Files
kopia/internal
Jarek Kowalski 331c7f6298 epoch: changed how very slow index writes are handled (#1587)
Previously epoch manager would discard (on read) any index blobs
associated with epoch N written after epoch N+2 has already started.

There was no signal to the writer that the write is actually bad. This
change reports slow writes as a failure and disables read-time filtering.

This is important for repositories copied using sync-to because it
may not necessarily preserve the modification time leading to problems
with restore.
2021-12-15 07:06:28 -08:00
..
2020-02-25 17:24:44 -08:00
2021-11-11 17:03:11 -08:00
2021-09-14 19:11:39 -07:00
2021-06-16 19:44:55 -07:00