Rob Gill
Rob Gill
The approach used by CentOS on these older processors is detected correctly. Under the patched kernel, the script reports: CVE-2017-5715 [branch target injection] aka 'Spectre Variant 2' * Mitigated according...
Oh, it gets much worse downstream from that. It's still not possible to clone repos from github.com via IPv6. Nor can FTL be retrieved from ftl.pi-hole.net.
> IPv6-only hosts/clients can live very well with DNS64+NAT64 environments. Yes but if only the v4 route was addressed, the installer would fail in messy ways on systems that lacked...
The problem appears to be on the display side rather than the logging. The correct offset is written to FTL.log and is present in Pihole-FTL.db From Pi-hole diagnosis: 2025-06-02 22:52:01...
> I think you found part of the issue That's right, it only addresses the incorrect display in diagnosis. eg `"Hostname of client => "PX!dU" contains (at least) one invalid...
> Please add the test to `.github/workflows/test.yml`, so it will run via GH action and we can see the status here. Recommended changes to tests done. Changes to scripting to...
Conflicts resolved, good to go.
> I also have a custom fork adding alpine support that includes a proper openrc script, and uses a virtual package for dependencies instead of requiring the full `abuild` install...
Drafting this for now, and will close it when #6275 merges
Replacement PR is well underway. Closing this.