Commit Graph
11197 Commits
Author SHA1 Message Date
Alexander Trost 4f97270e41 Merge pull request #2745 from SUSE/mds-operator-code-tidying
ceph file: tidy up file/mds operator code
2019-03-06 22:33:43 +01:00
Blaine Gardner 079cdb56ea ceph file: tidy up file/mds operator code
Separate `file` operator from `mds` tasks logically for easier code
navigation.

Signed-off-by: Blaine Gardner <blaine.gardner@suse.com>
2019-03-06 11:32:58 -07:00
Blaine Gardner 51cc7c2e85 Ceph integration tests: add more verbose logging
Signed-off-by: Blaine Gardner <blaine.gardner@suse.com>
2019-03-05 14:14:09 -07:00
Alexander Trost c430997134 Merge pull request #2762 from galexrt/flexvolume
docs: sorted flexvolume platform paths alphabetical
2019-03-05 16:43:37 +01:00
Alexander Trost ec6cd71fd8 docs: sorted flexvolume platform paths alphabetical
Sorted the flexvolume platform specific paths alphabetically.
Added some spacing around code blocks to prevent issues with certain
Markdown renderers.

Signed-off-by: Alexander Trost <galexrt@googlemail.com>
2019-03-05 07:19:41 +01:00
zhangyue baf27bdd8e cockroachdb: modify cockroachdb operator rbac access
Signed-off-by: zhangyue <zy675793960@yeah.net>
2019-03-05 14:16:10 +08:00
Alexander Trost a8095a888c Merge pull request #2455 from whereisaaron/chart-notes
Add to NOTES.txt the need for manually installed RBAC and links
2019-03-05 07:04:42 +01:00
Travis Nielsen 9fdea5e26b Merge pull request #2757 from travisn/container-image-env
Add CONTAINER_IMAGE env var to the ceph daemon pods
2019-03-04 16:11:14 -07:00
Alexander Trost feb3b0cac9 Merge pull request #2759 from galexrt/fix_2662
Fix indentation for Rancher Flexvolume docs
2019-03-05 00:07:49 +01:00
Alexander Trost 80ecfd710b Fix indentation in Rancher Flexvolume docs
Signed-off-by: Alexander Trost <galexrt@googlemail.com>
2019-03-04 23:52:36 +01:00
travisn 31848e23e7 add the CONTAINER_IMAGE env var to the ceph daemon pods
The mgr orchestrator modules need the container image to run
the ceph image for blinking the lights when a disk is down

Signed-off-by: travisn <tnielsen@redhat.com>
2019-03-04 13:33:41 -07:00
Travis Nielsen 322e493422 Merge pull request #2656 from lbogdan/docs/retain
[ceph docs] Add note of RBD image retention after the corresponding PV is deleted
2019-03-04 11:14:43 -07:00
Alexander Trost 4bb3636fbc Merge pull request #2374 from emonty/magnum-docs
Add configuration note about OpenStack Magnum flexvolume
2019-03-04 19:02:06 +01:00
Bogdan Luca d84b1b8c82 Add note of RBD image retention after the corresponding PV is deleted.
Signed-off-by: Bogdan Luca <luca.bogdan@gmail.com>
2019-03-04 17:34:48 +02:00
Travis Nielsen a145f4dcb7 Merge pull request #2703 from travisn/osds-per-device-with-filter
Clean up provisioning of OSDs on devices
2019-03-04 07:58:56 -07:00
Monty Taylor e87a457489 doc: Add note about OpenStack Magnum flexvolume
OpenStack Magnum uses Atomic Host as its default deployment. The Atomic
docs for rook point to the openshift docs, which are a bit complex to
follow if one is not actually using OpenShift. For magnum, the path
/var/lib/kubelet/volumeplugins is the correct path, so it's easier to
just point to that directly rather than sending users to OpenShift docs.

Signed-off-by: Monty Taylor <mordred@inaugust.com>
2019-03-04 14:35:42 +00:00
Alexander Trost 42c71768c0 Merge pull request #2428 from k911/improve-registry-example
docs: Improve HA docker registry example
2019-03-04 14:26:05 +01:00
Alexander Trost 409a5129e2 Merge pull request #2614 from chaosaffe/master
update image for coreos reboot daemonsets
2019-03-04 14:08:51 +01:00
Alexander Trost 5bc314be26 Merge pull request #2753 from nakabonne/fix-service-name
Fix service name in doc
2019-03-04 14:04:16 +01:00
Ryo Nakao 4bc2315c66 Fix service name in doc
Signed-off-by: Ryo Nakao <nakabonne@gmail.com>
2019-03-03 22:12:11 +09:00
Travis Nielsen 581766eb91 Merge pull request #2704 from travisn/tests-no-check-substr
Remove unreliable string checks from integration tests
2019-03-02 13:57:53 -07:00
travisn b24ebc8168 tests: remove unnecessary string returns from kubectl calls
Various kubectl helper methods return strings from calls to
create, delete, or apply resources. There is no need for this.
It is sufficient and complete to check the err from these calls to determine failure.

Signed-off-by: travisn <tnielsen@redhat.com>
2019-03-01 16:16:41 -07:00
travisn 6ca573bb34 tests: remove unreliable substring checking
The integration tests check for the return of substrings from calls to kubectl
in a number of places. On occasion these return something slightly different
and cause an unnecessary failure in the integration tests.

Signed-off-by: travisn <tnielsen@redhat.com>
2019-03-01 16:16:41 -07:00
Travis Nielsen 816d698655 Merge pull request #2742 from noahdesu/dmcrypt
ceph: share host ipc namespace with provision pod
2019-03-01 13:21:05 -07:00
Noah Watkins 6597531ed0 ceph: share host ipc namespace with provision pod
this is needed as cryptsetup in the pod and udev on the host coordinate
via a semaphore that must be visible in both contexts.

note that additionally the host needs to have dmcrypt compiled with udev
synchronization, which at least one minikube version does not have.

fixes: #923

Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
2019-03-01 10:50:19 -08:00
Travis Nielsen eaf6b97b59 Merge pull request #2702 from SUSE/mds-all-in-operator
ceph mds: configure completely from operator
2019-03-01 06:59:15 -07:00
Madhu Rajanna 901143349b Add cephfs csi doc
updated ceph-csi-drivers.md document to
test cephfs csi driver.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-03-01 12:57:43 +05:30
Alexander Trost 8e1bc5bd97 Merge pull request #2737 from galexrt/helm_2735
Added ROOK_HOSTPATH_REQUIRES_PRIVILEGED as value to Helm Chart
2019-02-28 22:09:08 +01:00
Alexander Trost dcc5d2b5ea Added ROOK_HOSTPATH_REQUIRES_PRIVILEGED as value to Helm Chart
Signed-off-by: Alexander Trost <galexrt@googlemail.com>
2019-02-28 20:42:51 +01:00
Blaine Gardner 4eb4fb7aa6 ceph mds: configure completely from operator
Configure the Ceph mds daemon completely from the operator a la the
recent changes to the Ceph mon and mgr operators.

Create the mds deployments first and then
create the keyring secrets for them with their owner reference as the
corresponding deployment. This will mean that the secrets do not need to
be micromanaged. When the deployment is deleted, the secret is also
deleted. This has not been necessary for the mons or the manager since
the mons share a keyring with a lifespan of the cluster, as does the
mgr, which currently has single-mgr support only.

Signed-off-by: Blaine Gardner <blaine.gardner@suse.com>
2019-02-28 12:31:43 -07:00
Travis Nielsen 8dcc3576e9 Merge pull request #2734 from JoeWrightss/patch-2
Fix misspell in doc
2019-02-28 09:48:54 -07:00
zhoulin xie 36c35b6968 Fix misspell in doc
Signed-off-by: zhoulin xie <zhoulin.xie@daocloud.io>
2019-03-01 00:26:55 +08:00
Travis Nielsen dc26762e24 Merge pull request #2719 from ZYecho/nfs-test
test: add nfs with multi export integration test
2019-02-27 21:27:41 -07:00
zhangyue 87b4f45fd9 test: add nfs with multi export integration test
Signed-off-by: zhangyue <zy675793960@yeah.net>
2019-02-28 09:02:52 +08:00
Travis Nielsen e4d42eece1 Merge pull request #2728 from jtlayton/wip-orch-init
Retry orchestrator initialization if command is unrecognized
2019-02-27 16:22:41 -07:00
Jeff Layton c0910dab2c ceph: retry "ceph orchestrator set backend rook" if command returns EINVAL
Use similar method as that used by the dashboard initialization code.

Fixes: https://github.com/rook/rook/issues/2617
Signed-off-by: Jeff Layton <jlayton@redhat.com>
2019-02-27 10:55:24 -05:00
Jeff Layton 56abc00b78 ceph: fix invalidArgErrorCode definition
We should probably not hardcode this value here, and instead use
whatever the syscall module thinks this is.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
2019-02-27 10:53:10 -05:00
Alexander Trost 7b872026de Merge pull request #2709 from SergeyKaydalov/edgefs-vdev-management
Edgefs VDEV management document
2019-02-27 06:29:55 +01:00
travisn bdc3cf8146 osd: fix the device filter and improve device provisioning reliability
All devices detected by the discovery pod were being passed to the OSD provisioning pod
thus not always honoring the desired device list that should be provisioned.
Now the provisioning pod will be given the desired state from the crd,
then apply that state depending on the actual devices detected.
Also added a helper to ensure OSDsPerDevice is always valid.

Signed-off-by: travisn <tnielsen@redhat.com>
2019-02-26 16:11:25 -07:00
Travis Nielsen 5ef1557b8b Merge pull request #2706 from gabibbo97/dashboard-readme
Add missing documentation for dashboard options in Cluster CRD
2019-02-26 15:32:52 -07:00
Sergey Kaydalov 4fcfad1dee Edgefs VDEV management document
Signed-off-by: Sergey Kaydalov <kayserg@gmail.com>
2019-02-26 20:30:42 +02:00
Travis Nielsen be2882dc39 Merge pull request #2711 from Madhu-1/fix-csi-dep
Fix ceph-csi deployment document
2019-02-26 11:18:32 -07:00
John Mulligan c3e863d1a9 build: support specifying custom 'docker' command in run script
Allow the default 'docker' command to be overridden by an environment
variable in the "build/run" shell script and the supporting library
file.

Signed-off-by: John Mulligan <jmulligan@redhat.com>
2019-02-25 18:49:11 -05:00
John Mulligan 49f3b68dea build: support specifying custom sha-256 hash command
Create a makefile (and shell) variable that can be used to invoke a
different command line, such as 'sha256sum' instead of the default
'shasum -a 256'. The shasum tool is left as the default but, at least on
fedora, this comes from a perl package and it may be that sha256sum,
which comes from coreutils would allow building on a more minimal
install.

Signed-off-by: John Mulligan <jmulligan@redhat.com>
2019-02-25 18:49:11 -05:00
John Mulligan 0fd73d82d9 images: support specifying custom 'docker' command
In the image makefiles use the "DOCKERCMD" variable in place of
calls to docker command line tool. This allows custom paths
to docker or other docker-compatible tools to be used for builds.

Signed-off-by: John Mulligan <jmulligan@redhat.com>
2019-02-25 18:49:11 -05:00
Travis Nielsen c616ba5f48 Merge pull request #2708 from leseb/nautilus-beast
rgw: change default frontend on nautilus
2019-02-25 15:35:27 -07:00
Madhu Rajanna de4da4fbad Fix markdown formatting issue
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-02-25 15:48:17 +05:30
Madhu Rajanna d2d699efb0 update namespace creation
rook-ceph-system namespace is
created as part of the operator,update
document to check namespace is present,
if ns is not present create it.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-02-25 15:48:12 +05:30
Sébastien Han 03587352d5 rgw: change default frontend on nautilus
As per: https://github.com/ceph/ceph/pull/26599, Beast is now the
default fronted for rados gateway.
Newly created cluster as of Nautilus will use it by default.

Resolves: https://github.com/rook/rook/issues/2707
Signed-off-by: Sébastien Han <seb@redhat.com>
2019-02-24 17:22:32 +01:00
Giacomo Longo d273525451 Add missing documentation for dashboard options in Cluster CRD
As can be seen in pkg/apis/ceph.rook.io/v1/types.go the Cluster CRD documentation is missing details about the dashboard configuration

Signed-off-by: Giacomo Longo <gabibbo97@gmail.com>
2019-02-24 16:04:04 +01:00