Commit Graph
7527 Commits
Author SHA1 Message Date
Travis Nielsen 0d8fd9d8a4 Merge pull request #8607 from leseb/refact
ceph: refactor operator initialization sequence and add more controllers (csi, lib-bucket-prov, discover daemon, flex)
2021-09-17 11:38:35 -06:00
Sébastien Han b89730d895 ceph: refactor operator initialization sequence
This commit is a large refactor on how the operator starts, stops and
how it starts various sub-components such as the ceph-csi driver. It
also refines the way we cancel orchestrations. We don't use breakpoints
anymore but send our self a SIGUP to reload our controller runtime
manager.
The reload will happen under different circonstances like:

* a new adminission controller secret is created/deleted/changed
* a CephCluster CR is edited

As mentioned earlier, the csi driver now has its own controller, just
like flex. It reacts to change in the operator config map for particular
ROOK_CSI_ fields.

A second new controller for the operator's general config has been
created, it manages:

* the logging level
* the ceph CLI command timeout
* the discovery daemon

The operator reacts much more rapidly to cancellation events by stopping
the manager's context and reloading it.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-17 16:57:12 +02:00
Sébastien Han 78ef8768c2 Merge pull request #8738 from leseb/yaml-parse
ceph: remove unnecessary package
2021-09-17 16:48:09 +02:00
Sébastien Han 68a4bc2d1b ceph: remove unnecessary package
We don't need to use github.com/ghodss/yaml since
"k8s.io/apimachinery/pkg/util/yaml" provides the same functionality and
we already import it.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-17 16:10:47 +02:00
Sébastien Han 32984f7297 Merge pull request #8746 from leseb/force-16.2.5
ci: force a particular ceph version
2021-09-17 16:09:12 +02:00
Travis Nielsen b3f5ec6dcf Merge pull request #8732 from leseb/cephfs-mirror-doc
doc: fix cephfs-mirror documentation
2021-09-17 07:54:41 -06:00
Sébastien Han ae291afb2f ci: force a particular ceph version
Let's force v16.2.5 since the CI is broken with 16.2.6. This gives us
time to continue to merge work and work on fixing deployments with
16.2.6 in parallel.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-17 15:28:48 +02:00
Sébastien Han fd28497cb4 docs: fix cephfs-mirror documentation
The steps to configure the peers are detailed in the CephFilesystem
section. Only the CephFilesystem is holding the peer configuration, not
the CephFilesystemMirror which only controls the bootstrap of the
daemon.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-17 09:50:13 +02:00
Travis Nielsen e2e3fc83fc Merge pull request #8673 from leseb/stale-bot
bot: change issue and pr stale days
2021-09-16 08:31:31 -06:00
Blaine Gardner 79e8cf0eea Merge pull request #8715 from BlaineEXE/prune-unused-crd-permission-in-clusterrole
ceph: remove unused crd permission from roles
2021-09-16 08:04:58 -06:00
Sébastien Han 875aec74cc bot: bump action to v4
Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-16 10:04:56 +02:00
Sébastien Han 796abd7805 bot: change issue and pr stale days
Let's be a bit more aggressive to ensure a consistent amount of
in-progress PRs and issues.
So now PR will go stale after 30 days and issues after 60 days.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-16 10:04:46 +02:00
Sébastien Han bcae365cb9 Merge pull request #8698 from sp98/fix-pdb-reconcile
ceph: reconcile osd pdb if allowed disruption is 0
2021-09-15 10:47:54 +02:00
Sébastien Han 1632bfaa5a Merge pull request #8711 from subhamkrai/dry-run-check
ci: dry-run is deprecated and replaced with --dry-run=client
2021-09-15 10:06:48 +02:00
Sébastien Han c4a6473fc8 Merge pull request #8435 from BlaineEXE/upgrade-doc-add-mirroring-changes
docs: ceph: add peer spec migration to upgrade doc
2021-09-15 09:26:12 +02:00
Travis Nielsen 7f3eebe578 Merge pull request #8695 from travisn/master-docs-latest
docs: Publish master docs to latest path
2021-09-14 14:27:56 -06:00
Blaine Gardner 7c511328e4 docs: update rbd mirror docs for block pool config
Remove legacy documentation for configuring RBD mirroring. While we
still support legacy mirroring configs, we want to encourage new users
to use the CephBlockPool configuration for mirroring.

Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
2021-09-14 13:39:28 -06:00
Blaine Gardner 83c05324ce ceph: remove unused crd permission from roles
Remove unused ability to modify CRDs from Rook cluster roles. Rook (as
an operator) should not have the ability to modify CRDs used by the
components it controls.

Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
2021-09-14 12:52:06 -06:00
subhamkrai 038031ddbd ci: dry-run is deprecated and replaced with --dry-run=client
--dry-run is deprecated and replaced with --dry-run=client

Signed-off-by: subhamkrai <srai@redhat.com>
2021-09-14 16:13:27 +05:30
Santosh Pillai 7480f6ba62 ceph: reconcile osd pdb if allowed disruption is 0
Rook checks for down OSDs by checking the `ReadyReplicas` count
in the OSD deployement. When an OSD pod goes into CBLO due to
disk failure, there is a delay before this `ReadyReplicas` count
becomes 0. The deplay is very small but may result in rook missing
OSD down event. As a result no blocking PDBs will be created and
only default PDB with `AllowedDisruptions` count as 0 is available.
This PR tries to solve this. The OSD pdb reconciler will be
reconciled again if `AllowedDisruptions` count in the main
PDB is 0.

Signed-off-by: Santosh Pillai <sapillai@redhat.com>
2021-09-14 15:56:59 +05:30
Travis Nielsen 235d5a384b Merge pull request #8693 from travisn/cleanup-docs
docs: Refactor documentation for ceph provider
2021-09-13 11:25:22 -06:00
Travis Nielsen ac6a6d60ba Merge pull request #8699 from subhamkrai/update-script
build: use pkg/operator/ceph/csi/spec.go for csi images
2021-09-13 10:42:27 -06:00
Travis Nielsen 6b7062eda5 docs: refactor documentation for ceph provider
With the nfs and cassandra providers moving to their own repo
we can simplify the docs a bit. Some obsolete documentation
is also removed.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2021-09-13 10:40:47 -06:00
Travis Nielsen 661ed66f53 Merge pull request #8694 from travisn/commit-prefixes
bot: Add more ceph commit prefixes
2021-09-13 10:36:15 -06:00
Travis Nielsen a32b11c572 bot: add more ceph commit prefixes
With the storage providers moved to their own repo,
we add more possible prefixes for commits to the
ceph operator, including the ceph daemons and csi
driver.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2021-09-13 09:29:34 -06:00
Travis Nielsen 99811e52e1 Merge pull request #8267 from lyind/master
ceph: fix lvm osd db device check
2021-09-13 09:07:13 -06:00
subhamkrai 284dbc0a0e build: use pkg/operator/ceph/csi/spec.go for csi images
it has happened more than once that the manifests do not
reflect the images so now reading from `pkg/operator/ceph/csi/spec.go`
as this what code uses.

Closes: https://github.com/rook/rook/issues/8687
Signed-off-by: subhamkrai <srai@redhat.com>
2021-09-13 17:16:52 +05:30
Jonas Zeiger 0e72a7c2bf ceph: fix lvm osd db device check
Signed-off-by: Jonas Zeiger <jonas.zeiger@talpidae.net>
2021-09-13 12:19:16 +02:00
Sébastien Han 1df2e9372e Merge pull request #8685 from BlaineEXE/emphaize-unit-tests-in-contributor-guides
docs: emphasize unit tests in development guide
2021-09-13 10:03:44 +02:00
Travis Nielsen fee7e82416 docs: publish master docs to latest path
Rename the master docs to latest. This will be easier
to understand for users who are new to the project.
The url will also contain latest, which will be more
clear to reference instead of the master path. This
also follows the pattern of the ceph docs site.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2021-09-10 17:55:14 -06:00
Blaine Gardner dc3c84ef27 docs: emphasize unit tests in development guide
Emphasize unit tests in the development guide, and provide guidelines to
help contributors create good unit tests.

Also clarify that Rook's CI will run integration tests and that users
aren't expected to run integration tests locally.

Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
2021-09-10 14:30:32 -06:00
Blaine Gardner 94eb4519a0 Merge pull request #8596 from subhamkrai/offline-installation
ci: image list for offline installation
2021-09-09 13:53:49 -06:00
Travis Nielsen 533cb6c07c Merge pull request #8674 from parth-gr/cluster-crd
docs: correct indentation for topologyKey
2021-09-09 12:46:11 -06:00
Travis Nielsen 0cd63aaf29 Merge pull request #8665 from travisn/roadmap-1.8
docs: Update ceph operator roadmap for v1.8
2021-09-09 12:34:24 -06:00
Travis Nielsen c5a55e6a88 docs: storage provider status mentions new repos
With the storage providers all moving to separate repos, we clarify
the status and location of each storage provider. Also, no need to
mention the deprecated storage providers anymore.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2021-09-09 12:23:48 -06:00
Travis Nielsen d93397ca55 docs: update ceph operator roadmap for v1.8
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2021-09-09 12:23:47 -06:00
Blaine Gardner eed91efdc2 Merge pull request #8667 from thotz/enhanceusernits
ceph: addressing nits from #8211
2021-09-09 12:10:51 -06:00
Blaine Gardner 91ec1ac4db docs: ceph: add peer spec migration to upgrade doc
Add a section to the upgrade doc instructing users to (and how to)
migrate `CephRBDMirror` `peers` spec to individual `CephBlockPools`.
Adjust the pending release notes to refer to the upgrade section now,
and clean up a few references in related docs to make sure users don't
miss important documentation.

Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
2021-09-09 11:46:29 -06:00
Travis Nielsen d193a23f7e Merge pull request #8677 from leseb/fix-arrray
ci: fix pod list
2021-09-09 11:05:22 -06:00
Sébastien Han 3c1c3409b1 Merge pull request #8668 from Muyan0828/fix/osd-full-alert-query
ceph: fix CephOSDCriticallyFulla and CephOSDNearFull Alert Query
2021-09-09 19:01:06 +02:00
Travis Nielsen af63a36647 Merge pull request #8666 from hall/master
ceph: add networking.k8s.io/v1 Ingress chart compatability
2021-09-09 10:59:58 -06:00
Sébastien Han 0e961c4bdc ci: stop build 3 times
This time for real.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-09 17:51:03 +02:00
Sébastien Han 73c340ded7 ci: fix pod list
We should not use .items[0].metadata.name if the array length is 0. This
is the case when nothing has been initialized yet. Instead, we should
use .items[*].metadata.name, the wildcard ensures to always return 0
even if nothing is present yet.

Fixes: https://github.com/rook/rook/issues/8676
Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-09 17:24:56 +02:00
Sébastien Han 031be4be19 Merge pull request #8675 from subhamkrai/ok-continue
ceph: modify the log info when ok to continue fails
2021-09-09 17:22:13 +02:00
subhamkraiandZeaone 0657804491 ceph: modify the log info when ok to continue fails
correct typo in logging, it was showing `ok-to-stop`
instead of `ok-to-continue` when 'continueUpgradeAfterChecksEvenIfNotHealthy' is true

Co-Authored-by: Zeaone <zeaone@ZeaonedeMacBook-Pro.local>
Signed-off-by: subhamkrai <srai@redhat.com>
2021-09-09 19:42:06 +05:30
parth-grandFabio Nitto fdae56dd3b docs: correct indentation for topologyKey
fixed the incorrect indentation for topologyKey
in ceph-cluster-crd doc and storage-class-device-set doc

Co-authored-by: Fabio Nitto
Signed-off-by: parth-gr <paarora@redhat.com>
2021-09-09 19:33:50 +05:30
chengli f2c9793fca ceph: fix CephOSDCriticallyFulla and CephOSDNearFull Alert Query
Add hostname to CephOSDCriticallyFull and CephOSDNearFull rules's expr

Signed-off-by: chengli <chengli@alauda.io>
2021-09-09 21:58:14 +08:00
Travis Nielsen 3e14dbbb85 Merge pull request #8670 from leseb/print-output
ceph: print the output on errors
2021-09-09 07:32:21 -06:00
Bryton Hall 5e4ed30bc9 ceph: add networking.k8s.io/v1 Ingress chart compatability
Kubernetes cluster versions 1.22+ must use the v1 Ingress version.

Signed-off-by: Bryton Hall <email@bryton.io>
2021-09-09 09:11:49 -04:00
Sébastien Han b7f55362b2 ceph: print the output on errors
Sometimes the error does not tell much, so as `exit status 1` and
printing the output along returning the error is useful.

For instance, I saw a job failing with no osd and the prepare job had
those lines:

```
exec: Running command: lsblk /dev/sdb1 --bytes --nodeps --pairs ....
inventory: skipping device "sdb1". exit status 1
```

We need to understand more about the lsblk issue.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-09-09 12:42:01 +02:00