For an advanced network configuration, the mon endpoints
expected for the ceph public network may be different from
the public IP address that is set on the K8s node. When
host networking is enabled in rook, the IP address of the
node was only allowed previously. Now, the IP address
used for a mon when scheduled on a node can be determined
by setting an annotation on each node where mons can run:
network.rook.io/mon-ip: <ip-address>
This annotation must be set before the mons are created.
Mon endpoints cannot be changed after they are already
running.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
(cherry picked from commit e90ed3b912)
Add IPv6 support for multus validation tool. Also test that IPv6 support
works by specifying one of the NetAttachDefs with an IPv6 address range.
Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
(cherry picked from commit 08a27940ed)
Multus requires that the 'ip' tool be installed in the Rook image.
Today, it is present in the upstream Ceph base image that is used, but
that is expected to change in the future. The Ceph project intends to
switch to CentOS 9 minimal images, and the 'ip' tool will likely be
removed from the base image at that point.
However, Rook CI has occasionally had issues with the current 'dnf
install' command when CentOS repos go down, or when there is otherwise
some problem. Because the package is already installed today, there is
no need to hamstring Rook builds when CentOS is having problems. But we
do want to make sure that Rook builds don't silently succeed in the
eventual future when 'ip' tool is removed from the Ceph image.
For now, replace the 'dnf install' with a check to verify that 'ip' tool
is installed, and add a shorter form of this note as a comment above it
to help Rook maintainers know how to resolve future 'ip' tool removal.
Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
(cherry picked from commit c4e99c1462)
When the clusters reach full, nearfull, or backfill full thresholds
ceph will raise health warnings and stop allowing IO or backfill
depending on the threshold. These settings require special ceph
commands instead of being generic ceph config. Allow these settings
to be set from the CephCluster CR in the spec.storage
section.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
(cherry picked from commit e479b051bc)
This PR adds a `settings` option in the `Module` struct
which can help users to futher customize the `balancer`
module with a differet mode.
Signed-off-by: sp98 <sapillai@redhat.com>
(cherry picked from commit 1ad37ecce9)
Add a note to remove LimitNOFILE setting in containerd service configuration.
This setting causes multiple issues such as slow `ceph status` commands (and failing healthchecks),
various timeouts, and mons failing/timing out.
Upstream containerd already removed the setting. NixOS hasn't incorporated the change.
This note can be removed once https://github.com/NixOS/nixpkgs/pull/313507 makes it to a stable release.
Signed-off-by: Onur Cakmak <occ@occ.me>
(cherry picked from commit d1e139cdb3)
This reduces the safe-to-remove retry interval from 1m to 15s.
Closes#14247
Signed-off-by: Brian Dowling <bjd-dev@simplicity.net>
(cherry picked from commit ef0ed93134)
fixing 'secert' to `secret` and adding workds like
`NotIn, addin` to skip work list.
Signed-off-by: subhamkrai <srai@redhat.com>
(cherry picked from commit 85c25bbc6e)
When IPv6-only network mode is enabled and when mons are configured to
listen only on msgrV2 port, Rook has to add the port to the mon
--public-bind-addr in order to force mons to stop listening on the
msgrV1 port. This can cause problems for IPv6 clusters where adding
':<port>' to the end of an IP addr looks like a valid IPv6 address.
In IPv6 mode, Rook needs to set this to '[<ip>]:<port>' with the
brackets to be a non-ambiguous IPv6 address.
In IPv4 mode, Rook needs to keep its existing behavior. IPv4 mode should
also continue to be the defualt assumption if the network config isn't
specified.
If 'dualStack: true' is specified, Rook is informed that the environment
is dual stack, but Rook doesn't know which IP family to expect from the
pod IP. In this scenario, it is safest for Rook not to apply the port,
since assuming either IPv4 or IPv6 format could be incorrect. In this
scenario, mons may still listen on msgrV1 ports, but this is preferable
to an environment where mons perpetually crash.
Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
(cherry picked from commit 573a47e736)
SVG cleanup job uses the cephSubvolumeGroup env name
from spec.Name. But sometimes it can be empty. In that
case rook should use the CR name.
Signed-off-by: sp98 <sapillai@redhat.com>
(cherry picked from commit 33ce8f0c8e)
after cross-storageclass clone/restore, now we can
clone across different storageclass.
this commit updates the documentation indicating
the same.
Signed-off-by: yati1998 <ypadia@redhat.com>
(cherry picked from commit 945da179e2)
The last pr missed removing the extra comma after the platform
so, instead of ` linux/amd64,linux/arm64,` it should be
` linux/amd64,linux/arm64`.
Signed-off-by: subhamkrai <srai@redhat.com>
(cherry picked from commit ea159ad9b7)
since we moved to ubuntu 22.04, ci push image build is failing
due wrong parameter `--platforms linux/amd64 linux/arm64,` it should
be instead `--platforms linux/amd64, linux/arm64`.
Signed-off-by: subhamkrai <srai@redhat.com>
(cherry picked from commit f89540c4ef)
Since pacific is no longer supported, we can remove pacific specific
code.
Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
(cherry picked from commit f776c0b35e)
Afte pylint was updated to version 3.2.0
a set of new errors appeared. This makes
ci passing by either fixing the problem
or ignoring it in the check
Signed-off-by: NymanRobin <robin.nyman@est.tech>
(cherry picked from commit dcec080468)
LVM-based OSDs on PVCs are legacy and are not well tested.
Therefore, we add status to the cephcluster CR that at least
gives a clue that the OSDs should be replaced.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
(cherry picked from commit 3628e83892)
The CI step that runs markdown lint accidentally got removed as part of
PR updates. Add the CI step back in.
Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
(cherry picked from commit 115e0aa430)
Update the version of Go used for govulncheck to get the latest CVE fix
present in Go v1.22.3.
Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
(cherry picked from commit f21f3f549c)
Travis Nielsen
Blaine Gardner
mergify[bot]
Michael Adam
sp98
Onur C. Cakmak
Brian Dowling
subham rai
subhamkrai
NymanRobin
yati1998
Satoru Takeuchi