…_updated

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

The name last_update_ondisk is misleading as it suggests a local
property like last_update_applied rather than a pg-global
property.  Clarify the name and add a much more specific comment.

Signed-off-by: Samuel Just <sjust@redhat.com>

Avoid maintaining pg_committed_to if pg is not active.  We can't
guarantee that last_update won't become divergent, so it doesn't
provide useful information.

Signed-off-by: Samuel Just <sjust@redhat.com>

The purpose of this rename is merely to clarify that the necessary
condition on ec roll-forward is that the pg has committed up to that
point.  Along with subsequent commits, this will clarify that both ec
and replicated pools propagate pg_committed_to for related if not
identical reasons.  Because EC::submit_transaction already did

  op->roll_forward_to = std::max(min_last_complete_ondisk, rmw_pipeline.committed_to);

there's no difference in behavior as rmw_pipeline.committed_to is
updated immediately after the notification to the PG that the write
completed.

Signed-off-by: Samuel Just <sjust@redhat.com>

…last_complete_ondisk

This commit updates the bulk of the interface pathways in crimson and
classic to refer to pg_committed_to rather than min_last_complete_ondisk
and changes the replica side to maintain pg_committed_to instead.

This commit shouldn't actually cause any behavior change -- we're still
passing min_last_complete_ondisk (which is a valid lower bound for
pg_committed_to!).

Signed-off-by: Samuel Just <sjust@redhat.com>

…nsaction

This commit actually changes the bound we're propagating.  This solves
two bugs:
- Using min_last_complete_ondisk caused replicas to be two update rounds
  behind rather than one
- Replicas don't actually have enough information to set
  min_last_complete_ondisk on activation, so we couldn't serve replica
  reads until the first write.  pg_committed_to, on the other hand, is
  fine as the activation last_update cannot become divergent.

Moreover, last_complete won't advance past missing objects causing
min_last_complete_ondisk to be blocked by any replica missing object.
Note that the replica read pathway seperately checks whether the target
is missing locally, so that property was not needed.

Fixes: https://tracker.ceph.com/issues/65086
Fixes: https://tracker.ceph.com/issues/65085
Signed-off-by: Samuel Just <sjust@redhat.com>

…Missing and related

Signed-off-by: Samuel Just <sjust@redhat.com>

…ries

This matches the behavior for normal IOs.

Signed-off-by: Samuel Just <sjust@redhat.com>

It wouldn't actually be wrong for the primary to trim the log right up
to the pg_committed_to bound it is propagating (though it generally
won't).

Signed-off-by: Samuel Just <sjust@redhat.com>

See comment for details. Modifies ECBackend::submit_transaction to
use the passed pg_committed_to unconditionally, adds a comment to
explain, and adds a comment to RMWPipeline::pg_committed_to to
clarify that it may lag PeeringState::pg_committed_to.

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Fixes: https://tracker.ceph.com/issues/65299
Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

…notify_t

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

Signed-off-by: Samuel Just <sjust@redhat.com>

…remove

We don't support balanced reads on ec pools.  Additionally, the yaml
actually specifies 'balanced_reads' rather than 'balance_reads' and
therefore has no actual effect.

Signed-off-by: Samuel Just <sjust@redhat.com>