You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Oct 4, 2020. It is now read-only.
I'm thinking -- restore via extract as a job. Basically runs "borg extract" with access to exactly one archive and without leaking our keys (as usual) on a client (or arbitrary host). Perhaps support short-lived SSH keys for this (deleted from the server after the job completes).
The text was updated successfully, but these errors were encountered:
For interactive browsing a more efficient access than tape-like is needed, which is why I started to work on metadata indexing by storing offsets into msgpack streams in a trie (HAT). This should work reasonably well but requires a pass over the stream to generate the trie. Some tests have shown that these compress rather well, so it might not be an issue to generate these by default when syncing the server cache (bc.job.backup); but also on-demand in the interface.
This also somewhat depends on the whole repository arbiter idea (#14).
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Kinda pointless to not have this.
I'm thinking -- restore via extract as a job. Basically runs "borg extract" with access to exactly one archive and without leaking our keys (as usual) on a client (or arbitrary host). Perhaps support short-lived SSH keys for this (deleted from the server after the job completes).
The text was updated successfully, but these errors were encountered: