Rather than deriving the identity of template variables by string parsing the
name, identify its relationship to the underlying struct using reflection.
Signed-off-by: Jonathon Anderson <janderson@ciq.com>
- Updated `wwctl upgrade` to handle updates
- Maintained `.Container` and `.ContainerName` in tstruct
- Added `ContainerName()` methods to node and profile objects
- Added `--container`, `-C` aliases to wwctl commands (`<node|profile> <add|set>`)
- Added `wwctl container` alias
- Added support for `container_exit.sh` if `image_exit.sh` is not found
Signed-off-by: Jonathon Anderson <janderson@ciq.com>
Before #1628, all default constructors for nodes and profiles attempted to
construct objects with pre-populated empty fields (e.g., for pointers to member
structs and for collections that default to nil). Changes to MergeNode in #1629
changed this behavior intentionally to simplify the generation of minimal json;
but this had the unintended side-effect of breaking existing templates that may
now encounter a nil pointer, particularly when accessing member structs.
Introduced here, the Expand() methods on Node and Profile are now responsible
for populating such fields, and Node.Expand() is now called during template
processing.
Signed-off-by: Jonathon Anderson <janderson@ciq.com>
- Make Resources an `interface{}` to support arbitrary yaml
- Remove `wwctl resource` as incompatible with arbitrary yaml
- Revert changes to host overlay templates
- Remove NFS mount options from warewulf.conf
- Replace NFS support / resource prefix with "fstab" resource
- Move resources to profiles
- Migrate warewulf.conf mounts to nodes.conf with `wwctl upgrade`
- Add documentation
Signed-off-by: Jonathon Anderson <janderson@ciq.com>
introduced wwbool and don't export
Nodes and NodeConfs. This requires new
explict Yaml (un)marshaling as the standard
marshaller won't touch these fields
Signed-off-by: Christian Goll <cgoll@suse.com>
This is a significant change in the undelying data model!
nodeDb, err := node.New()
will result in a structure which contains the on disk
values. Only
nodeDb.FindAllNodes() or nodeDb.GetNode(id) will give
the nodes with its merged in profiles.
Signed-off-by: Christian Goll <cgoll@suse.com>
This is needed for printing out the values, as at this part in the
code there is no access to the NodeConf and only there the information
if the Entry is ment as a slice or not is available.
Signed-off-by: Christian Goll <cgoll@suse.com>
If a node contains several profiles and these profiles have
overlays attached, these profiles are now merged. Overlays
can't be excluded as this can not be displayed to the user
Signed-off-by: Christian Goll <cgoll@suse.com>
Every struct in a NodeConf with `lopt:"foo" set is now added as a
command-line flag with the RecursiveCreateFlags call. For maps a struct
with the key UNDEF is added so that it can be parsed out.
As the flags for the command-line need variables which hold the values,
for every map an element map[UNDEF] is added. When now calling the
internal add, these map element can be filtered out and replace by the
given name. (e.g., --netname)
* rewrote node/profile add for recursive functions
* rewrote node/profile set for recursive functions
* rewrote node/profile list for recursive functions
Signed-off-by: Christian Goll <cgoll@suse.com>
when transforming a NodeInfo struct to a NodeConf (set the NodeConf from
the NodeInfo) nested structures were seperately handled. This means a
there were seperate handles for e.g *Netdevs and *IpmiConfig. The new
getter traverses through the datastucture and sets the right value for a
Entry struct, but calls itself again if a Pointer is detected.
This adds the posiblity to add new nested structures to NodeConf and
NodeInfo without the need of seperately handling them in the transformer
functions.
Signed-off-by: Christian Goll <cgoll@suse.de>
* added the additional type flag, which can be bool,IP
* UNDEF is recognized for the bool flag
* 0.0.0.0 must be used as UNDEF for IP flag
Signed-off-by: Christian Goll <cgoll@suse.de>
