Commit Graph
4472 Commits
Author SHA1 Message Date
Travis Nielsen f79bc3d6a0 noobaa: remove design docs due to inactivity
The NooBaa operator was proposed for addition to Rook, but
the team went a different direction and created an independent
operator. Removing the stale docs.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-17 14:30:56 -06:00
Travis Nielsen 2708eb457e ozone: move design doc to a subfolder
For consistency with the design docs, move the Ozone
design docs to a subfolder.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-17 14:30:56 -06:00
Travis Nielsen 8c2c257b34 Merge pull request #4984 from moricho/gomodules
Move to Go Modules
2020-03-17 13:26:11 -06:00
moricho c11d3d1cc9 codegen: modify codegen.sh to work correctly for go modules
This modify codegen.sh to work correctly.
`make codegen` temporarily download code-generator into vendor directory and use it
to refer to the local version of rook.

Signed-off-by: moricho <ikeda.morito@gmail.com>
2020-03-18 00:37:23 +09:00
Sébastien Han 4ae313f024 Merge pull request #4992 from travisn/doc-clarifications
Restructure prerequisites and clarify ceph docs
2020-03-17 09:07:12 +01:00
moricho 7d0215d820 ci: switch to use go.mod.check target
This switches to use `go.mod.check` target instead of `go.vendor.check` target.

Signed-off-by: moricho <ikeda.morito@gmail.com>
2020-03-17 16:48:43 +09:00
moricho 437a0b1d9b go: versionup sig-storage-lib-external-provisioner/controller
version up sig-storage-lib-external-provisioner/controller to v4.1.0 and
modify related files

Signed-off-by: moricho <ikeda.morito@gmail.com>
2020-03-17 16:47:55 +09:00
moricho 8907e220ea apis: regenerate files & modify k8sutil tests
This regenerate files with code-generator/deepcopy-gen and modify the related tests.

Signed-off-by: moricho <ikeda.morito@gmail.com>
2020-03-17 16:47:53 +09:00
moricho 2beb731a7e go: move to gomodules
This switches Rook to use `go mod` instead of `dep` for
dependencies management.

Signed-off-by: moricho <ikeda.morito@gmail.com>
2020-03-17 16:47:53 +09:00
Travis Nielsen cd4c77fa7d docs: update ceph toolbox to match example yaml
The ceph toolbox no longer needs mounts to /dev and /udev. The
docs had not previously been updated with this change.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-16 15:15:44 -06:00
Travis Nielsen a251959061 docs: restructure prerequisites and clarify ceph docs
The docs were in need of some restructuring wrt prerequisites. They
were mingled with general k8s and Ceph requirements. For clarity,
we need to separate the ceph prereqs from general k8s requirements.

There is now only a single prereqs page. All other topics should be
found under other sections, such as specific to Ceph. Having subtopics
under the general prereqs caused users to frequently not even notice
the main prereqs topic.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-16 15:15:44 -06:00
Travis Nielsen 2c70e10bd6 docs: add cluster-on-pvc and remove cluster-minimal examples
The cluster-minimal.yaml example is more confusing than helpful. Most users
seem to think that minimal includes a fully working cluster with OSDs.

The cluster-on-pvc.yaml example is frequently used and was missing from the
examples doc.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-16 15:13:56 -06:00
Travis Nielsen 575a73daa9 ceph: updates from a documentation review
The documentation for Ceph configuration needed some
clarification and cleanup from options that are no longer needed
or apply.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-16 15:13:55 -06:00
Travis NielsenandRan Rubin 735292e8e3 ceph: flex driver is not always a prerequisite
This configuration is required when using the flex CSI driver (required for K8s 1.12
or earlier). Going forward for K8s 1.13 and newer it is recommended to use the CSI
RBD driver, making Flexvolume configuration redundant.

Co-authored-by: Ran Rubin <ranrubin@gmail.com>
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-16 15:13:55 -06:00
Travis Nielsen 8a19c73036 ceph: clarify docs for erasure coding
Setting up the topology for the CRUSH map was outdated in the EC doc.
Now a link to the node labels is provided for the configuration.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-16 15:13:52 -06:00
Travis Nielsen 9c5c78cf01 Merge pull request #5025 from jmolmo/blinking_lights
Ceph: Added rook manager module rights for read pod logs
2020-03-16 13:57:13 -06:00
Travis Nielsen 017a560db3 Merge pull request #5021 from samkulkarni20/yb-mem-limits
YugabyteDB: Add resource limits to YugabyteDB pods
2020-03-16 11:12:07 -06:00
Juan Miguel Olmo Martínez c87a01bb8c Ceph: Added rook manager module rights for read pod logs
This is needed to allow the ceph manager rook module to read the output
of the <lsmcli> command used to switch disk identification light on/off on
physical disk devices

[test ceph]

Signed-off-by: Juan Miguel Olmo Martínez <jolmomar@redhat.com>
2020-03-16 18:03:23 +01:00
Sébastien Han a459a2e220 Merge pull request #5030 from leseb/no-full-err
ceph: remove dead code
2020-03-16 15:19:49 +01:00
Travis Nielsen 2724791d6c Merge pull request #5019 from HaichaoLii/arm64
test: Kubeadm-install add support for arm64
2020-03-16 07:45:39 -06:00
Sébastien Han f44583b2b9 ceph: remove dead code
As part of the transition to support Ceph release **as of** Nautilus, we
left over that portion of code.
We don't need it anymore.

Signed-off-by: Sébastien Han <seb@redhat.com>
2020-03-16 14:01:24 +01:00
Sébastien Han 316c114597 Merge pull request #5018 from travisn/mgr-delete-pod-access
Grant access to the Ceph mgr to restart pods
2020-03-16 11:45:13 +01:00
Haichao Li 24027b969e test: Kubeadm-install add support for arm64
Only kubectl for amd64 is currently available.This modification adds
arm64 version of kubectl that can successfully deploy kubernetes on
arm64 machine.

Signed-off-by: Haichao Li <haichao.li@arm.com>
2020-03-16 16:15:59 +08:00
Sameer Kulkarni 909770a346 YugabyteDB: Add resource limits to YugabyteDB pods
Current YugabyteDB operator code creates Master and TServer pods without any resource requests/limits (specifically CPU and memory).
This causes the operator to run into soft/hard memory limit issue. The fix adds recommended resource requests and limits as defaults to the pods it creates.

Closes: https://github.com/yugabyte/yugabyte-db/issues/3884
Signed-off-by: Sameer Kulkarni <samkulkarni20@gmail.com>
2020-03-14 12:23:14 +05:30
Travis Nielsen 7de388ce92 ceph: grant access to the mgr to delete pods
The Ceph mgr modules need access to restart the daemon pods
when the configuration changes for the daemons.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-13 17:08:27 -06:00
Travis Nielsen 24f3649149 Merge pull request #5024 from travisn/keepalive-label
Add keepalive label to exempt issues from the stalebot
2020-03-13 10:31:23 -06:00
Travis Nielsen ab705fe548 stalebot: add keepalive label to exempt issues
Some issues should not be closed by the stale bot, whether because an
issue is taking a long time to design, implement, or if we need an
owner for the issue, but we know we need the feature or fix.

The keepalive label will cause the bot to skip closing stale issues.

This label should be used sparingly

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-13 10:20:54 -06:00
Travis Nielsen 783507588e Merge pull request #5023 from leseb/allow-size-1
ceph: allow setting pool size 1 on octopus
2020-03-13 10:17:41 -06:00
Sébastien Han 99f6402563 ceph: allow setting pool size 1 on octopus
Left from https://github.com/rook/rook/pull/4895
Also more cleanup on the ceph.conf since we config is in the mon store.

Signed-off-by: Sébastien Han <seb@redhat.com>
2020-03-13 15:48:08 +01:00
Sébastien Han 8a14f4072a Merge pull request #4994 from CloudAndHeat/feature/improve-placement-docs
Clarify placement configuration for OSDs
2020-03-13 10:53:23 +01:00
Sébastien Han 5648555809 Merge pull request #5010 from leseb/crash-rook-cluster-label
ceph: add missing 'rook_cluster' label to crash
2020-03-12 15:47:12 +01:00
Sébastien Han 1edc3adc58 ceph: add missing 'rook_cluster' label to crash
The crash deployment was missing the 'rook_cluster' label to indicate
which ceph cluster it is part of.

Signed-off-by: Sébastien Han <seb@redhat.com>
2020-03-12 11:43:20 +01:00
Sébastien Han 9198663462 Merge pull request #5005 from travisn/yugabytedb-test-image
Yugabyte tests should use image built by rook
2020-03-12 08:45:56 +01:00
Travis Nielsen dabf0b6618 yugabyte: tests should use image built by rook
The integration tests for yugabyte were pointing to an image
outside of the rook repo. Since that image changed unexpectedly
it broke the rook integration tests. The tests need to use
the image built by rook.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-11 17:44:05 -06:00
Travis Nielsen 077a9ae60b Merge pull request #4960 from leseb/bump-to-14.2.8
Ceph v14.2.8 is causing failure with OSDs on PVCs
2020-03-11 14:18:04 -06:00
Sébastien Han 723671f58c ceph: run osd initContainer in priv mode
To access a special block device the container must run privileged.

Signed-off-by: Sébastien Han <seb@redhat.com>
2020-03-11 19:33:14 +01:00
Sébastien Han 14cb46ae39 ceph: bump to 14.2.8
Ceph 14.2.8 just got released so let's use it.

Signed-off-by: Sébastien Han <seb@redhat.com>
2020-03-11 19:32:48 +01:00
Sébastien Han a14339af2e Merge pull request #5001 from leseb/osd-udev
ceph: osd re-add udev bindmount
2020-03-11 19:32:18 +01:00
Sébastien Han 619f301eb7 ceph: osd re-add udev bindmount
The OSD needs to talk to the udev socket to get information about a device.
like serial, vendor etc. So that it can populate the output of:
`ceph osd metadata $n` correctly.

Signed-off-by: Sébastien Han <seb@redhat.com>
2020-03-11 18:35:47 +01:00
Sébastien Han 04979f8c60 Merge pull request #4909 from umangachapagain/config-watcher
Ceph: adds watch to rook-ceph-operator-config ConfigMap
2020-03-11 09:34:16 +01:00
Blaine Gardner b626de1758 Merge pull request #4989 from travisn/topology-label
Properly lookup the topology.kubernetes.io label for OSD topology
2020-03-10 15:40:29 -04:00
Travis Nielsen fd89d0dfb3 ceph: remove dead code for finding valid nodes
There are no longer callers of the helper to find nodes in a list
of Rook nodes.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-10 12:06:34 -06:00
Travis Nielsen ba07a63033 ceph: properly lookup the K8s topology labels for OSDs
The OSDs pick up on several topology labels for CRUSH hierarchy.
The GA label topology.kubernetes.io was partially implemented, but
not picked up by the OSDs. Now the OSDs will pick up both the topology
labels from pre-1.17 such as failure-domain.beta.kubernetes.io/zone
and topology.kubernetes.io/zone.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-10 12:06:34 -06:00
Jonas Schäfer d9d9216f2f docs: clarify placement configuration for OSDs
OSD placement is not controlled with the Cluster CRD placement
key, but with a storage class specific placement key. The
cluster-wide placement rules are not merged into the OSD placement
rules.

This change reflects this fact in the documentation.

[skip ci]

Closes: https://github.com/rook/rook/issues/4882
Signed-off-by: Jonas Schäfer <jonas.schaefer@cloudandheat.com>
2020-03-10 11:20:29 +01:00
Umanga Chapagain c54c0d16cf Ceph: adds watch to rook-ceph-operator-config ConfigMap
watches the rook-ceph-operator-config ConfigMap and updates
CSI driver

Signed-off-by: Umanga Chapagain <chapagainumanga@gmail.com>
2020-03-10 12:46:54 +05:30
Umanga Chapagain 06ba93e629 Ceph: refactor GetOperatorSetting for reuse
Signed-off-by: Umanga Chapagain <chapagainumanga@gmail.com>
2020-03-09 21:41:56 +05:30
Travis Nielsen ce18843885 Merge pull request #4978 from travisn/manifest-tool-update
Update manifest tool version to v1.0.0
2020-03-06 12:26:17 -07:00
Travis Nielsen de06c5c2ec build: update manifest tool version to v1.0.0
The manifest tool is needed for pushing images to dockerhub.
The tool has been failing the past two days due to running
an old version.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2020-03-06 11:59:27 -07:00
Sébastien Han c580e6b496 Merge pull request #4977 from leseb/orchestrator-mod-name
ceph: orchestrator cli name change
2020-03-06 18:04:47 +01:00
Sébastien Han 8b0fdf31a9 ceph: orchestrator cli name change
With https://github.com/ceph/ceph/pull/33131, the CLI has been rename
'orch' instead of 'orchestrator'.
Octopus only.

Signed-off-by: Sébastien Han <seb@redhat.com>
2020-03-06 17:01:16 +01:00