This happens during boot or node failover. Overriding the
directory service status forces nsswitch to be configured
correctly for the relevant directory service.

Sometime the command hang

nsswitch.conf generation needs to happen after we set the service
state to HEALTHY. This ensures that ldap is added to the passwd
and group entries.

NAS-113395 / 12.0 / add nvme mapping support to R50B

* raise collectd partition partition threshold to 1.5GB (#7124)


(cherry picked from commit b4686291

)

* Allow larger reporting databases
Co-authored-by: Caleb St. John <30729806+yocalebo@users.noreply.github.com>

This commit adds changes to make sure SED disks are unlocked on HA systems. With the changes introduced, we make sure that we always try to unlock SED disks when backup becomes master node eliminating any race conditions which might come synchronizing if master has already unlocked sed disks or not. Secondly if user changes sed disks password and initiates a failover assuming the system would unlock SED disks, it would not happen as we don't try to unlock sed disks when backup becomes master.

* Use ypcat to get NIS users rather than pwd and grp modules

Ypcat is less error-prone and most importantly will not hang
like NSS lookups when NIS is extravagantly misconfigured
or domain unavailable.

* Remove nis from nsswitch if service becomes faulted

A faulted NIS service may block SSH access to the server
and cause user / group lookups to hang for up to 3 minutes
per request. Hence, it's safer to reconfigure the nsswitch
in case we fail health check.

* Fix default NIS cache entry

* fix logger typo

* fix logger typo
Co-authored-by: Caleb St. John <30729806+yocalebo@users.noreply.github.com>

NAS-112863 / 12.0 / Alert when ARM_INFO is not set for NVDIMM with specification 2.2 or l…

Fixed alert typo

DOCS-2793 / 12.0-U6.1 / Added sleep after umount to make sure it is ready for the next test. (#7824)

(cherry picked from commit f377944090acba7866bd8d64008fcfe2968ebe1c)
Co-authored-by: caleb <yocalebo@gmail.com>

* Bug fix and add more security headers

Add more security headers to the nginx config. Also fixes a misconfig where previously added security headers wouldn't be inherited in the /ui subpath.

* Remove some headers

Remove headers CSP, hardened STS, X-Content-Type-Options and X-XSS-Protection

* Only remove CSP after misunderstanding the comment in the PR
Co-authored-by: Sebastian <sebastian.danielsson@protonmail.com>

* dont start nfs always when bindip is set

* be smarter about logging when service is stopped

* all() to any()

* recreate user homedir on do_update (if needed)

(cherry picked from commit cfa15e6897de7238412feab05af9e3a80d79429b)

* flake8

(cherry picked from commit 0c9b83480dd7f841d11cd36a86c7d24a2e526988)

* address review

(cherry picked from commit 9d503287cf94c42d7c53b213a666cac45fbc96c6)
Co-authored-by: caleb <yocalebo@gmail.com>

Besides being UNAVAIL, they can also be FAULTED, etc.

Incorrect permissions on this directory will prevent winbindd
from starting.

(cherry picked from commit a04410c23b6484884d7a6b61dbf4191cc3a2c1fb)
Co-authored-by: Andrew Walker <awalker@ixsystems.com>

Co-authored-by: caleb <yocalebo@gmail.com>

(cherry picked from commit f08e50771f8e769087ee5692c6226713ef0162e2)
Co-authored-by: themylogin <themylogin@gmail.com>