Commit Graph

87 Commits

Author SHA1 Message Date
ShyamsundarR
5c4abf8347 Add topology support to ceph-csi
Signed-off-by: ShyamsundarR <srangana@redhat.com>
2020-04-14 14:14:29 +00:00
Humble Chirammal
34fc1d847e Changes to accommodate client-go changes and kube vendor update
to v1.18.0

Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
2020-04-14 10:50:12 +00:00
Niels de Vos
c3cf6be6a7 util/conn_pool: open a connection with requested user
Use the Credentials.ID in combination with the keyfile to connect to the
Ceph cluster. This makes it possible to use different users for
different tasks on the cluster.

Fixes: #904
Signed-off-by: Niels de Vos <ndevos@redhat.com>
2020-04-09 15:10:55 +00:00
Niels de Vos
19cc28ddea util/cephcmds: have GetOMapValue() return ErrPoolNotFound
On occasion the e2e tests fail as there is an unexpected error while
deleting an RBD image. The particular tests forcefully removes the pool
where the RBD image is stored. Deleting a volume that has been removed
already (or when its parent pool has been wiped), should succeed.

By catching the error that a pool does not exist (anymore), the
provisioner responds to the DeleteVolume request with succes.

Signed-off-by: Niels de Vos <ndevos@redhat.com>
2020-04-08 12:44:42 +00:00
Madhu Rajanna
8b70e68f65 Fix issue related to http status code in vault
the status code for success should be between
200 to 300.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2020-04-06 11:01:25 +00:00
Niels de Vos
a1de56dbd3 tests: in case 'go test' is run in a container, skip TestGetPIDLimit()
In (standard, non-privileged) container environments the /sys/fs/cgroup
mountpoint is not available. This would cause the tests to fail, as
TestGetPIDLimit() tries to write to the cgroup configuration.

The test will work when run as root on a privileged container or
directly on a host (as Travis CI does).

Setting the CEPH_CSI_RUN_ALL_TESTS environment variable to a non-empty
value will cause the test to be executed.

Signed-off-by: Niels de Vos <ndevos@redhat.com>
2020-04-02 06:08:03 +00:00
Madhu Rajanna
84aa1ba392 Use Error instead of Errorf
If the string formatting is not required
use Error.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2020-03-30 08:53:16 +00:00
Humble Chirammal
b1dfcb4d7e Correct static errors and source code comments.
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
2020-03-30 05:13:24 +00:00
xu.chen
399f0b0d89 Audit log and follow klog standard 2020-03-27 09:24:52 +00:00
root
ae4d269836 fix typos 2020-03-24 15:43:03 +00:00
Niels de Vos
90f81516ee util/conn_pool: add tests for ConnPool
Signed-off-by: Niels de Vos <ndevos@redhat.com>
2020-03-11 16:09:10 +00:00
Niels de Vos
397825c665 util: add ConnPool for connection re-use
By using the ConnPool it is not needed to re-connect every time to the
Ceph cluster when (rbd) operations are executed through the go-ceph/rbd
API.

Signed-off-by: Niels de Vos <ndevos@redhat.com>
2020-03-11 16:09:10 +00:00
Reinier Schoof
a4532fafd0 added volumeNamePrefix and snapshotNamePrefix as parameters for storageClass
this allows administrators to override the naming prefix for both volumes and snapshots
created by the rbd plugin.

Signed-off-by: Reinier Schoof <reinier@skoef.nl>
2020-02-25 05:03:51 +00:00
Madhu Rajanna
8dcb6a6105 Handle Delete operation if pool not found
If the backend rbd or cephfs pool is already deleted
we need to return success to the  DeleteVolume RPC
call to make it idempotent.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2020-02-11 15:58:15 +00:00
Vasyl Purchel
669dc4536f Reduce encryption KMS configuration SC parameters
* moves KMS type from StorageClass into KMS configuration itself
 * updates omapval used to identify KMS to only it's ID without the type

why?

1. when using multiple KMS configurations (not currently supported)
automated parsing of kms configuration will be failing because some
entries in configs won't comply with the requested type
2. less options are needed in the StorageClass and less data used to
identify the KMS

Signed-off-by: Vasyl Purchel vasyl.purchel@workday.com
Signed-off-by: Andrea Baglioni andrea.baglioni@workday.com
2020-02-10 15:21:11 +00:00
Vasyl Purchel
419ad0dd8e Adds per volume encryption with Vault integration
- adds proposal document for PVC encryption from PR448
- adds per-volume encription by generating encryption passphrase
  for each volume and storing it in a KMS
- adds HashiCorp Vault integration as a KMS for encryption passphrases
- avoids encrypting volume second time if it was already encrypted but
  no file system created
- avoids unnecessary checks if volume is a mapped device when encryption
  was not requested
- prevents resizing encrypted volumes (it is not currently supported)
- prevents creating snapshots from encrypted volumes to prevent attack
  on encryption key (security guard until re-encryption of volumes
  implemented)

Signed-off-by: Vasyl Purchel vasyl.purchel@workday.com
Signed-off-by: Andrea Baglioni andrea.baglioni@workday.com

Fixes #420
Fixes #744
2020-02-05 05:18:56 +00:00
Humble Chirammal
3af1e26d7c Update to kube v1.17
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
2020-01-17 12:06:02 +00:00
Madhu Rajanna
dcafdb519e discard umount error if directory is not mounted
if the directory is not mounted return nil
during umount of mountPoint

Discard error if error is os.IsNotExist

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-12-17 13:33:41 +00:00
Vasyl Purchel
166eaf700f Adds PVC encryption with LUKS
Adds encryption in StorageClass as a parameter. Encryption passphrase is
stored in kubernetes secrets per StorageClass. Implements rbd volume
encryption relying on dm-crypt and cryptsetup using LUKS extension

The change is related to proposal made earlier. This is a first part of
the full feature that adds encryption with passphrase stored in secrets.

Signed-off-by: Vasyl Purchel vasyl.purchel@workday.com
Signed-off-by: Andrea Baglioni andrea.baglioni@workday.com
Signed-off-by: Ioannis Papaioannou ioannis.papaioannou@workday.com
Signed-off-by: Paul Mc Auley paul.mcauley@workday.com
Signed-off-by: Sergio de Carvalho sergio.carvalho@workday.com
2019-12-16 08:12:44 +00:00
Woohyung Han
8a16f740d6 Update golangci-lint version to v1.21.0
Signed-off-by: Woohyung Han <techhanx@gmail.com>
2019-12-12 04:57:14 +00:00
Madhu Rajanna
118f34525e Remove deprecated containerized
As we are moving towards v2.0.0 I think
it's a good time to remove the deprecated flag.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-12-11 14:44:35 +00:00
Madhu Rajanna
dfc3562e29 Add Version flag to cephcsi
This will be helpful if someone wants
to check the cephcsi version

output
```
docker run quay.io/cephcsi/cephcsi:v1.2.1 --version
Cephcsi Version: v1.2.1
Git Commit: 4b871366327d63e27fc1abfb699f0faaf0fc16b9
GoVersion: go1.12.5
Compiler: gc
Platform: linux/amd64
```

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-11-06 12:35:28 +00:00
Stefan Haas
6a2717ce20 Added forcecephkernelclient as startup parameter to force enabling ceph
Signed-off-by: Stefan Haas <shaas@suse.com>
2019-10-16 06:47:10 +00:00
Madhu Rajanna
239822f147 reuse existing code for size Roundoff
This PR addresses the review comments in
https://github.com/ceph/ceph-csi/pull/644

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-10-11 11:07:39 +00:00
Madhu Rajanna
7274bd09e5 Fix volsize for cephfs and rbd
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-10-11 08:22:27 +00:00
Madhu Rajanna
b8568a5bb9 Add a check for nil secrets
Improve the error message if
secrets are not provided in request

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-09-27 05:10:01 +00:00
Madhu Rajanna
6aac399075 Change the logic of locking
if any on going opearation is seen,we
have to return Abort error message

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-09-20 07:37:17 +00:00
Madhu Rajanna
e395080cdc Add req-ID to logging
with this log format we can easily identify
the logs per request

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-09-11 13:45:40 +00:00
Madhu Rajanna
ed9330d2f6 rename Key to CtxKey
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-09-11 13:45:40 +00:00
Madhu Rajanna
f4b38228ae Remove volumemounter flag from cephfs
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-09-05 07:20:50 +00:00
Niels de Vos
dd668e59f1 Address security concerns reported by 'gosec'
gosec reports several issues, none of them looks very critical. With
this change the following concerns have been addressed:

[pkg/cephfs/nodeserver.go:229] - G302: Expect file permissions to be 0600 or less (Confidence: HIGH, Severity: MEDIUM)
  > os.Chmod(targetPath, 0777)

[pkg/cephfs/util.go:39] - G204: Subprocess launched with variable (Confidence: HIGH, Severity: MEDIUM)
  > exec.Command(program, args...)

[pkg/rbd/nodeserver.go:156] - G302: Expect file permissions to be 0600 or less (Confidence: HIGH, Severity: MEDIUM)
  > os.Chmod(stagingTargetPath, 0777)

[pkg/rbd/nodeserver.go:205] - G302: Expect file permissions to be 0600 or less (Confidence: HIGH, Severity: MEDIUM)
  > os.OpenFile(mountPath, os.O_CREATE|os.O_RDWR, 0750)

[pkg/rbd/rbd_util.go:797] - G304: Potential file inclusion via variable (Confidence: HIGH, Severity: MEDIUM)
  > ioutil.ReadFile(fPath)

[pkg/util/cephcmds.go:35] - G204: Subprocess launched with variable (Confidence: HIGH, Severity: MEDIUM)
  > exec.Command(program, args...)

[pkg/util/credentials.go:47] - G104: Errors unhandled. (Confidence: HIGH, Severity: LOW)
  > os.Remove(tmpfile.Name())

[pkg/util/credentials.go:92] - G104: Errors unhandled. (Confidence: HIGH, Severity: LOW)
  > os.Remove(cr.KeyFile)

[pkg/util/pidlimit.go:74] - G304: Potential file inclusion via variable (Confidence: HIGH, Severity: MEDIUM)
  > os.Open(pidsMax)

URL: https://github.com/securego/gosec
Signed-off-by: Niels de Vos <ndevos@redhat.com>
2019-09-04 11:48:37 +00:00
Madhu Rajanna
a81a3bf96b implement grpc metrics for ceph-csi
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-08-30 06:50:32 +00:00
Daniel-Pivonka
01a78cace5 switch to cephfs, utils, and csicommon to new loging system
Signed-off-by: Daniel-Pivonka <dpivonka@redhat.com>
2019-08-29 14:04:31 +00:00
Madhu Rajanna
38ca08bf65 Context based logging for rbd
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-08-26 06:19:24 +00:00
Daniel-Pivonka
81c28d6cb0 implement klog wrapper
Signed-off-by: Daniel-Pivonka <dpivonka@redhat.com>
2019-08-21 14:36:41 +00:00
Madhu Rajanna
0da4bd5151 start controller or node server based on config
if both controller and nodeserver flags are set/unset
cephcsi will start both server,

if only one flag is set, it will start relavent
service.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-08-19 06:11:43 +00:00
Madhu Rajanna
89732d923f move flag configuration variable to util
remove unwanted checks
remove getting drivertype from binary name

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-08-19 06:11:43 +00:00
Niels de Vos
31648c8feb provisioners: add reconfiguring of PID limit
The container runtime CRI-O limits the number of PIDs to 1024 by
default. When many PVCs are requested at the same time, it is possible
for the provisioner to start too many threads (or go routines) and
executing 'rbd' commands can start to fail. In case a go routine can not
get started, the process panics.

The PID limit can be changed by passing an argument to kubelet, but this
will affect all pids running on a host. Changing the parameters to
kubelet is also not a very elegant solution.

Instead, the provisioner pod can change the configuration itself. The
pod is running in privileged mode and can write to /sys/fs/cgroup where
the limit is configured.

With this change, the limit is configured to 'max', just as if there is
no limit at all. The logs of the csi-rbdplugin in the provisioner pod
will reflect the change it makes when starting the service:

    $ oc -n rook-ceph logs -c csi-rbdplugin csi-rbdplugin-provisioner-0
    ..
    I0726 13:59:19.737678       1 cephcsi.go:127] Initial PID limit is set to 1024
    I0726 13:59:19.737746       1 cephcsi.go:136] Reconfigured PID limit to -1 (max)
    ..

It is possible to pass a different limit on the commandline of the
cephcsi executable. The following flag has been added:

    --pidlimit=<int>       the PID limit to configure through cgroups

This accepts special values -1 (max) and 0 (default, do not
reconfigure). Other integers will be the limit that gets configured in
cgroups.

Signed-off-by: Niels de Vos <ndevos@redhat.com>
2019-08-13 14:43:29 +00:00
ShyamsundarR
925bda2881 Move mounting staging instance to a sub-path within staging path
This commit moves the mounting of a block volumes and filesystems
to a sub-file (already the case) or a sub-dir within the staging
path.

This enables using the staging path to store any additional data
regarding the mount. For example, this will be extended in the
future to store the fsid of the cluster, and maybe the pool name
to map unmap requests to the right image.

Also, this fixes the noted hack in the code, to determine in a
common manner if there is a mount on the passed in staging path.

Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-08-13 14:07:52 +00:00
Madhu Rajanna
dfbdec4b6a add validation to check if stagingPath exists
It's CO responsibility to create the
stagingPath as per the CSI spec.

The CO SHALL ensure
// that the path is directory and that the process serving the
// request has `read` and `write` permission to that directory. The
// CO SHALL be responsible for creating the directory if it does not
// exist.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-07-29 12:52:10 +00:00
Humble Devassy Chirammal
c7d990a96b
Merge pull request #460 from Madhu-1/fix-pluginapath
Fix pluginpath for cephfs
2019-07-29 14:02:18 +05:30
ShyamsundarR
bd204d7d45 Use --keyfile option to pass keys to all Ceph CLIs
Every Ceph CLI that is invoked at present passes the key via the
--key option, and hence is exposed to key being displayed on
the host using a ps command or such means.

This commit addresses this issue by stashing the key in a tmp
file, which is again created on a tmpfs (or empty dir backed by
memory). Further using such tmp files as arguments to the --keyfile
option for every CLI that is invoked.

This prevents the key from being visible as part of the argument list
of the invoked program on the system.

Fixes: #318

Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-07-25 12:46:15 +00:00
Madhu Rajanna
a5164cfa41 Avoid keyring message while logging
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-07-25 09:48:09 +00:00
Madhu Rajanna
778cfb3090 provide option to set pluginpath for cephfs
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-07-25 14:47:42 +05:30
Madhu Rajanna
f4c80dec9a Implement NodeStage and NodeUnstage for rbd
in NodeStage RPC call  we  have to map the
device to the node plugin and make  sure  the
the device will be mounted to  the global path

in  nodeUnstage request unmount the device from
global path and unmap the device

if the volume mode is block  we will be creating
a file inside a stageTargetPath  and it will be
considered  as the global path

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-07-24 12:49:21 +00:00
Humble Devassy Chirammal
5d5a6c4d91
Merge pull request #469 from Madhu-1/driver-version
Update driver version during build time
2019-07-24 14:41:45 +05:30
ShyamsundarR
e5e332eded Use correct file descriptor to parse errors
File descriptors in use to parse errors from a few command
invocations were incorrect. This led to inability to detect
certain errors cases and act accordingly.

One of the easiest noticeable issues was when an image is deleted
but its RADOS keys and maps are still intact. In such cases
the DeleteVolume call always errored out unable to find the
image rather than, proceed with cleaning up the RADOS objects
and returning a success.

The original method of using stdout was incorrect, as the command
was tested from within a shell script and the scripts STDIN/OUT/ERR
was redirected to understand behavior. This is now tested using just
the CLI in question, and also examining Ceph code, and further
testing a couple of edge conditions by deleting backing images
for PVs

Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-07-16 07:51:10 +00:00
Madhu Rajanna
3f8bd3b2a6 Update driver version during build time
update driver version and add git commit
to the image. This will help us to identify
what latest git commit image contains.

Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
2019-07-12 15:54:52 +05:30
Poornima G
32ea550e3a Modify CephFs provisioner to use the ceph mgr commands
Currently CephFs provisioner mounts the ceph filesystem
and creates a subdirectory as a part of provisioning the
volume. Ceph now supports commands to provision fs subvolumes,
hance modify the provisioner to use ceph mgr commands to
(de)provision fs subvolumes.

Signed-off-by: Poornima G <pgurusid@redhat.com>
2019-07-12 05:42:41 +00:00
ShyamsundarR
c4a3675cec Move locks to more granular locking than CPU count based
As detailed in issue #279, current lock scheme has hash
buckets that are count of CPUs. This causes a lot of contention
when parallel requests are made to the CSI plugin. To reduce
lock contention, this commit introduces granular locks per
identifier.

The commit also changes the timeout for gRPC requests to Create
and Delete volumes, as the current timeout is 10s (kubernetes
documentation says 15s but code defaults are 10s). A virtual
setup takes about 12-15s to complete a request at times, that leads
to unwanted retries of the same request, hence the increased
timeout to enable operation completion with minimal retries.

Tests to create PVCs before and after these changes look like so,

Before:
Default master code + sidecar provisioner --timeout option set
to 30 seconds

20 PVCs
Creation: 3 runs, 396/391/400 seconds
Deletion: 3 runs, 218/271/118 seconds
  - Once was stalled for more than 8 minutes and cancelled the run

After:
Current commit + sidecar provisioner --timeout option set to 30 sec
20 PVCs
Creation: 3 runs, 42/59/65 seconds
Deletion: 3 runs, 32/32/31 seconds

Fixes: #279
Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-07-01 14:10:14 +00:00