* fix: do not loose data when Kafka reports that offset is above watermark
This can happen in certain cluster rebalance settings.
This is also linked to https://github.com/influxdata/rskafka/issues/147
but for the upstream issue I currently have no idea how to fix it, so
let's at least harden IOx against it.
Fixes#5128.
* refactor: panic for `SequenceNumberAfterWatermark`
* chore: return a struct with named and documented fields from `compact_persisting_batch`
* docs: Remove extra 'the' and fix a typo
Co-authored-by: Carol (Nichols || Goulding) <carol.nichols@gmail.com>
* feat: `QueryChunk::as_any`
* feat: allo `ChunkPruner::prune_chunks` to fail
* feat: limit per-table chunk data for every query
Closes#5211.
* fix: address review comments
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
* refactor: remove min_sequnce_number
* fix: typos
* fix: remove min_sequencer_number from new files from merging main
* fix: add back throwing error if the compactor compacts files persisted by the ingester after the ingester sends max seq_num back to querier
* test: add test_compactor_collision back but modify the input to make it work woth new changes
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
- remove `IOxSessionContext::default()` because untracked contexts
should only be created by tests
- remove `Option<IOxSessionContext>` because it is a typed workaround
for `IOxSessionContext::default`
Tests should use `IOxSessionContext::testing` and all _normal_ users
should create proper contexts.
I suspect this will help tracing or at least prevent silent regressions.
See #5129.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
* refactor: remove parquet chunk ID to `ChunkMeta`
* refactor: return `Arc` from `QueryChunk::summary`
This is similar to how we handle other chunk data like schemas. This
allows a chunk to change/refine its "believe" over its own payload while
it is passed around in the query stack.
Helps w/ #5032.
* test: add test for timestamps in kafka write buffer
* refactor: move timestamp batching test to generic tests
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
Instead of using some hand-rolled timestamp-based logic (or just
"unknown") all over the place, just use logic introduced in #5017.
This requires slightly improved table summaries within the querier that
at least has min/max for the timestamp column. For that, the former
`IngesterChunk`-specific `calculate_summary` method was extended to
`create_basic_summary` to include that data and is now also used by
`QuerierParquetChunk`.
Note: `QuerierRBChunk` already has detailled metrics that are provided
by the read buffer implementation.
Should we ever need even better pruning for `QuerierParquetChunk` (or
`IngesterChunk`) then we _only_ need add extra data to the table
summaries.
Closes#4976.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
Went through and remove all lazy_static uses with once_cell (while waiting for the project to compile). There are still dependencies using lazy_static so it is still in the crate graph but at least there isn't an explicit dependency on it (and it is easier to update to `std::lazy::Lazy` once that is stable).
* refactor: change level 1 to level 2 preparing for next design changes
* fix: make level-2 consistent everywhere
* chore: remove unused comments
* refactor: change all the name level_1 to level_2 to completely replace 1 with 2 to amke everything consistent
* chore: add correspinding constants for the comapction levels in the comments
Co-authored-by: Dom <dom@itsallbroken.com>
* fix: avoid using min_time, which can be negative, for ChunkId. Using object store id which is uuid instead
* chore: Apply suggestions from code review
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
* chore: run fmt
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
The ingester maintains a rough "total memory in use" counter it uses to
try and limit the amount of memory the ingester is using overall.
When a partition is persisted, this total memory usage value is adjusted
to account for releasing the partition memory. Prior to this commit, the
ordering was:
* Writes increase the memory counter
* maybe_persist() is called to trigger persistence
* A partition is identified for persistence
* Partition memory usage is released back to the total memory counter
* Persistence starts
This meant that the partitions in the process of being persisted were
not accounted for in the ingester's total memory counter, and therefore
we could significantly overrun the configured memory limit.
After this commit, the ordering is:
* Writes increase the memory counter
* maybe_persist() is called to trigger persistence
* A partition is identified for persistence
* Persistence starts
* Persistence completes
* Partition memory usage is released back to the total memory counter
This ensures persisting partitions are sill tracked in the total memory
counter, causing pauses to correctly fire.
Changes the ingester to use the partition key derived in the router, and
transmitted over through the kafka API boundary.
This should have no observable behavioural change, but be more resilient
as we're no longer assuming the partitioning algorithm produces the same
value in both the router (where data is partitioned) and the ingester
(where data is persisted, segregated by partition key).
This is a pre-requisite to allowing the user to specify partitioning
schemes.
* refactor: store per-file column set in catalog
Together with the table-wide schema and the partition-wide sort key, this should
be everything we need to read a parquet file directly into memory
without peeking any file-level metadata.
The querier will use this to directly load parquet files into the read
buffer.
**WARNING: This requires a catalog wipe!**
Ref #4124.
* refactor: use proper `ColumnSet` type
* refactor: stream partitions from ingester
Ref #4849.
* refactor: do not collect record batched on the ingester side
Ref #4849.
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
* refactor: use new ingester<>querier wire protocol
Use and document the new and more flexible ingester<>querier wire
protocol.
Note that the ingester does NOT stream the response data yet, but the
internal data structures would allow that. A follow-up change will
adjust the ingester code to stream the data.
Ref #4849.
* fix: typos
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
* refactor: clarify naming and public interface
* test: add schema assertion to `ingester_response_to_record_batches`
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
* refactor: prepare new ingester<>querier protocol on the querier side
This changes the querier internals to work with the new protocol. The
wire protocol stays the same (for now). There's a (somewhat hackish)
adapter in place on the querier side that converts the old to the new
protocol on-the-fly. This is an intermediate step before we actually
change the wire protocol (and in a step after that also take advantage
of the new possibilites on the ingester side).
Ref #4849.
* docs: explain adapter
* chore: TEMP Update DataFusion to pre-release
* chore: update arrow et al to 16.0.0
* chore: Run cargo hakari tasks
* fix: update reader read_dictionary API
* chore: Update to real Datafusion release
* fix: Update parquet API
* fix: update test
Co-authored-by: CircleCI[bot] <circleci@influxdata.com>
This commit changes the code base to use a new reference-counted
PartitionKey type wrapper, instead of passing a bare String around.
This allows the compiler to type check & verify usage of the partition
key, instead of passing a bare string around. By reference counting the
underlying string, we reduce memory usage for some use cases.
* feat: Change data type of catalog Postgres partition's sort_key from a string to an array of string
* test: add column with comma
* fix: use new protonuf field to avoid incompactible
* fix: ensure sort_key is an empty array rather than NULL
* refactor: address review comments
* refactor: address more comments
* chore: clearer comments
* chore: Update iox_catalog/migrations/20220607102200_change_sort_key_type_to_array.sql
* chore: Update iox_catalog/migrations/20220607102200_change_sort_key_type_to_array.sql
* fix: Rename migration so it will be applied after
Co-authored-by: Marko Mikulicic <mkm@influxdata.com>
* fix: do not return readable until a write is completely readable
* docs: Add diagram with partially buffered write
* refactor: account for actively buffering during update rather than fixup
* fix: fixup
* fix: use checked_sub
Co-authored-by: Marco Neumann <marco@crepererum.net>
* fix: checked_sub calculation
Co-authored-by: Marco Neumann <marco@crepererum.net>
Reduces memory usage in the ingester during persist operations by
streaming the results of the snapshot merge/sort/dedupe directly to
the parquet file.
Prior to this commit the output of the compact was buffered in memory
before being wrote to the parquet file.
* test: "optimize" ingesterrecord batches in query tests
It seems that I had the right idea in #4656 but wasn't able to trigger
https://github.com/influxdata/conductor/issues/955 because the query
tests do not "optimize" the record batches in the same way the actual
gRPC implementation does. If we apply the same transformation we indeed
end up with the same error.
* fix: all batches within the ingester flight response must have same schema
* refactor: simplify and reuse code
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
* feat: enable debugging of failed querier->ingester requests
- extend `query-ingester` CLI to allow usage of predicates
- on failed requests: log all information that required for the CLI
- test the "ingester fails" scenario
* test: explain
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
* docs: improve
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
* refactor: move b64 pred. serde into a single crate
Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>
Removes the min/max timestamp fields from the IoxMetadata proto
structure embedded within a Parquet file's metadata.
These values are redundant as they already exist within the Parquet
column statistics, and precluded streaming serialisation as these
removed min/max values were needed before serialising the file.
Remove the redundant row_count from the IoxMetadata structure that is
serialised into the Parquet file.
The reasoning is twofold:
* The Parquet file's native metadata already contains a row count
* Needing to know the number of rows up-front precludes streaming
Ok, so... this needed lots of... channels. Channels everywhere.
The stream method on TestWriteBufferStreamHandler previously assumed it
would only be called once. In a test where reset_to_earliest is called,
stream might be called again to get the reset stream.
We want to be able to control which of the streams gets which
operations, so that's why the macro now takes a vec of vec of
operations-- one vec of operations per expected call to stream, and the
stream will send all the operations in its vec.
The test thread needs to wait for the handler stream to consume the last
item from the last receiver stream, so when the
TestWriteBufferStreamHandler has set up the last expected call to
stream, pass back the last transmitter and have it wait until it's at
full expected capacity (which means all operations have been consumed by
the receiver).
Carol (Nichols || Goulding)
Jake Goulding
Andrew Lamb
kodiakhq[bot]
Marco Neumann
dependabot[bot]
Marko Mikulicic
Nga Tran
Raphael Taylor-Davies
Markus Westerlind
Dom Dwyer
Ryan Russell
Paul Dix