[RELEASE] ScyllaDB 2026.1.15

The ScyllaDB team is pleased to announce the release of ScyllaDB 2026.1.15, a production-ready patch release for ScyllaDB 2026.1 Long Term Support (LTS) Release.

More information on ScyllaDB Version Support Policy is available here.

Related Links

Bug Fixes

The following issues are fixed in this release.

Compaction

  • A newly produced SSTable could be picked up and removed by a second, unrelated compaction before the first compaction had finished with it, leading to an “Unable to remove input SSTable” failure. Such SSTables are now protected from concurrent compaction until the first compaction releases them, closing this window.
    scylladb#31718

  • nodetool stop COMPACTION no longer stopped major compactions, so you could only stop a running major compaction through the REST API. nodetool stop COMPACTION now stops both regular (automatic) and major compactions again. A new REGULAR type stops only regular compactions (nodetool stop REGULAR), and MAJOR stops only major compactions. nodetool and the REST API now share the same compaction-type validation, so nodetool rejects an invalid type right away with a clear error message.
    scylladb#32067

  • When a row had a shadowable tombstone and a regular tombstone with timestamps in different time windows, the timestamp-based splitting writer wrote both into the same time window. This could cause Time Window Compaction Strategy (TWCS) reshape to repeat endlessly. Each tombstone part is now written to the time window that matches its own timestamp, so reshape finishes as expected.
    scylladb#32202

  • If a node shut down early during startup, before compaction was enabled, the compaction manager’s shutdown path could leave configuration-update actions running. Shutdown now drains those actions, so a node stops cleanly after an early startup error.
    scylladb#32016, scylladb#32131

CQL

  • When a secondary index was temporarily out of sync with its base table, for example because of replication lag, a query that used the index could return rows that did not match the WHERE clause. ScyllaDB now checks every row read through an index against the base row before returning it, so indexed queries always return results that match their predicates. With this change, indexed queries with a single predicate use the same filtering path as indexed queries that already had more than one predicate.
    scylladb#31905

  • SELECT ... FROM MUTATION_FRAGMENTS(<table>) ignored GROUP BY and returned one row aggregated over all fragments instead of one row per group. Aggregates over more than one internal page also came back as several partial rows. For example, a COUNT over 25,000 fragments returned 10000, 10000, and 5000. Aggregate and GROUP BY queries on MUTATION_FRAGMENTS() now page and group correctly and return the expected results.
    scylladb#32011

Reliability

  • When commitlog replay encountered damaged segments, the cache for large (fragmented) commitlog entries could mix fragments from different shards, because fragment IDs are only unique per shard. The cache is now kept per shard, so large entries are replayed correctly.
    scylladb#32203

Stability

  • Some internal queries in system tables, such as lookups in system.repair_tasks and system.topology_requests, put parameter values directly into the query text. Each distinct value was cached as a separate prepared statement that was never evicted, so memory use grew steadily over time on long-running nodes. These queries now use bound parameters, which keeps the internal statement cache bounded.
    scylladb#32054

Storage

  • A backup task still waiting for the snapshot lock ignored abort requests. It kept running, then started uploading after the Scylla Manager run that created it had stopped. Over time this could build up a large queue of leftover backup tasks on a node. Queued backup tasks now respond to abort, so stopping a backup run also cancels its pending tasks.
    scylladb#31926

  • The S3 chunked download path retried errors immediately without pacing, and retried non-retryable errors indefinitely, which could leave a read waiting forever. It also didn’t report refusals to the request throttling controller. Downloads now use the standard retry strategy with backoff and a bounded number of retries, and report to the throttling controller. Reads from object storage now behave predictably when the endpoint is under pressure.
    scylladb#32053

  • Diagnostics are added for the case where an SSTable component fails to open with ENOENT. The error message no longer misleadingly says “Could not create” when the file was actually opened read-only, and additional information is logged to help distinguish a genuinely missing file from other causes, making it easier to triage issues involving unlinked or re-added SSTables.
    scylladb#31172

Tablets

  • During tablet migration, a tablet could move on to its next migration step before every node had confirmed it was done with the previous step. As a result, requests started under the previous step could still be running on replicas while new requests already used the tablet’s new location. For example, two lightweight transactions (LWT) on the same data could run at the same time instead of one after the other, and interrupted repair or streaming work could overlap with the next migration step. Tablet migration now waits until all nodes confirm before it moves on, so migration steps no longer overlap. Because of this, an unavailable node can hold up tablet migration until it is added to the ignore list.
    scylladb#31848, scylladb#26864, scylladb#26375

  • After a tablet split or merge, tablet migration could stall for up to 60 seconds (until the next periodic load-stats refresh) if a replica had not yet reported its size, for example a node that had just joined. A load-stats refresh is now triggered immediately in this case, so migration no longer waits.
    scylladb#32013

  • When the tablet load balancer planned a group of migrations, it checked the per-shard streaming limits using the number of migrations rather than their combined streaming weight. Rebuild migrations carry a higher weight, so source and target shards could go over tablet_streaming_read_concurrency_per_shard. The check now uses the combined weight, so streaming stays within the configured limits.
    scylladb#32206

  • A DROP KEYSPACE or DROP TABLE that ran while tablet migration was streaming sstables into the same table could lead to a node restart. Streaming now holds the table for the duration of the operation, so the drop waits for in-flight streaming to finish.
    scylladb#32014