The process used to build the snapshots no longer works because the box it runs on is outdated. Analysing the web logs shows the majority of traffic to these links is from search engine bots. With those removed, there is about 1 hit per day from (probable) humans. Most users needing a tarball are better served by using official releases. Those needing latest code are better served by using git checkout. The tarball snapshots are not compelling enough to invest time in fixing the script that produces them. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com&gt; --- NB when pushing this, I would delete the existing files from the download location so we aren't left with stale files, and disable the (now broken) cronjob. docs/downloads.html.in | 15 --------------- 1 file changed, 15 deletions(-) diff --git a/docs/downloads.html.in b/docs/downloads.html.in index 2956a4f60e..657cd6a6f8 100644 --- a/docs/downloads.html.in +++ b/docs/downloads.html.in @@ -433,21 +433,6 @@ <li><a href="https://libvirt.org/sources/">libvirt.org HTTPS server</a></li> </ul> - <h2><a id="hourly">Hourly development snapshots</a></h2> - - <p> - Once an hour, an automated snapshot is made from the git server - source tree. These snapshots should be usable, but we make no guarantees - about their stability; furthermore, they should NOT be - considered formal releases, and they may have transient security - problems that will not be assigned a CVE. - </p> - - <ul> - <li><a href="ftp://libvirt.org/libvirt/libvirt-git-snapshot.tar.xz">... FTP server</a></li> - <li><a href="https://libvirt.org/sources/libvirt-git-snapshot.tar.xz"&... HTTP server</a></li> - </ul> - <h2><a id="schedule">Primary release schedule</a></h2> <p> -- 2.17.0