* feat: batch partition catalog requests in querier
This is mostly wiring that builds on top of the other PRs linked to #8089.
I think we eventually could make the batching code nicer by adding
better wrappers / helpers, but lets do that if we have other batched
caches and this patterns proofs to be useful.
Closes#8089.
* test: extend `test_multi_get`
* test: regression test for #8286
* fix: prevent auto-flush CPU looping
* fix: panic when loading different tables at the same time
---------
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
* chore: update expected output for `COUNT` aggregates with `FILL(null)`
See #8232
* fix(influxql): fill count aggregates with 0 by default
When gap-filling a COUNT aggregate any missing rows should be filled
with 0, unless otherwise directed by a FILL clause. To do this the
projection on the aggregate plan is modiefied to coalesce any COUNT
fields with 0 unless a FILL value has been specified in the query.
* chore: add more tests
* chore: add explanation of COUNT gap filling with multiple measurements
* fix: update test introduced with merge
---------
Co-authored-by: Stuart Carnie <stuart.carnie@gmail.com>
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
Ensure that advanced syntax window functions that contain a selector,
rather than an aggregate, function are considered valid and generate
a correct plan.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
I'm going to make a change in the future that removes the access to the
namespace ID from this code, and it's not needed anyway as partitions
are uniquely identifiable by only table ID and partition key.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
This adds a second level of wait to the WAL draing & rotate before
allowing shutdown to proceed. Only once the returned receiver has
notified the called that the notification has been handled may it await
the empty waker set.
* chore: Update datafusion to get new grouping
* chore: Update for new API
* chore: update tests
* fix: new API
* fix: state type
---------
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
* fix: add retry loop to avoid CI flake in build-catalog test
* fix: Update influxdb_iox/tests/end_to_end_cases/debug.rs
---------
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
When a long running query is in process and the querier is shutting
down, it might happen that the executor (= thread pool and tokio
executor responsible for the CPU-bound DataFusion execution) is shut
down while the query is running. From a "systems interaction" PoV I
think this is totally fine and I would like to avoid some weird
ref-counting. Or in other words: if the system is shutting down, shut it
down.
However the error was treated as "internal" which is not useful. The
client should rather be informed that its server was gone and that it is
OK (and desired) to retry. So as per
<https://grpc.github.io/grpc/core/md_doc_statuscodes.html> I think this
should signal "unavailable".
This change wires the error code in such a way that the gRPC service
layer can properly inspect it and then changes the error mapping.
Ref https://github.com/influxdata/idpe/issues/17917 .
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
While I cannot reproduce the CI flakyness locally (probably because the
local system is fast enough), looking at the test convinced me that the
ingester should not persist.
Closes#8245.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
This is mostly wiring that builds on top of the other PRs linked to #8089.
I think we eventually could make the batching code nicer by adding
better wrappers / helpers, but lets do that if we have other batched
caches and this patterns proofs to be useful.
Closes#8089.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
Now that user-defined window functions are available, change the
DERIVATIVE and NON_NEGATIVE_DERIVATIVE function implementations to
use user-defined windows functions. This should improve performance
by allowing the entire window to be processed in one go, rather
than processing one row at a time.
The implementation is also moved out of the planner module alongside
the other user-defined window functions.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
There are a bunch of dependencies in `Cargo.lock` that are related to
mysql. These are NOT compiled at all, and are also not part of `cargo
tree`. The reason for the inclusion is a bug in cargo:
https://github.com/rust-lang/cargo/issues/10801
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
* refactor: refactor `Schemas` type in InfluxQL planner
* chore: update after merging main
---------
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
kodiakhq[bot]
dependabot[bot]
Marco Neumann
Joe-Blount
Joe-Blount
Martin Hilton
Carol (Nichols || Goulding)
Fraser Savage
Andrew Lamb
Carol (Nichols || Goulding)
Chunchun Ye
Christopher M. Wolff