the offset must be relative to the start of THIS large block, not to the
start of the disk — otherwise every large block after index 0 lands past
the end of destinationBlock and copy() silently writes nothing (zeroed block).
affect qcow2 incremental backup with NBD disable
NBD + qcow2 and Vhd backups don't go through this code path
The previous static/js/matomo.js snippet never loaded the tracker
script, so nothing was ever recorded; it also pointed at site 5, the
XO website. Replace it with the client module from the Vates VMS docs
on the new dedicated docs site (23): cookieless page-view tracking that
follows SPA navigations, and instrumentation of the lunr search box
reporting each settled query with its suggestion count, so zero-result
searches surface missing content. The feedback widget mirrors votes and
thumbs-down reasons as doc_feedback events.
Also fills in the Formbricks IDs for the "XO Docs page feedback"
survey, enabling the feedback widget.
Swizzles DocItem/Footer to show a 'Was this page helpful?' widget under
every doc article, ported from the Vates VMS docs. Votes post to the
self-hosted Formbricks instance via its public client API; a thumbs-down
asks why (outdated, unclear, missing info, inaccurate) plus optional
free text. The vote is recorded on click so it survives the visitor
leaving before the follow-up. Page path rides along as a hidden field
for per-page filtering.
* ensure cache is generated completely at the end of the backup job
* ensure a broken cache is regenerated even without merge during the cleanup phase
* ensure a broken cache is regenerated on request
add new option and script to launch a load test :
number of vm is configurable
concurrency is configurable
number of backup run is configurable
will do multiple run, then a restore
- Show only traffic rules of VMs. Don't include `vm-snapshot` or `vm-template`
- For VM selection, display all VMs and disable those without a VIF, with a
message displayed
A VM backup log status is calculated depending on its subtask status. Previously, we made that it would be a failure if at least one of its subtask is a failure.
However, in case of successful retries, the intended behaviour is to have failed subtasks, which should not affect the parent task status. The first commit of this PR fixes that.
The second commit simplifies the getStatus function and its usage.
---------
Co-authored-by: Florent BEAUCHAMP <florent.beauchamp@vates.tech>
