cephcsi need to add mount the cephfs subvolume
as the readonly when the PVC type is ROX to
provide only readonly access to the users
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
With the current code base, the subvolumegroup will
be created once, and even for a different cluster,
subvolumegroup creation is not allowed again.
Added support multiple subvolumegroups creation by
validating one subvolumegroup creation per cluster.
Fixes: #1123
Signed-off-by: Yug Gupta <ygupta@redhat.com>
The previous function used to remove omap keys apparently did not
return errors when removing omap keys from a missing omap (oid).
Mimic that behavior when using the api.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
For any function that sets more than one key on a single oid setting
them as a batch will be more efficient.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
For any function that removes more than one key on a single oid removing
them as a batch will be more efficient.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
Taking this appraoch means that any function that must get more than one
key's value from the same oid can be more efficient by calling out to
ceph only once.
To be cautious and avoid missing things we always request ceph return
more keys than we actually expect to be set on the oid. If there are
unexpected keys there, we will not miss the keys we want if we first hit
an unexpected key if we were to limit ourselves to iterating only over
the number of keys we're expecting to be on the object.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
Convert the business-logic of the journal to use the new go-ceph based
omap manipulation functions.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
These new omap manipulation functions (get/set/remove) are roughly
equivalent to the previous command-line based approach but rely
on direct api calls to ceph.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
These types have private fields but we need to construct them outside of
the util package. Add New* methods for both.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
if the PVC access mode is ReadOnlyMany
or single node readonly, mounting the rbd
device path to the staging path as readonly
to avoid the write operation.
If the PVC acccess mode is readonly, mapping
rbd images as readonly.
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
go-ceph v0.3 adds constants for ImageFeature values and their names.
Instead of hardcoding "layering" in several places, use the constant
given by librbd.
The rbdVolume.ImageFeatures does not seem to be used anywhere after the
conversion. Stashing the image metadata does include the ImageFeatures
as these are retrieved when getting the image information. It is safe to
drop ImageFeatures altogether and only use the imageFeatureSet instead.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
It seems that convering the release component from the unix.Utsrelease
type leaves some trailing "\x00" characters.
While splitting the string to compare kernel versions, these additional
characters might prevent converting the string to an int. Strip the
additional characters before returning the string.
Note:
"\x00" characters are not visible when printing to a file or screen.
They can be seen in hex-editors, or sending the output through 'xxd'.
Fixes: #1167
Signed-off-by: Niels de Vos <ndevos@redhat.com>
Go 1.13 contains support for error wrapping. To support wrapping,
fmt.Errorf now has a %w verb for creating wrapped errors, and three
new functions in the errors package ( errors.Unwrap, errors.Is and
errors.As) simplify unwrapping and inspecting wrapped errors.
With this change, If we currently compare errors using ==, we have to
use errors.Is instead. Example:
if err == io.ErrUnexpectedEOF
becomes
if errors.Is(err, io.ErrUnexpectedEOF)
https://tip.golang.org/doc/go1.13#error_wrapping
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
IneffAssign warns about the two following statements:
Line 147: warning: ineffectual assignment to supported (ineffassign)
Line 148: warning: ineffectual assignment to ok (ineffassign)
Reported-by: https://goreportcard.com/report/github.com/ceph/ceph-csi
Updates: #975
Signed-off-by: Yug Gupta <ygupta@redhat.com>
This prevents the need to open the IOContext for additional operations
on the image.
It also addresses a leak of the IOContext in case `rbdVolume.open()` was
called. The method only returned the `rbd.Image` without the possibility
to close the related IOContext.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
There is a bug in current code where the devicePath
is always empty and the rbd image unmap never
happens if nodeplugin fails to mount the rbd image
to the stagingpath.
This is a fix to unmap the rbd image if some issue
occurs after rbd image is mapped.
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
When mounting fails, the node-plugin should give a suggestion to check the
kernel logs so that users can report problems better.
Edited the existing log to include the message in both rbd and cephfs.
Fixes: https://github.com/ceph/ceph-csi/issues/1006
Signed-off-by: Mudit Agarwal <muagarwa@redhat.com>
util: golint warns about exported methods to have a
comment or to unexport them.
e2e: golint warns about package comment to be of the form
"Package e2e ..."
Reported-by: https://goreportcard.com/report/github.com/ceph/ceph-csi
Updates: #975
Signed-off-by: Yug Gupta <ygupta@redhat.com>
InvalidPoolID has recently been added, and can be used in other location
too. As GetPoolID is updated with this patch set, return InvalidPoolID
on errors too.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
GetPoolID() did not return ErrPoolNotFound in case the pool could not be
found. This has been addressed as well, so that looking for an existing
pool behaves the same for checking by Name or ID.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
Before, the one CSIJournal type was handling both configuration and
providing methods to make changes to the journal. This created the
temptation to modify the state of the global configuration object to
enact changes through the method calls.
This change creates a new type `journal.Connection` that takes the
monitors and credentials to create a short(er)-lived object to actually
read and make changes on the journal. This also avoid mixing the
arguments needed to connect to the cluster with the arguments needed
for the various journal read & update calls.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
By using switch/case it is easier to follow the error checking of the
genVolFromVolID() function. In case a new error is added as a return of
the function, it will be simpler to add checking for it.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
The SetNamespace setter function was called only once, immediately after
the creation of a volume journal object in cephfs only.
Remove this function so that it is no longer implied that this field can
be mutated after the journal is created. In it's place, use an extended
"constructor" NewCSIVolumeJournalWithNamespace that takes a namespace
value at create-time only.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
The function SetCSIDirectorySuffix was used only one per (long-lived,
gloabl) journal object. It is simpler to construct the journal objects
with this needed parameter:
1. As it is required to function and non-optional AFAICT
2. Removes the temptation to mutate global object
3. Reduces LOC with exact same functionality
4. SetCSIDirectorySuffix would not behave correctly if called a 2nd time
anyway.
Point 4. means that if you called the function twice to change the
suffix when you previously had "csi.volumes.alice", you'd get
"csi.volumes.alice.bob" instead of "csi.volumes.bob" what one would
expect.
Signed-off-by: John Mulligan <jmulligan@redhat.com>
The problem happens when multiple PVCs with the
same UUID are attached/mounted on a node. This
can happen after creating a PVC from a snapshot,
or cloning a PVC.
make nouuid as the default mount option if
the format type is xfs to avoid mounting
issues.
updates: #966
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
The gocyclo linter complains about the high complexity of the
CreateVolume() function:
> pkg/rbd/controllerserver.go:133:1: cyclomatic complexity 21 of func `(*ControllerServer).CreateVolume` is high (> 20) (gocyclo)
By splitting it up and separeting the creation of an exisint CSI Volume
object in buildCreateVolumeResponse(), the gocyclic linter does not
complain any longer.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
