lp:~xavpaice/charm-hw-health

Author: Xav Paice
Author Date: 2022-03-02 04:43:03 UTC

Add arg to charmcraft for lxd containers

If running in lxd, we cannot use charmcraft to make lxds, so we add a
switch to prevent charmcraft using lxd.

Author: Xav Paice
Author Date: 2022-02-02 23:56:13 UTC

Add charmcraft.yaml

Author: Xav Paice
Author Date: 2021-09-27 06:08:48 UTC

Retry check_ipmi_sensor on failure

Alters the cron task which runs check_ipmi_sensor to sleep for a second
then retry, to filter out intermittent issues where some BMCs show
errors then immediately recover.

Author: Xav Paice
Author Date: 2021-07-09 04:27:46 UTC

Correction to config.yaml

Correction to the help text for sel_check_exclusions.
Refactor variable name in tools.py to make the associated code easier to
read.

Author: Xav Paice
Author Date: 2021-05-28 04:04:42 UTC

Update check_ipmi_sensor from upstream

Updates the check_ipmi_sensor submodule, and includes it in the charm
code itself as a file copied whenever we run 'make submodules-update'.
This means that changes to the file get reviewed rather than relying on
the upstream source to review the code safely, given that it runs as
root.

Author: Adam Dyess
Author Date: 2021-04-27 21:57:50 UTC

Fix linting issues around testing cron.d updated

Author: Xav Paice
Author Date: 2021-04-13 01:05:43 UTC

Fix creation of ipmi cache dir if exists

Author: Xav Paice
Author Date: 2021-02-14 20:51:39 UTC

Fix doc text for ipmi_check_options

Adding quotes to the text after --seloptions causes the command to fail,
this fixes the doc to prevent future travellers finding that out.

Author: Xav Paice
Author Date: 2021-01-21 22:55:08 UTC

Update submodules to https url

Some environments do not have ssh access to Github, change the submodule
URL to use https.

Author: Xav Paice
Author Date: 2020-08-12 10:07:56 UTC

*draft* fix and test cron_ilorest.py

if the ilorest command returns health other than OK, the cron_ilorest.py isn't picking that up - in particular, a number of the items already have status 'absent' but that's != 'OK' and the check isn't complaining. I've patched that much, but when in unit tests and using the sample outputs, when I edit other things to be 'someothertextnotOK', it's not picking that up.

Author: Xav Paice
Author Date: 2020-08-12 00:46:46 UTC

Change the ilorest output file name

To make the output of cron_ilorest.py consistent with other scripts, and
compatible with the nrpe check which templates the filename as
toolname.out, this changes the ilorest output name from ilorest.nagios
to ilorest.out.

Author: Xav Paice
Author Date: 2020-02-19 03:04:28 UTC

remove HP from supported vendors

The support for HP Raid adapters has not been coded into the charm yet.
Leaving it in the list of supported vendors results in deployment to HP
hardware failing, even though there are still some useful tools for that
platform. See LP#1855095.

Author: Xav Paice
Author Date: 2020-02-18 20:42:07 UTC

fix typo causing unit tests to fail

Author: Xav Paice
Author Date: 2020-01-27 14:42:35 UTC

Update ipmi cron job to cater for stale lockfile

Author: Xav Paice
Author Date: 2020-01-27 13:11:51 UTC

Update README for tools resource info

Author: Xav Paice
Author Date: 2019-06-24 06:03:48 UTC

Fix for lp1833897 handle separate filesystem for /var

Author: Xav Paice
Author Date: 2019-06-24 06:37:14 UTC

fix unit tests for clear-sel action

Author: Xav Paice
Author Date: 2019-05-23 23:42:39 UTC

Merge branch 'fix_cron' into tip

Author: Xav Paice
Author Date: 2019-05-23 23:33:08 UTC

Fix the clear-sel action

The clear-sel action wasn't working at all, fixed some minor errors to
make it actually run.

Author: Xav Paice
Author Date: 2019-05-23 23:31:54 UTC

Fix cronjob for ipmi

The cronjob for ipmi status was running as the 'nagios' user which
doesn't have the perms needed, and so failed. This changes it to root.

Author: Xav Paice
Author Date: 2019-03-13 00:46:09 UTC

Fix upgrade-charm hook

Fix error where install() was called by the removal of the
hw-health.installed flag, but that flag was being removed by a routine
called when it was set, if the hook name is upgrade-charm (even if the
install() only just set the flag within the same hook run.

Author: Xav Paice
Author Date: 2019-03-12 20:25:19 UTC

Remove sas3ircu when not needed

Previously, sas3ircu was added for all Supermicro machines regardless of
whether the controller is installed, generating errors for machines
without the controller. This now relies on the PCI device being
present.

Closes-bug: 1819733

Author: Xav Paice
Author Date: 2019-02-25 05:49:13 UTC

Add smartctl nrpe checks

Adds a Nagios plugin and nrpe check using it for smartctl checks against
a list of disks that support it. One check for all disks, report
critical if the status is anything other than OK. Adds a config item to
optionally disable the check, should we want to.

Author: Xav Paice
Author Date: 2019-02-25 05:49:13 UTC

Add smartctl nrpe checks

Adds a Nagios plugin and nrpe check using it for smartctl checks against
a list of disks that support it. One check for all disks, report
critical if the status is anything other than OK. Adds a config item to
optionally disable the check, should we want to.

Author: Xav Paice
Author Date: 2019-02-28 02:14:43 UTC

Fix qa tests for Juju actions

Fixes some lint errors plus the unit test for actions.py, and fixes the
environment used for running actions.py to the charm venv.

Author: Xav Paice
Author Date: 2019-02-20 20:14:22 UTC

add ipmi sensor monitor

Adds a tool based on freeipmi command out put to determine the state of
hardware as viewed via IPMI. This should be reasonably generic across
any hardware, and therefore the default is to enable this. There's a
config option added to disable if we wish.

Author: Xav Paice
Author Date: 2019-02-20 20:14:22 UTC

fix unit test for cron_mdadm