mirror of
https://github.com/retspen/webvirtcloud
synced 2026-11-09 19:56:48 +00:00
fa5a8c4d6e
- The page lists the volumes first: that refreshes the pool, and libvirt measures a pool's space only when it is refreshed, so the size and usage shown were as of the previous refresh. - A space that cannot be read says so instead of 0 bytes, and the pool's status, read from the same pool info, no longer fails the page after it; a stopped pool says its space and volumes are unavailable until it is started, instead of 0 bytes and nothing. - The API's pool is measured after the same refresh, and a stopped pool lists no volumes there instead of failing. - A clone the job refuses (a name in use, a volume gone) is a message on the page, not a server error. - An ISO image is recognised by its name (storages.jobs.is_iso): libvirt calls its format raw, so the template's "iso" check never disabled its Clone button. The clone job refuses one too; the upload uses the same rule. - One clone dialog serves every volume, filled and reset by the button that opens it, instead of one dialog per volume in the page; its fields have labels and its title names the dialog. - Volume rows are searched by name only (data-search), with a message when none matches, in a live region for screen readers; names show in full. The search script loads only with the volume table: on a stopped pool's page it failed for want of a search box. An empty pool says "This pool has no volumes". - The unused wvmStorage.get_volumes_details goes: update_volumes is the same. |
||
|---|---|---|
| .. | ||
| api | ||
| migrations | ||
| templates | ||
| tests | ||
| __init__.py | ||
| forms.py | ||
| jobs.py | ||
| models.py | ||
| upload.py | ||
| views.py | ||