Library
150 results
ElasticsearchIntermediateDynamic mapping made your field text, so term queries miss, and mappings can't be changed in place. Use keyword fields, then reindex behind an alias with zero downtime.
#indexing#schema-design#troubleshooting
SQL ServerBeginnerError 8152 never said which column. Message 2628 on SQL Server 2019+ (or trace flag 460 on 2016/2017) names it. Then fix the data or the column, and watch for silent truncation.
#data-types#troubleshooting#t-sql
MongoDBBeginnerA server selection timeout means the driver can't reach the cluster; a pool timeout means you've run out of connections. How to diagnose each, including the serverless one-client-per-request trap.
#configuration#nosql#troubleshooting
MySQLBeginnerThe error points at a query that returned arbitrary values. Fix it with aggregates, primary-key grouping, ANY_VALUE() or window functions, and keep the SQL mode on.
#configuration#troubleshooting
RedisIntermediateOne slow command blocks every client. Find it with SLOWLOG and LATENCY DOCTOR, then replace KEYS with SCAN, split big keys, delete with UNLINK and tame fork pauses.
#caching#performance#troubleshooting
PostgreSQLBeginnerAfter an import or migration, every insert collides with an existing id because the sequence is behind. Reset one or all sequences, and prevent it with identity columns.
#data-types#migrations#troubleshooting
Neo4jBeginnerMERGE creates the whole pattern if any part is missing, and it can't stop concurrent duplicates without a constraint. How to merge correctly and clean up existing duplicates.
#concurrency#data-modeling#troubleshooting
Memcached is a fast, simple, multi-threaded blob cache; Redis is a data-structure server with persistence and replication. How to choose, and the pitfalls both share.
#caching#performance
ScyllaDBIntermediateA query without the partition key means a full cluster scan. When ALLOW FILTERING is fine, and how query-first tables, materialized views and global indexes fix the rest.
#data-modeling#indexing#nosql
TimescaleDBIntermediateSince TimescaleDB 2.13 continuous aggregates are materialized-only by default. Refresh policies, real-time aggregation and backfilling explained.
#maintenance#performance#troubleshooting
MilvusBeginner"Collection not loaded" and missing new rows come from load state and the default Bounded consistency level. How to fix both, and why flush() after every insert isn't the answer.
#configuration#troubleshooting#vector-search
