| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
| |
Continuation of #3530.
Signed-off-by: Ponnuvel Palaniyappan <pponnuvel@gmail.com>
|
|
|
|
|
|
|
| |
Prevously the lsblk would return single FSTYPE multiple times as I
forgot to add -s to it. Quick fix for this.
Signed-off-by: Lukas Herbolt <lukas@herbolt.com>
|
|
|
|
|
|
|
|
|
| |
Closes: #3552
Signed-off-by: Pablo Fernández Rodríguez <pafernan@redhat.com>
use f-strings
Signed-off-by: Pablo Fernández Rodríguez <pafernan@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Add new section fstype section under hardware devices which lists
devices by filesystem, based on lsbl -nrpo output. Devices with no
filesystem are placed into unknown section and Ext2/3/4 are put into
section called ext4.
Signed-off-by: Lukas Herbolt <lukas@herbolt.com>
|
|
|
|
|
|
|
|
|
| |
The sapconf command is not distributed in RHEL 8 and 9,
so the code is never executed, and its safe to remove.
Related: RH: RHEL-27830 and RHEL-27831
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Ensure that we format the package versions to the pep440 standard.
This is necessary as we are using `parse_version()` from the `packaging`
package to verify that the sos collector node version is equal or bigger
than the versions we are collecting from the nodes. If we pass the wrong
format to `parse_version()`, we are unable to do the comparison and
unable to run `sos report` on those nodes.
This addresses the issue of the Ubuntu packaging version naming with `+`
and `~` signals as well as generic versioning schemes.
Resolves: #3544
Signed-off-by: David Negreira <david.negreira@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If no API url is specified, it will use
localhost always. In some cases, specially
when using Ansible, we'll want to specify
another API URL.
Related: RH: RHEL-24523
Closes: #3517
Co-authored-by: Alberto Losada Grande <alberto.losada@redhat.com>
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
| |
Continuation of #3530.
Signed-off-by: Ponnuvel Palaniyappan <pponnuvel@gmail.com>
|
|
|
|
| |
Signed-off-by: Ponnuvel Palaniyappan <pponnuvel@gmail.com>
|
|
|
|
|
|
|
|
|
|
| |
Collect useful info by executing some more commands in nfs-utils
package:
- nfsconf for the current configuration.
- mountstats for various NFS client statistics.
- rpcctl for SunRPC connection information.
Signed-off-by: Seiichi Ikarashi <s.ikarashi@fujitsu.com>
|
|
|
|
| |
Signed-off-by: Costa Shulyupin <costa.shul@redhat.com>
|
|
|
|
|
|
| |
Related: Discussion #3513
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
| |
Signed-off-by: David Negreira <david.negreira@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
As part of the project's effort to modernize on f-strings, this commit
replaces all usage of .format() with f-string equivalents. The vast
majority of these are in-place syntax changes, but a limited number
change the order of list items or some formatting tricks in order to
appease PEP8.
None of these conversions change the underlying logic of the flows they
appear in.
Related: #3472 discussion
Signed-off-by: Jake Hunsaker <jacob.r.hunsaker@gmail.com>
|
|
|
|
| |
Signed-off-by: Jake Hunsaker <jacob.r.hunsaker@gmail.com>
|
|
|
|
|
|
|
|
|
| |
Some parsers should skip strings <=3 length and some mappings should
obfuscate just full words.
Relevant: #3403
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
As obfuscation of substrings of words is redundant and can lead to
leaking the secure string from the obfuscated word (e.g.
enobfuscateduser1ment), cleaner should match whole words only, in
relevant parsers/mappings.
Related: RHEL-2399
Resolves: #3403
Closes: #3496
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Obfuscation of entries like "map" or "po" is redundant and rather causes
user confusion and longer cleaner run.
Stop obfuscating those strings in relevant mappings (only).
Relevant: #3403
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Running various flatpak commands can start gvfs mount, which is
undesirable outcome of sos. Calling the commands with
GVFS_REMOTE_VOLUME_MONITOR_IGNORE=1 env.variable prevents that
behaviour.
Relevant: RHEL-14328
Closes: #3528
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
| |
Resolves: SET-463
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
| |
Signed-off-by: David Negreira <david.negreira@canonical.com>
|
|
|
|
| |
Signed-off-by: David Negreira <david.negreira@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently, we are not supporting microk8s with the
same options as provided by kubernetes. This PR
extends the UbuntuKubernetes to get information
from microk8s.
Following the review, it also extends microk8s plugin
to add the credentials file to the copy spec.
Closes: #3500
Co-authored-by: Arif Ali <arif.ali@canonical.com>
Signed-off-by: Pedro Guimaraes <pedro.guimaraes@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Ceph release 18.2.0 added a couple of new commands 'counter dump'
and 'counter schema' and will supersede 'perf dump' and 'perf schema'
which are deprecated now. For sosreport, we obviously need both
commands to collect data on existing deployments.
At some point in the future, we can remove 'perf dump' and
'perf schema' from the list of commands for Ceph plugins.
Signed-off-by: Ponnuvel Palaniyappan <pponnuvel@gmail.com>
|
|
|
|
|
|
|
|
|
|
| |
If running tests or sos without the snap or deb installed, then this
will fail with TypeError as the sosreport package will not be
installed. This will ensure we catch this, and use the default path.
Resolves: SET-338
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Adding a plugint to collect:
- infinidat storage logs
- some info that could be useful for debugging purposes
Resolves: SET-454
Signed-off-by: Alejandro Santoyo <alejandro.santoyo@canonical.com>
|
|
|
|
|
|
|
|
|
| |
Similar to how we do in openstack_nova, collect commands from as many
of the openstack projects.
Resolves: SET-337
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The list of modules was too broad, and depending on the different attributes
configured the list of modules to be loaded could also be different.
Liaising with the LXD team suggested to check for the `snap.lxd.daemon`
service rather then the modules to collect `lxd.buginfo`. This is due to the
fact that LXD is socket activated, and running any lxd based command could
start the service and hence load the modules. Gating this with either the
`snap.lxd.daemon` for snap or `lxd` service for debs respectively should
prevent this from happening.
This also no obfuscates the certificate that is retrieved from `lxd.buginfo`
Resolves: SET-377
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Closes: #2241
Resolves: #2257
Resolves: SET-99
Resolves: LP #1891296
Co-authored-by: Arif Ali <arif.ali@canonical.com>
Signed-off-by: Adam R Bell <adam.bell@canonical.com>
Signed-off-by: Arif Ali <arif.ali@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
| |
The option --config was moved to --kubeconfig in
OCP 4.7.
Related: RH: RHEL-24402
Co-authored-by: Alberto Losada Grande <alberto.losada@redhat.com>
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The authentication method for RHEL uploads to the
customer portal is changing in 2024 to Device Auth
tokens, from user/password basic authorization.
To accomplish this, one new class is created:
DeviceAuth (deviceauth.py), that takes care of
managing OID token authentication.
Closes: RH: SUPDEV-63
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
| |
Signed-off-by: ajaymohite25 <ajaymohite244@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It's part of a Ceph mgr module `prometheus` and the example output is as
follows. That would be helpful to understand the timeline and the
history of issues in Ceph.
[ceph healthcheck history ls]
```
Healthcheck Name First Seen (UTC) Last seen (UTC) Count Active
MON_DISK_LOW 2024/02/08 12:53:46 2024/02/08 12:53:46 1 Yes
MON_DOWN 2024/02/08 12:58:46 2024/02/08 12:58:46 1 No
2 health check(s) listed
```
Signed-off-by: Nobuto Murata <nobuto.murata@canonical.com>
|
|
|
|
|
|
|
|
|
| |
Without SoSPredicate, the plugin was starting
the fwupd service.
Related: RH: RHEL-24342
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
| |
Capture machine-id and lastupload files.
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
'wbinfo -t/--check-secret' is one of the basic commands
to check if the trust secret for domain is still valid.
'net ads testjoin' is used to check if participation
in a domain is still valid.
Related: #RHEL-23665
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
| |
Signed-off-by: ajaymohite25 <ajaymohite244@gmail.com>
|
|
|
|
|
|
|
|
| |
There are some configuration files which could affect NetworkManager's
behaviors. "NetworkManager --print-config" is useful to verify the
current configuration at a glance.
Signed-off-by: Hiroyuki Yasuhara <h.yasuhara@fujitsu.com>
|
|
|
|
|
|
|
|
|
|
| |
The file is libmount private (and optional)
but may provide important information as it stores
userspace-specific mount data
Related: #3497
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
| |
Signed-off-by: Adam Vest <adam.vest@canonical.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Adds a new preset for `aap_eda`, or Ansible Automation Platform Event
Driven Controller.
This preset will set the `containers_common` plugin to collect output
for the `eda` user.
Related: #3486
Signed-off-by: Jake Hunsaker <jacob.r.hunsaker@gmail.com>
|
|
|
|
|
|
|
|
|
|
| |
When using own preset, loggers verbosity is set even before they are
configured. In such a case, we must guard accessing them properly.
Resolves: #3491
Relevant: RHEL-22395
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Adding the file 'aap_eda.py' for the sos report collects
the files used for troubleshooting issues at
Ansible Automation Platform Event Driven Controller
Related: RH AAP-12898
Resolves: #3486
Signed-off-by: Rudnei Bertol Junior <rudnei@redhat.com>
|
|
|
|
|
|
|
|
| |
RHEL Atomic Host is RHEL7 only, and we since
we don't ship sos 4.* in RHEL 7, we can drop
this policy and the associated sos plugin.
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
| |
Add passwords stored in /sys/kernel/config/target/ to
the list of forbidden paths.
Related: RH RHEL-19056
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|
|
|
|
|
|
|
|
|
| |
No URI shall contain a secret or token so there is no need to scrub that
file.
Resolves: #3477
Signed-off-by: Pavel Moravec <pmoravec@redhat.com>
|
|
|
|
|
|
|
|
|
|
| |
containerd uses CNI to configure the container's network.
The CNI uses different configuration files that by default are
installed in `/etc/cni/net.d`, these files are important to
understand how the network is configured.
Signed-off-by: Antonio Ojea <aojea@google.com>
|
|
|
|
| |
Signed-off-by: Trevor Benson <trevor.benson@gmail.com>
|
|
|
|
| |
Signed-off-by: Trevor Benson <trevor.benson@gmail.com>
|
|
|
|
|
|
|
|
|
| |
Add DebianPlugin to the list of imports and
the class UbuntuKubernetes.
Related: #3484
Signed-off-by: Jose Castillo <jcastillo@redhat.com>
|