Commit Graph
2093 Commits
Author SHA1 Message Date
travisn cfa0f43f61 check available nodes for mons only when necessary
Signed-off-by: travisn <tnielsen@redhat.com>
2018-10-17 10:59:43 -06:00
Travis Nielsen a69396af5b Merge pull request #2222 from travisn/discover-devices-interval
Reduce frequency of device discovery
2018-10-17 10:57:48 -06:00
travisn 01dc5656f3 reduce frequency of device discovery
Signed-off-by: travisn <tnielsen@redhat.com>
2018-10-17 09:56:47 -06:00
Alexander Trost f322b9fa9e Merge pull request #2218 from anguslees/fmtfix
Don't misinterpret command output as string format
2018-10-17 10:19:48 +02:00
Alexander Trost 6f6f474b9e Merge pull request #2200 from MartinForReal/master
use "COPY" instead of "ADD" in Dockerfile
2018-10-16 12:45:31 +02:00
Fan, Shang Xiang 3d7c650492 build: comply with docker best practice.
Signed-off-by: Fan, Shang Xiang <sxfan@alauda.io>
2018-10-16 13:10:02 +08:00
Angus Lees 9c95ed5f6f Don't misinterpret command output as string format
Previous code tried to interpret the string output as a `fmt.Sprintf`
string formatter, mangling `%` into `%!(MISSING)`, etc.

Thanks to go, I don't belive these are exploitable (unlike the similar
error in C).

Since this seemed to be a common error in the codebase, I did a quick
audit by visually inspecting the results of `git grep 'f([^"]'`.  I
don't have a good suggestion for automated tests to prevent this in
future :(

Example error (look for `(MISSING)`):
```
E0927 05:31:07.618429   11227 driver-call.go:237] Failed to unmarshal output for command: unmount, output: "2018-09-27 05:31:07.191711 I | exec: Running command: df --type ceph /var/lib/kubelet/pods/95461479-c216-11e8-bcf0-02030782ac80/volumes/ceph.rook.io~rook/oe-scratch\n2018-09-27 05:46:43.808596 I | Filesystem                                                 1K-blocks     Used Available Use%!M(MISSING)ounted on\n2018-09-27 05:46:43.808659 I | 10.107.25.147:6790,10.109.173.79:6790,10.104.85.255:6790:/ 151678976 49410048 102268928  33%!/(MISSING)var/lib/kubelet/pods/95461479-c216-11e8-bcf0-02030782ac80/volumes/ceph.rook.io~rook/oe-scratch\n{\"status\":\"Success\"}\n", error: invalid character '-' after top-level value
```

Signed-off-by: Angus Lees <gus@inodes.org>
2018-10-15 15:59:13 +11:00
Travis Nielsen 0c32013739 Merge pull request #2210 from SUSE/ceph-mds-enable-downsizing
ceph mds: allow mds deployment scale-down
2018-10-13 07:45:39 -06:00
Blaine Gardner 96f9a9c4ad ceph mds: allow mds deployment scale-down
fix #2193

Fix regression introduced by PR allowing MDS config creation in init
container. This allows the new deployment-based mds clusters to scale
down where currently they can only remain constant or scale up.

Signed-off-by: Blaine Gardner <blaine.gardner@suse.com>
2018-10-12 14:53:08 -06:00
Travis Nielsen ce66b38fc6 Merge pull request #2212 from jtlayton/rook-cluster-name-env
Ensure that $ROOK_CLUSTER_NAME is set for mgr daemon
2018-10-11 13:35:45 -06:00
Jeff Layton 6bab32b6f9 Ensure that $ROOK_CLUSTER_NAME is set for mgr daemon
We need this environment variable in the main container, not the init
container, as it's used by the rook ceph-mgr module.

Also fix up the env var counts in the testcase.

Fixes: ca71b7a311 (Ceph mgr: Set up config in init container)
Signed-off-by: Jeff Layton <jlayton@redhat.com>
2018-10-11 12:46:36 -04:00
Travis Nielsen 0f229c72a6 Merge pull request #2204 from jtlayton/nfs-f28
nfs: rebase nfs image onto fedora 28
2018-10-10 18:37:01 -06:00
Travis Nielsen d69fcfcd8a Merge pull request #2202 from mickymiek/integrations-tests-k8s1.12
Add Kubernetes 1.12 to Jenkins tests
2018-10-10 14:00:16 -06:00
mickymiek f93128b2d9 make sure integration tests works on k8s 1.12
Signed-off-by: mickymiek <meunie_m@etna-alternance.net>
2018-10-10 17:12:51 +02:00
Travis Nielsen 7683277e54 Merge pull request #2191 from travisn/rgw-init-container
Generate config in init container for RGW
2018-10-10 01:07:25 -04:00
travisn b43f0056ed rgw: generate config in init container
Signed-off-by: travisn <tnielsen@redhat.com>
2018-10-09 21:43:24 -06:00
Jeff Layton 522949e55f nfs: rebase nfs image onto fedora 28
f26 has been EOL'ed for a while now, so we shouldn't be relying on it.

Update to fedora 28. This necessitates installing libnsl2 for ganesha,
which is not installed in the base image these days.

Long term, we should probably find some way to keep this in sync with
the Fedora release cadence, or rebase onto something more stable like
Centos.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
2018-10-09 10:10:50 -04:00
Travis Nielsen 7de9624e91 Merge pull request #2153 from SUSE/ceph-mds-use-init-container
Ceph mds/file: Set up config in init container
2018-10-08 11:24:14 -06:00
Alexander Trost eed927ec96 Merge pull request #2093 from danielcb/update-documentation
docs: add documentation about imagePullSecrets #1713
2018-10-08 16:12:54 +02:00
Alexander Trost fb55786871 Merge pull request #2194 from galexrt/fix_merge_resources
Fix MergeResourceRequirements() func
2018-10-08 14:11:12 +02:00
Alexander Trost 34be9cf79e Merge pull request #2195 from solidnerd/fix-quickstart-link
Fix quickstart link in kubernetes examples
2018-10-07 15:54:17 +02:00
solidnerd 98d527fadc Fix quickstart link in kubernetes examples
Signed-off-by: solidnerd <niclas@mietz.io>
2018-10-07 15:50:22 +02:00
Alexander Trost 010c6747ab Fix MergeResourceRequirements() func
Resources are now correctly merged when "not set".

Signed-off-by: Alexander Trost <galexrt@googlemail.com>
2018-10-06 17:49:03 +02:00
Blaine Gardner ce803fbf9d Ceph mds/file: Set up config in init container
Progress toward issue #2003.
Includes design from design doc PR #1578

Use init containers to create configuration for Ceph mgrs. There is only
one init container in this design. The init container calls the Rook
binary to create Ceph config files which are then shared with the mds
daemon main container.

Once this init is run, the main mds daemon is run. Leaving room to use
the Ceph-versioned image in the future, call `ceph-mds --foreground ...`
to run the Ceph mds.

The refactor to using an init container also necessitated refactoring
the mdses replicaset implementation to a deployment-per-pod
implementation due to a chicken-egg problem. With a single container (in
the before times) the Rook binary was able to call the ceph-mds daemon
with an id generated from the pod name. Since the pod name is not known
before runtime, and the id is one of the few params that must be
specified to Ceph daemons on run, it is necessary to know the id
beforehand; thus the move to a deployment architecture following the
likes of the mon and mgr daemons.

Signed-off-by: Blaine Gardner <blaine.gardner@suse.com>
2018-10-05 19:01:12 -06:00
Travis Nielsen 8b018f6df4 Merge pull request #2180 from rootfs/no-log-flex
don't print console logs during flex driver call
2018-10-05 10:29:59 -06:00
Travis Nielsen 24112c4f04 Merge pull request #2166 from travisn/remove-toolbox-image
Start toolbox pod with the ceph image and stop building the toolbox image
2018-10-04 13:54:38 -06:00
Huamin Chen 97d20f9833 don't print console logs during flex driver call
Signed-off-by: Huamin Chen <hchen@redhat.com>
2018-10-03 08:46:28 -04:00
travisn efadffda82 docs: update usage of ceph toolbox
Signed-off-by: travisn <tnielsen@redhat.com>
2018-10-02 16:42:55 -06:00
travisn efd581e7ed start toolbox pod with the ceph image instead of a separate image
Signed-off-by: travisn <tnielsen@redhat.com>
2018-10-02 15:43:53 -06:00
Bassam Tabbara d7b03a59c1 Merge pull request #2179 from bassam/pr-readme
update CNCF level in README
2018-10-02 07:56:58 -07:00
Bassam Tabbara 43358f9e6b update cncf level in README
Signed-off-by: Bassam Tabbara <bassam@upbound.io>
2018-10-02 07:55:38 -07:00
Travis Nielsen 1e1ccd4aa7 Merge pull request #2163 from slayerjain/2148
ceph: Update operator to use node hostname rather than name-name.
2018-10-01 13:19:41 -06:00
Bassam Tabbara 84561c0469 Merge pull request #2172 from bassam/pr-vendor
add vendor checks
2018-10-01 10:51:37 -07:00
Jared Watts 09fc6430e8 Merge pull request #2150 from davidkarlsen/patch-1
Fix invalid claim name
2018-10-01 10:14:30 -07:00
travisn 7d97b6ad76 osd: use the node hostname labels instead of node names
Signed-off-by: travisn <tnielsen@redhat.com>
2018-10-01 08:09:06 -06:00
Bassam Tabbara eabc9bde1a simplify dependencies and add vendor.update target
remove all cosntraints that are not really needed from Gopkg.toml.
we should only be adding constraints that are direct dependencies
of the project and we require a specific version/revision of.

added a vendor.update makefile target that can be used to update
dependencies to the latest.

Signed-off-by: Bassam Tabbara <bassam@upbound.io>
2018-09-30 16:01:43 -07:00
Bassam Tabbara 727f46d124 ensure all vendor deps have constraints
all vendor dependencies in Gopkg.toml now have constraints. this will
ensure that they do not float as people run `make vendor`. if you need
a new package be sure to change the contraint explicitly.

Signed-off-by: Bassam Tabbara <bassam@upbound.io>
2018-09-30 09:56:46 -07:00
Bassam Tabbara 2464e32857 add vendor checks
this commit updates dep to version 0.5, and adds a check during
CI runs that enforces that the vendor dependencies have not changed.
There is a now a makefile target `make vendor.check` that performs
this check. Also removed the requirement that PR run make vendor as
this will be done by CI.

Signed-off-by: Bassam Tabbara <bassam@upbound.io>
2018-09-30 08:10:42 -07:00
Alexander Trost 70cbd4a675 Merge pull request #2167 from asm72/master
docs: fix typo
2018-09-28 20:19:18 +02:00
Andrea Simonini 2ce2d8f771 docs: fix typo
Signed-off-by: Andrea Simonini <andrea.simonini@gmail.com>
2018-09-28 15:22:20 +02:00
Travis Nielsen d2ab770bcc Merge pull request #2059 from rootfs/csi-doc
add ceph csi driver proposal
2018-09-26 13:20:49 -06:00
Travis Nielsen 9c46d1da26 Merge pull request #1800 from ajarr/design-nfs-ganesha-crd
Add Ceph NFS-Ganesha CRD design doc
2018-09-26 13:19:29 -06:00
Travis Nielsen 6f63d16e69 Merge pull request #2146 from travisn/osd-run-with-daemon-image
Enable filestore on device scenario to run with ceph image
2018-09-26 13:19:08 -06:00
Travis Nielsen 19da8856f0 Merge pull request #2159 from rootfs/no-logging
Revert "log the output of blkid to see if device should be formatted"
2018-09-26 11:03:10 -06:00
travisn 08f795f8b3 sanitize underscores in the volume name
Signed-off-by: travisn <tnielsen@redhat.com>
2018-09-26 10:36:36 -06:00
travisn 15f6ee5b14 osd: enable filestore on device scenario to run with ceph image
update osd specs to be consistent with other daemons

Signed-off-by: travisn <tnielsen@redhat.com>
2018-09-26 10:36:16 -06:00
Huamin Chen 3832708c4f Revert "log the output of blkid to see if device should be formatted"
This reverts commit c67542829b.

Signed-off-by: Huamin Chen <hchen@redhat.com>
2018-09-26 12:42:49 +00:00
Daniel Bross 96fe5521c1 docs: add documentation about imagePullSecrets
Adds documentation about applying an imagePullSecret to a service
account to allow all pods spawned by the operator to have this
imagePullSecret. This is necessary to access an image
(e.g. ceph-agent) from a private/authenticated docker registry
without changing the current operator code.

Signed-off-by: Daniel Bross <daniel.c.bross@gmail.com>
2018-09-25 13:58:26 +02:00
Travis Nielsen aae5c97de3 Merge pull request #2145 from ganeshmaharaj/enable-glog
Enable glog logging
2018-09-24 23:04:28 -06:00
Travis Nielsen 2d315e7392 Merge pull request #2120 from noahdesu/expose-node-info-to-ceph
ceph: expose node,pod,ns names to ceph daemons
2018-09-24 17:11:30 -06:00