now not required
with `SnapPackageManager` now enabled for the `UbuntuPlugin`
Use package manager to detrmine is package is a snap rather than
`snap list`.
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
| |
httpd.conf can refer to files in either conf.d or directly in conf dir,
so let collect all configs by default.
Resolves: #3264
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
| |
adapt the specfile to new fedora packaging guidelines.
Signed-off-by: Sandro Bonazzola <sbonazzo@redhat.com>
|
|
|
|
| |
Signed-off-by: Sandro Bonazzola <sbonazzo@redhat.com>
|
|
|
|
|
|
|
|
|
| |
A MAC address can be followed by characters other than just blank char
or a newline. Let enhance the list accordingly.
Resolves: #3258
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When collecting an sos report from witin a container, and intending to
collect host-level information, symlink collection would previously
collect the container's version of the symlink target due to how they
would resolve within the `Archive.add_file()` flow.
Fix this by simply forcing the follow up collection of the sysroot
location of the symlink target, which would previously be aborted due to
the fact that sos had already collected the container's matching file.
This introduces a certain level of inefficiency by way of doubling the
copy operations, however ensuring that we only make a single copy
operation originally involves a quite fragile set of path mangling that
depends heavily upon how the symlinks are created on the host's
filesystem. For example a symlink using a relative path could be
resolved somewhat easily, but a symlink using an absolute path
introduces a certain amount of indeterminate resolution, which we cannot
allow for the vast majority of use cases that are performed outside of a
container.
Resolves: rhbz#2075720
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
| |
This commit marks the release of sos-4.5.4
Signed-off-by: Jake Hunsaker <jhunsake@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Update the apache plugin to only collect the defaults and update the
openstack bits to collect specifics for their service
move gnocchi.py to openstack_gnocchi.py
Closes: #2525
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
| |
When katello-agent is (still) enabled, "qpid-stat -q" provides a great
overview of pending requests and connected agents.
As there is no katello plugin, we must collect it in foreman directly.
Resolves: #3251
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
| |
Add a new plugin to support collecting data from microk8s on Ubuntu.
Signed-off-by: David Negreira <david.negreira@canonical.com>
|