Commit Graph
11197 Commits
Author SHA1 Message Date
Blaine Gardner 91480a717d Merge pull request #4060 from leseb/error-wrapf
ceph: use "github.com/pkg/errors"
2019-12-09 11:46:38 -07:00
Blaine Gardner af822018e3 ceph: osd: filestore: fix filename too long error
Add osd-max-object-{name,namespace}-len flags to filestore OSDs to avoid
`ERROR: osd init failed: (36) File name too long` errors.

Signed-off-by: Blaine Gardner <blaine.gardner@suse.com>
2019-12-09 10:41:54 -07:00
Sébastien Han 80ba8ab2ad ceph: change cep-nfs pid location
Let's put the PID under the directory owned by ganesha.

Closes: https://github.com/rook/rook/issues/3525
Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-09 18:21:37 +01:00
Travis Nielsen 0e87101807 Merge pull request #4450 from rohan47/nfs_image_update
WIP: NFS: updated the current nfs version to a newer version
2019-12-09 10:18:36 -07:00
Sébastien Han ac74216e13 Merge pull request #4454 from leseb/skio-ceph-dm
ceph: reject ceph dm devices
2019-12-09 18:06:38 +01:00
Sébastien Han 5ce2ed220e ceph: use "github.com/pkg/errors"
We now use the error package.
Kubernetes errors have been renamed kerrors since they are lower than
'errors'.

Closes: https://github.com/rook/rook/issues/4054
Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-09 16:58:32 +01:00
Sébastien Han 04ad1636c8 ceph: reject ceph dm devices
Since https://github.com/rook/rook/pull/4219, lv devices are now
presented to ceph-volume when preparing the device.
So on this initial run, this won't fail because the dm hasn't been
created yet but if an orchestration is re-triggered, the prepare pod
ensures idempotency with the ceph-volume batch command.
Unfortunately, c-v seems to have a bug where it doesn't read the dm to
detect whether or not they are ceph members already.
Ceph bug: https://tracker.ceph.com/issues/43209

Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-09 16:17:26 +01:00
Travis Nielsen 6a66e010ad Merge pull request #4445 from travisn/continue-on-node-missing
Continue orchestration on missing node for osd
2019-12-09 08:05:36 -07:00
Travis Nielsen a8f399af88 Merge pull request #3669 from mateuszlos/cephclient
Create Client CRD
2019-12-09 08:05:19 -07:00
Mateusz Los 0daec4a7d6 ceph: change rgw client prefix
add rgw prefix to objectstore user

Signed-off-by: Mateusz Los <los.mateusz@gmail.com>
2019-12-09 12:49:24 +01:00
Mateusz Los 50f3d29cf4 ceph: client controller tests and docs
ceph client CRD refactoring
Add info about client crd to documentation
Add tests for client controller CRD

Signed-off-by: Mateusz Los <los.mateusz@gmail.com>
2019-12-09 12:48:44 +01:00
Sébastien Han 95ddb4de62 Merge pull request #4451 from dspearson/spelling-fix
Correct spelling error.
2019-12-09 12:06:10 +01:00
Dominic Pearson f71a600e0c examples: correct spelling error.
Signed-off-by: Dominic Pearson <dsp@technoanimal.net>
2019-12-09 11:43:54 +01:00
Sébastien Han 96c1eeae23 Merge pull request #4446 from travisn/empty-device-filter
ceph: comment out empty deviceFilter property
2019-12-09 09:59:13 +01:00
rohan47 5a1e911458 NFS: updated the current nfs version to a newer version
Signed-off-by: rohan47 <rohgupta@redhat.com>
2019-12-09 12:19:33 +05:30
Mateusz Gozdek 65fd8ece2a ceph: Create initial client controller code
Create initial client controller code

Signed-off-by: Mateusz Los <los.mateusz@gmail.com>
2019-12-08 16:47:57 +01:00
Alex Punnen ae633f4828 docs: add 'git clone' step to quickstart guides
Add documentation to clone the correct release branch of the repo
        to reduce confusion from pulling the wrong branch.

Signed-off-by: Alex Punnen <alexcpn@gmail.com>
2019-12-07 11:19:29 +05:30
Dmitry Yusupov b12cc5ab85 Merge pull request #4448 from codytrey/master
removed dead link to edgefs-helm-operator
2019-12-06 20:42:51 -08:00
Cody Belcher 47a706288d Docs: edgefs-quickstart
removed dead link to edgefs-helm-operator.md from edgefs-quickstart.md

Signed-off-by: Cody Belcher <cody.t.belcher@gmail.com>
2019-12-06 22:10:14 -06:00
Travis Nielsen 1247e7add8 ceph: comment out empty deviceFilter property
The deviceFilter property needs to have a value rather
than be completely empty as it was in the settings.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2019-12-06 16:45:41 -07:00
Travis Nielsen 618028d2cd ceph: continue orchestration on missing node for osd
When a node has been removed, cordoned, or is otherwise
missing from the filters that rook uses to evaluate deploying
osds on nodes, the orchestration should continue instead
of failing. Otherwise, the orchestration will stay stuck
in a retry loop if any nodes are cordoned or removed in
the middle of an orchestration.

Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
2019-12-06 15:48:19 -07:00
Sébastien Han c7e52741f6 Merge pull request #4443 from leseb/new-branch-mergify
mergify: add backport-release-1.2 rules
2019-12-06 18:49:56 +01:00
Sébastien Han 77af8c6712 mergify: add backport-release-1.2 rules
Add a new rule for the upcoming release-1.2 branch.

Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-06 18:45:19 +01:00
Sergey Kaydalov b6b95d44ed edgefs: payloads on an S3 service
Added support for data chunk forwarding to an external S3 service

Signed-off-by: Sergey Kaydalov <kayserg@gmail.com>
2019-12-06 19:16:43 +02:00
Sébastien Han 555b205ebb Merge pull request #4439 from ottoyiu/patch-1
docs: fix typo on ceph bucket claim description
2019-12-06 11:36:11 +01:00
Otto Yiu 239028c010 docs: fix typo on ceph bucket claim description
Signed-off-by: Otto Yiu <otto@live.ca>
2019-12-05 22:57:14 -08:00
Travis Nielsen 6c543788ba Merge pull request #4219 from cybozu-go/support-lv-backed-pv
ceph: support OSD on PVC backed by LV
2019-12-05 14:26:23 -07:00
Travis Nielsen c75e64b954 Merge pull request #3998 from ruslanloman/master
Allow to run rook operator on host network
2019-12-05 10:51:06 -07:00
Sébastien Han 74a33a0299 Merge pull request #4153 from leseb/rook-multus-design
design: rook multus design
2019-12-05 18:28:02 +01:00
dulltzandSatoru Takeuchi dfe45ac6b0 ceph: support OSD on PVC backed by LV
"OSD on PVC" doesn't work for PV backed by LV. Fixing this problem
by the following changes.

- Rook accepts LVM disk type.
- If a LV-backed device is passed, Rook/Ceph invokes
  "ceph-volume lvm prepare" with "--data vg/lv"
  instead of "--data /path/to/device".
- If a LV-backed device is passed, Rook/Ceph suppresses
  activation/deactivation of VG that owns this LV.

Fixes: https://github.com/rook/rook/issues/4185
Signed-off-by: dulltz <isrgnoe@gmail.com>
Co-authored-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
2019-12-05 17:26:27 +00:00
Sébastien Han 3ebae50f84 design: add multus for ceph
Add the design doc for integrating multus with the ceph backend.

Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-05 18:16:37 +01:00
Sébastien Han e108f87b06 Merge pull request #4386 from leseb/location-prepare
ceph: do not restart the osds when upgrading the operator
2019-12-05 16:59:33 +01:00
Travis Nielsen 0a352f64d9 Merge pull request #4285 from cybozu-go/use-mapper-name
ceph: support device selection for OSDs by device path name patterns
2019-12-05 07:54:07 -07:00
Sébastien Han 470798c570 ceph: osd stop using rook binary and tini
This patch removes the need of using both tini and rook binary to run
the main osd container.
By removing this, we know can upgrade the operator image without
restarting the OSD since the osd pod does not use the rook image
anymore.
This patch only works when not running OSD on PVC since they are certain
actions that we cannot achieve with an init container.
A new init container has been introduced to do the activation of the
OSD, the activate files are placed in a empty dir mount and shared with
the main osd process.
Now the main osd process runs the ceph-osd daemon binary directly.

This currently has a few limitation:

* only works with bluestore OSDs
* only works when not running OSD on PVC

Closes: https://github.com/rook/rook/issues/4363
Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-05 14:39:51 +01:00
ruslanloman 59e17dcdf5 ceph: Allow to run rook operator on host network
There is use case when need to run rook operator on host network since
  ceph mon is not accessible from kubernetes pod network.

Signed-off-by: ruslanloman <ruslanloman@gmail.com>
2019-12-05 13:25:08 +02:00
Kazuhito MATSUDA 40d868ebd7 ceph: support device selection for OSDs by device path name patterns
Add devicePathFilter to Storage Selection Settings which enables
device selection with a regular expression for device path names.

Fixes: https://github.com/rook/rook/issues/4275
Signed-off-by: Kazuhito MATSUDA <kazuto.jinnai@gmail.com>
2019-12-05 07:02:44 +00:00
Travis Nielsen ab69d75cf2 Merge pull request #4354 from egafford/cleanup-csi-resources
ceph: clean up CSI resources on cluster deletion
2019-12-04 20:00:39 -07:00
Travis Nielsen 5b8eb3c18e Merge pull request #4396 from leseb/ganesha-pid
ceph: openshift ganesha fixes
2019-12-04 17:40:09 -07:00
Travis Nielsen b89911ec7f Merge pull request #4318 from rajatsingh25aug/toolbox
Ceph: Toolbox does not need to run privileged
2019-12-04 16:57:49 -07:00
Travis Nielsen 2c6a6a87d6 Merge pull request #3282 from elek/ozone-design
ozone: design doc for ozone operator
2019-12-04 15:36:28 -07:00
Sébastien Han 4a7258eb0d ceph: osd pass crush location from prepare pod
We shouldn't need to compute the location from the main OSD pod, since
the prepare pod runs on the same node we can populate the crush location
information from that pod and pass it along to the main osd container.

This is an initial commit that aims to eventually remove the need of
using the rook binary when running the main osd container.

Closes: https://github.com/rook/rook/issues/4362
Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-04 23:03:02 +01:00
RAJAT SINGH ca3ab62b9d Ceph: Modified toolbox.yaml so that it need not to run privileged
Removed unnecessary voumeMounts and privileged flag from toolbox.yaml file.

Signed-off-by: RAJAT SINGH <rajasing@redhat.com>
2019-12-05 01:57:50 +05:30
RAJAT SINGH 519b36286c File: Added file direct-mount.yaml
Changed all the occurances of toolbox.yaml to direct-mount.yaml in file direct-tools.md

Signed-off-by: RAJAT SINGH <rajasing@redhat.com>
2019-12-05 01:57:40 +05:30
Sébastien Han c5e79d4876 Merge pull request #4424 from leseb/lvm-filter-again
ceph: lvm filter again
2019-12-04 21:13:32 +01:00
Kazuhito MATSUDA 05ac9489b3 ceph: use /dev/mapper/<name> for ceph-volume instead of /dev/md-x
`ceph-volume` cannot handle a device correctly if the device has an alias
under `/dev/mapper/`.
This commit fixes it by checking whether the device has that kind of alias
and using the alias as an argument for `ceph-volume`.

Signed-off-by: Kazuhito MATSUDA <kazuto.jinnai@gmail.com>
2019-12-04 12:53:50 +00:00
Sébastien Han ad18158f81 ceph: revert "ceph: osd append instead of replace"
This reverts commit 7553318235.
The append puts the filter at the end of the file and it's out of any
lvm section, so it does not work anymore.

LVM complains with:

Configuration setting "filter" invalid. It's not part of any section.

Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-04 12:03:13 +01:00
Sébastien Han 12ebb8376e ceph: revert "ceph: add the missing newline in lvm.conf's filter configuration line"
This reverts commit eb908e8959.
See next commit for the reason of the revert.

Signed-off-by: Sébastien Han <seb@redhat.com>
2019-12-04 12:02:54 +01:00
Sébastien Han cdacb4d763 Merge pull request #4418 from travisn/osd-log-errors
Continue orchestration on osd update errors
2019-12-04 10:53:49 +01:00
Alexander Trost 3cc58f639a docs: fix storageclass name used
This corrects the used StorageClass name in the docs and example
snippets shown in the docs.

Signed-off-by: Alexander Trost <galexrt@googlemail.com>
2019-12-04 10:16:14 +01:00
Sébastien Han 0a01a60de9 Merge pull request #4398 from leseb/ceph-version-output
ceph: fix ceph version output
2019-12-04 08:20:57 +01:00