Ferenc Szili

This change adds the design document for making truncate table safe with tablets #16411

docker/env/hydra.sh crashes on Python 3.10

4

### Issue description - [ ] This issue is a regression. - [x] It is unknown if this issue is a regression. #### Steps to Reproduce Running `./docker/env/hydra.sh list-images -c...

truncate: make TRUNCATE TABLE safe with tablets

30

Currently truncating a table works by issuing an RPC to all the nodes which call `database::truncate_table_on_all_shards()`, which makes sure that older writes are dropped. It works with tablets, but is...

logging: Add row count to large partition warning message

1

When writing large partitions, that is: partitions with size or row count above a configurable threshold, ScyllaDB outputs a warning to the log: ``` WARN ... large_data - Writing large...

Currently, TRUNCATE TABLE has different implementations for tablets and for vnodes. With vnodes, the request coordinator first checks the availability of all the nodes in the cluster, and fails the...

truncate: failing SCYLLA_ASSERT during database::truncate

2

While running rolling upgrade SCT test, Scylla 2025.2 crashes during TRUNCATE TABLE with a failing SCYLLA_ASSERT(): ``` 2025-04-11T06:27:02.274+00:00 rolling-upgrade--ubuntu-focal-db-node-8bb9bf91-1 !INFO | scylla[20949]: scylla: replica/database.cc:2679: future replica::database::truncate(db::system_keyspace &, column_family &, const...

test_tablet_rf_change: OperationTimedOut during "ALTER KEYSPACE test WITH replication"

8

test_tablet_rf_change with parameter direction=="up" failed in debug mode. A lot of tests on CI are running slow currently, but I want to open this in case this is encountered again....