# read-replica

A read replica is a read-only copy of a database instance that asynchronously replicates updates from a primary instance and serves read traffic.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## PostgreSQL 19 Released

DevFeed: [PostgreSQL 19 Released](<https://devfeed.tech/articles/postgresql-19-released-144325.md>)

Original publisher: [Read original article](<https://simonwillison.net/2026/Nov/19/postgresql-19/>)

Author: Simon Willison

Published: 2026-11-19T20:43:57Z

Content type: release

Language: en

Sources: [Simon Willison's Weblog](<https://devfeed.tech/sources/simon-willison-s-weblog.md>)

Topics: [read-replica](<https://devfeed.tech/topics/read-replica.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-security-research](<https://devfeed.tech/tags/ai-security-research.md>), [ai-security-research-51](<https://devfeed.tech/tags/ai-security-research-51.md>), [applications](<https://devfeed.tech/tags/applications.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cookie](<https://devfeed.tech/tags/cookie.md>), [feature](<https://devfeed.tech/tags/feature.md>), [lag](<https://devfeed.tech/tags/lag.md>), [mediawiki](<https://devfeed.tech/tags/mediawiki.md>), [movie](<https://devfeed.tech/tags/movie.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [mysql-82](<https://devfeed.tech/tags/mysql-82.md>), [mysql-replication](<https://devfeed.tech/tags/mysql-replication.md>), [new-feature](<https://devfeed.tech/tags/new-feature.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [postgresql-161](<https://devfeed.tech/tags/postgresql-161.md>), [postgresql-19](<https://devfeed.tech/tags/postgresql-19.md>), [replica](<https://devfeed.tech/tags/replica.md>), [replication](<https://devfeed.tech/tags/replication.md>), [replication-32](<https://devfeed.tech/tags/replication-32.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [scaling-149](<https://devfeed.tech/tags/scaling-149.md>), [standby](<https://devfeed.tech/tags/standby.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>), [wal](<https://devfeed.tech/tags/wal.md>), [wikipedia](<https://devfeed.tech/tags/wikipedia.md>), [wikipedia-79](<https://devfeed.tech/tags/wikipedia-79.md>)

### AI overview

PostgreSQL 19 adds `WAIT FOR LSN` for applications that read from standbys: a session can wait for a replica to replay WAL through a specified log sequence number before querying, enabling read-your-writes consistency after a write on the primary. The excerpt's example captures the primary's WAL position after an update, saves it for a later request, then waits for that position before reading from the standby. An optional timeout causes an error if the replica has not caught up in time. The article relates this to MediaWiki's existing MySQL approach, which saves a primary binlog position and waits for the replica before reads, using `WAIT_FOR_EXECUTED_GTID_SET()` or, as a fallback, `MASTER_POS_WAIT()`.

### Source excerpt

PostgreSQL 19 Released Most years PostgreSQL ships a new release in September, but this year's release came two months later thanks mainly to the flurry of issues that the PostgreSQL team needed to address as part of the ongoing AI-reported vulnerability apocalypse. This release was definitely worth the wait. My favorite new feature is this one: For applications that send reads to standbys, PostgreSQL 19 adds "read-your-writes" consistency with the new WAIT FOR LSN command. A session on a standby can use WAIT FOR LSN to wait until the standby has replayed the write-ahead log (WAL) up to a specific log sequence number (LSN) before running a query, so the application can read data it just wrote on the primary. In practice, that looks like this: UPDATE movie SET genre = 'Dramatic' WHERE genre = 'Drama'; SELECT pg_current_wal_insert_lsn(); That returns a value like 0/0306EE20. Store that somewhere (a signed cookie would work) such that on a subsequent request your application can do: WAIT FOR LSN '0/0306EE20' WITH (TIMEOUT '0.1s'); SELECT * FROM movie WHERE genre = 'Drama'; The timeout is optional - if you provide one you'll get an error if the replica you are talking to doesn't catch up within the specified time. I first encountered this pattern in Wikipedia, which has a hand-rolled version of this built against MySQL replication: In addition to this, MediaWiki attempts to ensure that the user sees events occurring on the wiki in chronological order. A few seconds of lag can be tolerated, as long as the user sees a consistent picture from subsequent requests. This is done by saving the primary binlog position in the session, and then at the start of each request, waiting for the replica to catch up to that position before doing any reads from it. Here's the relevant MediaWiki code, which these days uses MySQL's WAIT_FOR_EXECUTED_GTID_SET() function and falls back to MASTER_POS_WAIT(). Via Lobste.rs Tags: mysql, postgresql, replication, scaling, wikipedia, ai-security-r

## WAL-G 3.0.10 released

DevFeed: [WAL-G 3.0.10 released](<https://devfeed.tech/articles/wal-g-3-0-10-released-141067.md>)

Original publisher: [Read original article](<https://www.postgresql.org/about/news/wal-g-3010-released-3448/>)

Published: 2026-11-18T00:00:00Z

Content type: release

Language: en

Sources: [PostgreSQL news](<https://devfeed.tech/sources/postgresql-news.md>)

Topics: [releases](<https://devfeed.tech/topics/releases.md>), [backups](<https://devfeed.tech/topics/backups.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [MariaDB](<https://devfeed.tech/topics/mariadb.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Redis](<https://devfeed.tech/topics/redis.md>)

Tags: [accounting](<https://devfeed.tech/tags/accounting.md>), [apache-cloudberry](<https://devfeed.tech/tags/apache-cloudberry.md>), [aws-sdk](<https://devfeed.tech/tags/aws-sdk.md>), [backup](<https://devfeed.tech/tags/backup.md>), [backup-and-recovery](<https://devfeed.tech/tags/backup-and-recovery.md>), [backup-storage](<https://devfeed.tech/tags/backup-storage.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [copy](<https://devfeed.tech/tags/copy.md>), [greenplum](<https://devfeed.tech/tags/greenplum.md>), [mariadb](<https://devfeed.tech/tags/mariadb.md>), [mongodb](<https://devfeed.tech/tags/mongodb.md>), [object-storage](<https://devfeed.tech/tags/object-storage.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [redis](<https://devfeed.tech/tags/redis.md>), [release](<https://devfeed.tech/tags/release.md>), [replication](<https://devfeed.tech/tags/replication.md>), [restore](<https://devfeed.tech/tags/restore.md>), [s3](<https://devfeed.tech/tags/s3.md>), [v2](<https://devfeed.tech/tags/v2.md>)

### AI overview

WAL-G 3.0.10 adds `wal-send`, which serves archived WAL through PostgreSQL's streaming replication protocol so standbys and `pg_receivewal` can read from object storage and follow new segments. `backup-push --wait-for-archive` waits for required WAL segments to reach the selected backup storage before marking a backup complete. The release also adds AWS SDK v2 S3 support with configurable timeouts, retryable errors, and checksums, plus optional server-side S3 copying and configurable transfer concurrency. Greenplum and Cloudberry updates include shared-storage accounting and fixes to segment backup, restore, and copy. Permanent backups remain protected from retention-based deletion, but an explicit delete target can remove them.

### Source excerpt

The WAL-G team is pleased to announce the release of WAL-G 3.0.10. WAL-G is an open-source backup and recovery tool for PostgreSQL, Apache Cloudberry, MySQL/MariaDB, MongoDB, Redis, and other databases. The main new PostgreSQL feature is wal-send. It serves archived WAL over PostgreSQL's streaming replication protocol, allowing standbys and pg_receivewal to read WAL directly from object storage and follow new segments as they arrive. Other PostgreSQL improvements include: Continuous WAL archival in daemon mode: the pipeline keeps working between requests, and retained lookahead reduces directory scanning (#2697, #2698). Optional backup-push --wait-for-archive: WAL-G waits for all WAL segments required by the backup to reach the selected backup storage before publishing the backup as complete (#2722). Backup-name reservation before upload, protecting against concurrent reuse of the same name (#2707). WAL prefetch files are published only after a successful download (#2696), and WAL source files are closed after upload failures (#2695). Standard X/X LSN notation in catchup-push --from-lsn (#2643). S3 support now uses AWS SDK v2 and adds configurable request timeouts, retryable error codes, and checksum settings. Backup copying can use server-side S3 copy and configurable transfer concurrency. Greenplum and Cloudberry improvements include per-backup accounting for shared AO/AOCS/PAX storage and fixes to segment backup, restore, and copy operations. One behavior change to note: permanent backups remain protected from retention-based deletion, but an explicit delete target can now remove them (#2808). WAL-G v3.0.10 is available for download, along with the full release notes, on our GitHub releases page. Have a nice day!

## Distribute Amazon RDS traffic between read replicas using Network Load Balancer

DevFeed: [Distribute Amazon RDS traffic between read replicas using Network Load Balancer](<https://devfeed.tech/articles/distribute-amazon-rds-traffic-between-read-replicas-using-network-load-balancer-138652.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/database/distribute-amazon-rds-traffic-between-read-replicas-using-network-load-balancer/>)

Author: Vijay Chakraborty

Published: 2026-11-17T15:51:53Z

Content type: tutorial

Language: en

Sources: [AWS Database Blog](<https://devfeed.tech/sources/aws-database-blog.md>)

Topics: [Load Balancing](<https://devfeed.tech/topics/load-balancing.md>), [Amazon RDS](<https://devfeed.tech/topics/amazon-rds.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Virtual Private Network](<https://devfeed.tech/topics/vpn.md>), [Cloud Networking](<https://devfeed.tech/topics/cloud-networking.md>), [Latency](<https://devfeed.tech/topics/latency.md>), [connection draining](<https://devfeed.tech/topics/connection-draining.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [affinity](<https://devfeed.tech/tags/affinity.md>), [amazon-aurora](<https://devfeed.tech/tags/amazon-aurora.md>), [amazon-cloudwatch](<https://devfeed.tech/tags/amazon-cloudwatch.md>), [amazon-ec2](<https://devfeed.tech/tags/amazon-ec2.md>), [amazon-elastic-container-service](<https://devfeed.tech/tags/amazon-elastic-container-service.md>), [amazon-elastic-kubernetes-service-amazon-eks](<https://devfeed.tech/tags/amazon-elastic-kubernetes-service-amazon-eks.md>), [amazon-eventbridge](<https://devfeed.tech/tags/amazon-eventbridge.md>), [amazon-rds](<https://devfeed.tech/tags/amazon-rds.md>), [amazon-route-53](<https://devfeed.tech/tags/amazon-route-53.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [aws-cloudformation](<https://devfeed.tech/tags/aws-cloudformation.md>), [aws-direct-connect](<https://devfeed.tech/tags/aws-direct-connect.md>), [aws-lambda](<https://devfeed.tech/tags/aws-lambda.md>), [aws-privatelink](<https://devfeed.tech/tags/aws-privatelink.md>), [aws-site-to-site-vpn](<https://devfeed.tech/tags/aws-site-to-site-vpn.md>), [elastic-load-balancing](<https://devfeed.tech/tags/elastic-load-balancing.md>), [health](<https://devfeed.tech/tags/health.md>), [lambda](<https://devfeed.tech/tags/lambda.md>), [load-balancer](<https://devfeed.tech/tags/load-balancer.md>), [read-replicas](<https://devfeed.tech/tags/read-replicas.md>), [technical-how-to](<https://devfeed.tech/tags/technical-how-to.md>), [traffic](<https://devfeed.tech/tags/traffic.md>)

### AI overview

An NLB can provide one stable endpoint for distributing connections across Amazon RDS read replicas, with health checks and Availability Zone affinity. An EventBridge rule invokes a Lambda function every minute to resolve replica DNS names and synchronize their IPs into the NLB target group; S3 JSON state and retry logic help avoid premature deregistration during transient DNS failures. With zonal DNS affinity enabled and cross-zone load balancing disabled, traffic can remain AZ-local in steady state, reducing inter-AZ transfer. The endpoint also supports hybrid access over Direct Connect or Site-to-Site VPN without PrivateLink or Route 53 Resolver endpoints when a custom domain is unnecessary. The approach works with RDS engines that support read replicas, but the NLB balances connections rather than queries, so long-lived connections can limit redistribution. In EKS, CoreDNS placement can undermine zonal affinity; NodeLocal DNSCache is suggested. Deregistration delay allows existing connections to drain.

### Source excerpt

Learn how to put Amazon RDS read replicas behind a Network Load Balancer with health-aware, Availability Zone affinity routing. An AWS Lambda function keeps the load balancer target group in sync with replica IP addresses, giving clients a single stable endpoint over Direct Connect or Site-to-Site VPN without AWS PrivateLink overhead.

## More load is faster: Tuning Postgres read replicas

DevFeed: [More load is faster: Tuning Postgres read replicas](<https://devfeed.tech/articles/more-load-is-faster-tuning-postgres-read-replicas-140981.md>)

Original publisher: [Read original article](<https://clickhouse.com/blog/tuning-postgres-read-replicas>)

Author: Colton Shepard

Published: 2026-11-17T00:00:00Z

Content type: article

Language: en

Sources: [ClickHouse Blog](<https://devfeed.tech/sources/clickhouse-blog.md>)

Topics: [read-replica](<https://devfeed.tech/topics/read-replica.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [lag](<https://devfeed.tech/tags/lag.md>), [latency](<https://devfeed.tech/tags/latency.md>), [postgres](<https://devfeed.tech/tags/postgres.md>), [read-replicas](<https://devfeed.tech/tags/read-replicas.md>), [settings](<https://devfeed.tech/tags/settings.md>), [testing](<https://devfeed.tech/tags/testing.md>), [tradeoffs](<https://devfeed.tech/tags/tradeoffs.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [wal](<https://devfeed.tech/tags/wal.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

The tests found that ordinary traffic can keep Postgres read replica lag to a few seconds, while very low write volume can leave replicas waiting for WAL segments to switch--up to the default 60-second archive timeout. In the tested archive-based setup, `archive_timeout=5s` and `wal_retrieve_retry_interval=500ms` are offered as a relatively safe starting point; polling faster gave diminishing returns because WAL-G fetches took about 0.8 seconds. Shorter timeouts increase object-storage re-fetching, replica CPU use, and network costs with synchronous HA. Forcing WAL switches reduced primary throughput by about 15% and added more overhead. Streaming replication was faster, but the article cites vacuum, disk, primary-load, and failover risks as reasons not to make it the default for read replicas.

### Source excerpt

Testing Postgres read replicas reveals why more write traffic can reduce lag, how WAL settings affect latency, and where aggressive tuning introduces tradeoffs.

## Gartner Analyst: "Your Next Database User Won't Be a Person"

DevFeed: [Gartner Analyst: "Your Next Database User Won't Be a Person"](<https://devfeed.tech/articles/gartner-analyst-your-next-database-user-won-t-be-a-person-137751.md>)

Original publisher: [Read original article](<https://flashdba.com/2026/11/16/gartner-analyst-your-next-database-user-wont-be-a-person/>)

Author: flashdba

Published: 2026-11-16T17:27:25Z

Content type: opinion

Language: en

Sources: [Flashdba](<https://devfeed.tech/sources/flashdba.md>)

Topics: [Database](<https://devfeed.tech/topics/database.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [database-scalability](<https://devfeed.tech/topics/database-scalability.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [database monitoring](<https://devfeed.tech/topics/database-monitoring.md>), [connection pool](<https://devfeed.tech/topics/connection-pool.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [after](<https://devfeed.tech/tags/after.md>), [age-of-ai](<https://devfeed.tech/tags/age-of-ai.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [alarm](<https://devfeed.tech/tags/alarm.md>), [budget](<https://devfeed.tech/tags/budget.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [database](<https://devfeed.tech/tags/database.md>), [databases](<https://devfeed.tech/tags/databases.md>), [databases-and-agentic-ai](<https://devfeed.tech/tags/databases-and-agentic-ai.md>), [first](<https://devfeed.tech/tags/first.md>), [gartner](<https://devfeed.tech/tags/gartner.md>), [governance](<https://devfeed.tech/tags/governance.md>), [inferencing](<https://devfeed.tech/tags/inferencing.md>), [next](<https://devfeed.tech/tags/next.md>), [operational](<https://devfeed.tech/tags/operational.md>), [performance](<https://devfeed.tech/tags/performance.md>), [user](<https://devfeed.tech/tags/user.md>)

### AI overview

The excerpt argues that gradual AI agent adoption is more likely to expose budget and governance problems before cloud operational databases fail. Much agent activity is read-heavy, and cloud platforms can absorb added load with replicas, serverless compute, or higher provisioned limits; this can keep queries and latency healthy while shifting the strain into infrastructure costs, often charged to teams that did not deploy the agents. Database failure can still occur beyond elasticity limits, but the article says organizations may pay for the approach well before that point. Governance is a separate, pre-existing gap: an agent using a service account may be identifiable only as that account, leaving records without attribution to the agent, its initiator, authority, or purpose. Higher volume multiplies these gaps without necessarily triggering an alert. The excerpt also notes that database choices are moving from central DBA teams to product and engineering leads as DBA headcount declines, while system owners may still receive the bill and audit findings.

### Source excerpt

A Gartner analyst asked what AI agents would break first. On cloud operational databases the budget breaks before the database - and governance already has.

## Don't let overload protection be the enemy of scaling

DevFeed: [Don't let overload protection be the enemy of scaling](<https://devfeed.tech/articles/don-t-let-overload-protection-be-the-enemy-of-scaling-131132.md>)

Original publisher: [Read original article](<https://medium.com/whatnot-engineering/dont-let-overload-protection-be-the-enemy-of-scaling-7f59abbcb070?source=rss----162aeca881b0---4>)

Author: Whatnot Engineering

Published: 2026-11-12T13:05:42Z

Content type: article

Language: en

Sources: [Whatnot Engineering](<https://devfeed.tech/sources/whatnot-engineering.md>)

Topics: [consistency](<https://devfeed.tech/topics/consistency.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [rate-limiting](<https://devfeed.tech/topics/rate-limiting.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [backend](<https://devfeed.tech/tags/backend.md>), [bottleneck](<https://devfeed.tech/tags/bottleneck.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [database](<https://devfeed.tech/tags/database.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [graphql](<https://devfeed.tech/tags/graphql.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [ip](<https://devfeed.tech/tags/ip.md>), [limit](<https://devfeed.tech/tags/limit.md>), [load](<https://devfeed.tech/tags/load.md>), [protection](<https://devfeed.tech/tags/protection.md>), [query](<https://devfeed.tech/tags/query.md>), [queue](<https://devfeed.tech/tags/queue.md>), [redis](<https://devfeed.tech/tags/redis.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [request](<https://devfeed.tech/tags/request.md>), [rest](<https://devfeed.tech/tags/rest.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [site-reliability-engineer](<https://devfeed.tech/tags/site-reliability-engineer.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [sre](<https://devfeed.tech/tags/sre.md>), [traffic](<https://devfeed.tech/tags/traffic.md>)

### AI overview

Whatnot replaced its concurrency limiter's precise, synchronous Redis counting after the old design remained a bottleneck even across nine primaries. The new distributed semaphore tolerates slightly stale counts: pods check current and previous window counters on replicas, accumulate request deltas locally, and batch INCRBY updates to primaries every 100 ms. Window keys can land on different shards, spreading hot traffic; Redis LRU eviction replaces per-request TTL updates. The limiter still throttles slow request types, returning 503 when a type exceeds its cap and 429 when an IP does. It can miss very short requests that start and finish between syncs, a tradeoff the team accepted because slow requests are the overload concern. In pre-Black-Friday 2025 load tests, the hottest primary across eight shards stayed below 50% CPU; ahead of Black Friday 2026, the semaphore handled four times the traffic that would overwhelm the old limiter, with headroom. Limits remain static; an adaptive controller is planned.

### Source excerpt

Marcin Mierzejewski | Reliability Platform (Poland) The concurrency limiter in front of our main API at Whatnot has one job: protect the backend from overload. As we scaled, though, it became the component most likely to fall over under a traffic surge. We had grown its Redis fleet from one instance to nine, and the hottest instance still peaked at 92% CPU in load tests. We had even configured that instance to send 60% of its traffic elsewhere, but it still stayed the hottest. Keeping track of every request was a losing effort. The way out was to stop counting so precisely. Concurrency limiting in a nutshell In short, we limit concurrency rather than rate. We count in-flight requests per request type (GraphQL operation names or REST endpoints), so a request type that slows down gets throttled harder. Why count requests in flight instead of requests per second? The number in flight is roughly the arrival rate times how long each request takes. If a database query slows from, say, 50 ms to 500 ms, the same traffic keeps ten times as many requests in flight. Each one holds a worker and a database connection while it waits, so the pile-up slows everything else down, which makes requests slower still. A rate limiter doesn't notice, because the request rate hasn't changed. A concurrency limiter does. The slow request type hits its cap on in-flight requests, and the extras get a fast 503 instead of joining the queue. That request type sheds load early, and the rest of the API keeps its workers and connections. A separate per-IP cap stops any one client from taking more than its share. For simplicity, we'll only talk about request-type concurrency limiting, unless stated otherwise. How the limiter became the bottleneck In 2024, my colleague Karol Gil wrote about how we initially built the limiter. That design worked, and a few weeks later it throttled a query storm that nearly tripled request volume. However, it paid for a precise, synchronous count on every request. Each A

## Managed cross-Region disaster recovery for Amazon OpenSearch Service

DevFeed: [Managed cross-Region disaster recovery for Amazon OpenSearch Service](<https://devfeed.tech/articles/managed-cross-region-disaster-recovery-for-amazon-opensearch-service-129794.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/big-data/managed-cross-region-disaster-recovery-for-amazon-opensearch-service/>)

Author: Cedric Pelvet

Published: 2026-11-12T00:45:45Z

Content type: article

Language: en

Sources: [AWS Big Data Blog](<https://devfeed.tech/sources/aws-big-data-blog.md>)

Topics: [Amazon OpenSearch Service](<https://devfeed.tech/topics/amazon-opensearch-service.md>), [Disaster Recovery](<https://devfeed.tech/topics/disaster-recovery.md>), [opensearch](<https://devfeed.tech/topics/opensearch.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Amazon CloudWatch](<https://devfeed.tech/topics/amazon-cloudwatch.md>)

Tags: [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [amazon-cloudwatch](<https://devfeed.tech/tags/amazon-cloudwatch.md>), [amazon-opensearch-service](<https://devfeed.tech/tags/amazon-opensearch-service.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [api](<https://devfeed.tech/tags/api.md>), [aws-command-line-interface-aws-cli](<https://devfeed.tech/tags/aws-command-line-interface-aws-cli.md>), [aws-key-management-service-aws-kms](<https://devfeed.tech/tags/aws-key-management-service-aws-kms.md>), [aws-management-console](<https://devfeed.tech/tags/aws-management-console.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [business-continuity](<https://devfeed.tech/tags/business-continuity.md>), [cross](<https://devfeed.tech/tags/cross.md>), [disaster-recovery](<https://devfeed.tech/tags/disaster-recovery.md>), [failover](<https://devfeed.tech/tags/failover.md>), [foundational-100](<https://devfeed.tech/tags/foundational-100.md>), [region](<https://devfeed.tech/tags/region.md>), [rpo](<https://devfeed.tech/tags/rpo.md>), [rto](<https://devfeed.tech/tags/rto.md>)

### AI overview

Amazon OpenSearch Service now provides managed cross-Region disaster recovery for domains running OpenSearch 3.7 or later. Pair same-account domains in different Regions with matching capacity; the service copies existing data and continuously replicates changes, with both domains readable and only the primary writable. Planned switchover coordinates a role swap with zero data loss; failover promotes the secondary without waiting, and may lose writes still in replication transit, bounded by replication lag. The stated targets are RPO under one minute and RTO under 20 minutes; example tests saw 8-10 seconds of loss on failover and writes resume in about a minute. Applications must redirect writes after promotion. Index templates, settings, roles, and ISM policies are not replicated and must be managed on both domains. UltraWarm, Writable Warm, and the OpenSearch Optimized engine for log analytics are unsupported. There is no additional service charge, though secondary-domain capacity and cross-Region transfer are billed.

### Source excerpt

Amazon OpenSearch Service now offers managed cross-Region disaster recovery for domains on OpenSearch 3.7 or later. It replicates data from a primary domain to a secondary domain in another Region with API-driven switchover and failover, providing an RPO under 1 minute and an RTO under 20 minutes.

## Laravel Read Replicas: Check if a Replica Has Caught Up

DevFeed: [Laravel Read Replicas: Check if a Replica Has Caught Up](<https://devfeed.tech/articles/laravel-read-replicas-check-if-a-replica-has-caught-up-131217.md>)

Original publisher: [Read original article](<https://laravel-news.com/laravel-read-replica-commit-ids>)

Author: Paul Redmond

Published: 2026-11-11T23:04:37Z

Content type: article

Language: en

Sources: [Laravel](<https://devfeed.tech/sources/laravel.md>)

Topics: [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Laravel](<https://devfeed.tech/topics/laravel.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [MariaDB](<https://devfeed.tech/topics/mariadb.md>)

Tags: [commit](<https://devfeed.tech/tags/commit.md>), [database](<https://devfeed.tech/tags/database.md>), [ids](<https://devfeed.tech/tags/ids.md>), [job](<https://devfeed.tech/tags/job.md>), [lag](<https://devfeed.tech/tags/lag.md>), [laravel](<https://devfeed.tech/tags/laravel.md>), [laravel-tutorials](<https://devfeed.tech/tags/laravel-tutorials.md>), [mariadb](<https://devfeed.tech/tags/mariadb.md>), [middleware](<https://devfeed.tech/tags/middleware.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [read-replicas](<https://devfeed.tech/tags/read-replicas.md>), [replica](<https://devfeed.tech/tags/replica.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

Laravel 13.40 adds commit ID checks that let applications route reads based on whether a replica has applied a write, rather than relying only on a fixed lag window. `getLastCommitId()` reads the primary's position; `hasCommitId()` checks the server used for reads and can check once or wait up to a whole-number timeout; `mergeCommitIds()` combines positions across writes. The article gives two patterns: session middleware stores pending IDs and uses the primary until a replica catches up, while queued jobs receive an ID, wait briefly, then fall back to the primary. Capture the ID after a transaction commits. The middleware adds a replica query per pending request and a primary query per writing request; a job should receive an ID rather than a serialized model to avoid loading it from a lagging replica before the check. Requirements vary: MySQL needs binary logging and GTIDs, MariaDB needs binary logging, SQLite and SQL Server are unsupported, and PostgreSQL 18 and earlier cannot use the wait option. Replica checks follow Laravel's connection routing and must be repeated on each request.

### Source excerpt

Laravel 13.40 adds commit IDs to MySQL, MariaDB, and PostgreSQL connections, so you can check whether a read replica has a write before you read from it. The post Laravel Read Replicas: Check if a Replica Has Caught Up appeared first on Laravel News. Join the Laravel Newsletter to get Laravel articles like this directly in your inbox.

## CoreWeave Database Is Here: Managed PostgreSQL Where Your AI Runs

DevFeed: [CoreWeave Database Is Here: Managed PostgreSQL Where Your AI Runs](<https://devfeed.tech/articles/coreweave-database-is-here-managed-postgresql-where-your-ai-runs-129191.md>)

Original publisher: [Read original article](<https://wf.coreweave.com/blog/coreweave-database-is-here-managed-postgresql-where-your-ai-runs>)

Author: CoreWeave

Published: 2026-11-11T16:05:56Z

Content type: release

Language: en

Sources: [Coreweave Blog Posts](<https://devfeed.tech/sources/coreweave-blog-posts.md>)

Topics: [PostgreSQL](<https://devfeed.tech/topics/postgresql.md>), [CloudNativePG](<https://devfeed.tech/topics/cloudnativepg.md>), [Managed Services](<https://devfeed.tech/topics/managed-services.md>), [Database](<https://devfeed.tech/topics/database.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [blog](<https://devfeed.tech/tags/blog.md>), [cks](<https://devfeed.tech/tags/cks.md>), [cloudnativepg](<https://devfeed.tech/tags/cloudnativepg.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [coreweave](<https://devfeed.tech/tags/coreweave.md>), [database](<https://devfeed.tech/tags/database.md>), [managed](<https://devfeed.tech/tags/managed.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>)

### AI overview

CoreWeave Database provides managed, PostgreSQL-compatible databases inside a CoreWeave Kubernetes Service (CKS) cluster, alongside AI workloads. Databases run on CPU nodes, store live data on CoreWeave Distributed File Storage, and use a primary with replicas across hosts for automatic failover. The service handles replication, patching, monitoring, and continuous backups, with point-in-time recovery within the retention window. Teams retain responsibility for data, schemas, queries, and database users; they control access through PostgreSQL roles and Kubernetes network policies. Connections use in-cluster endpoints, and the included pgvector extension supports embedding queries. The service uses CloudNativePG for PostgreSQL orchestration and adds CoreWeave console and Kubernetes workflows and managed storage and backups. The excerpt says existing PostgreSQL databases can be migrated using native tools, but gives no migration procedure or performance measurements.

### Source excerpt

CoreWeave Database is now generally available. Run managed PostgreSQL inside your CKS cluster, operated by CoreWeave, right next to your AI workloads.

## How a broken release name overloaded our symbolication database

DevFeed: [How a broken release name overloaded our symbolication database](<https://devfeed.tech/articles/how-a-broken-release-name-overloaded-our-symbolication-database-124324.md>)

Original publisher: [Read original article](<https://blog.sentry.io/broken-release-database-load/>)

Author: Michael Warkentin

Published: 2026-11-09T09:00:00Z

Content type: news

Language: en

Sources: [Sentry](<https://devfeed.tech/sources/sentry-blog.md>)

Topics: [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [Database](<https://devfeed.tech/topics/database.md>), [rdbms](<https://devfeed.tech/topics/rdbms.md>), [database monitoring](<https://devfeed.tech/topics/database-monitoring.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>)

Tags: [build](<https://devfeed.tech/tags/build.md>), [ci](<https://devfeed.tech/tags/ci.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [customer-data](<https://devfeed.tech/tags/customer-data.md>), [database](<https://devfeed.tech/tags/database.md>), [index](<https://devfeed.tech/tags/index.md>), [name](<https://devfeed.tech/tags/name.md>), [release](<https://devfeed.tech/tags/release.md>), [url](<https://devfeed.tech/tags/url.md>), [var](<https://devfeed.tech/tags/var.md>)

### AI overview

A broken release name caused millions of source-map URL index rows to accumulate, and an unbounded lookup repeatedly scanned them, driving database load and symbolication timeouts. Sentry capped each URL lookup at 1,000,000 rows and cached misses; about 0.5% of lookups hit the cap, while the busiest host's CPU fell from 76% to 19% and timeouts fell 98%. Other fixes included splitting a costly debug ID join, updating bundle dates without rewriting index rows, excluding generated file names from the URL index, and disabling Postgres JIT for a legacy lookup. The article says the unexpanded environment variable came from a customer build configuration and can occur with any CI provider. It recommends checking Sentry release names and ensuring variables are set and expanded; SDK 11.9.0 also ignores a commit variable whose value is its own name.

### Source excerpt

An unexpanded env var filled one release with 44M URL index rows. Capping lookups cut database CPU from 76% to 19% and symbolication timeouts by 98%.

## Database Sharding Clearly Explained

DevFeed: [Database Sharding Clearly Explained](<https://devfeed.tech/articles/database-sharding-clearly-explained-121476.md>)

Original publisher: [Read original article](<https://blog.levelupcoding.com/p/database-sharding-clearly-explained>)

Author: Nikki Siapno

Published: 2026-11-06T15:03:19Z

Content type: article

Language: en

Sources: [Level Up Coding System Design Newsletter](<https://devfeed.tech/sources/level-up-coding-system-design-newsletter.md>)

Topics: [sharding](<https://devfeed.tech/topics/sharding.md>), [database-scalability](<https://devfeed.tech/topics/database-scalability.md>), [table-partitioning](<https://devfeed.tech/topics/table-partitioning.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [connection pool](<https://devfeed.tech/topics/connection-pool.md>), [Memcached](<https://devfeed.tech/topics/memcached.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-engineering](<https://devfeed.tech/tags/ai-engineering.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [application](<https://devfeed.tech/tags/application.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automate](<https://devfeed.tech/tags/automate.md>), [automation](<https://devfeed.tech/tags/automation.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [cache](<https://devfeed.tech/tags/cache.md>), [caching](<https://devfeed.tech/tags/caching.md>), [connection-pooling](<https://devfeed.tech/tags/connection-pooling.md>), [consistent-hashing](<https://devfeed.tech/tags/consistent-hashing.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [data](<https://devfeed.tech/tags/data.md>), [data-movement](<https://devfeed.tech/tags/data-movement.md>), [database](<https://devfeed.tech/tags/database.md>), [sharding](<https://devfeed.tech/tags/sharding.md>)

### AI overview

Sharding splits data across multiple database servers to distribute storage and workload when one server remains a constraint after simpler measures such as query optimization, caching, vertical scaling, read replicas, and partitioning. Horizontal sharding divides rows while keeping the same schema; vertical sharding separates tables or workloads. The key design challenge is choosing a shard key that spreads both data and traffic evenly. Queries and transactions spanning shards require coordination and add latency and complexity, while changing shard counts can require difficult data movement. The text recommends keeping common operations within one shard, planning for rebalancing, monitoring shards individually, automating maintenance, and testing realistic workloads. Sharding can extend scalability, but replaces a single-server bottleneck with distributed-system and operational challenges.

### Source excerpt

The hardest part of sharding isn't splitting the data.

## Safe database change at fleet scale: canaries, maintenance waves, and recovery gates

DevFeed: [Safe database change at fleet scale: canaries, maintenance waves, and recovery gates](<https://devfeed.tech/articles/safe-database-change-at-fleet-scale-canaries-maintenance-waves-and-recovery-gates-121139.md>)

Original publisher: [Read original article](<https://severalnines.com/blog/safe-database-change-at-fleet-scale-canaries-maintenance-waves-and-recovery-gates/>)

Author: Sebastian Insausti

Published: 2026-11-06T11:37:15Z

Content type: tutorial

Language: en

Sources: [SeveralNines](<https://devfeed.tech/sources/severalnines.md>)

Topics: [Insight for DBAs](<https://devfeed.tech/topics/insight-for-dbas.md>), [ClusterControl](<https://devfeed.tech/topics/clustercontrol.md>), [canary release](<https://devfeed.tech/topics/canary-release.md>), [Disaster Recovery](<https://devfeed.tech/topics/disaster-recovery.md>), [Database](<https://devfeed.tech/topics/database.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>)

Tags: [canary](<https://devfeed.tech/tags/canary.md>), [change](<https://devfeed.tech/tags/change.md>), [clustercontrol](<https://devfeed.tech/tags/clustercontrol.md>), [configuration-management](<https://devfeed.tech/tags/configuration-management.md>), [custom](<https://devfeed.tech/tags/custom.md>), [database](<https://devfeed.tech/tags/database.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [maintenance](<https://devfeed.tech/tags/maintenance.md>), [modern-data-stack-operations](<https://devfeed.tech/tags/modern-data-stack-operations.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [scale](<https://devfeed.tech/tags/scale.md>), [scripts](<https://devfeed.tech/tags/scripts.md>)

### AI overview

For a PostgreSQL minor update across a fleet, the article uses preflight checks on every target, one representative canary, and sequential waves that stop at the first failed gate. In its three-cluster lab, a missing package in the approved mirror caused preflight to fail while all databases stayed on the old version; after the mirror was fixed, the canary and both wave targets passed acceptance. The tested canary had a brief write interruption, measured by one-second probe timestamps. ClusterControl runs upgrade jobs, while preflight, sequencing, validation, and go/no-go decisions rely on custom scripts or people; its upgrade job has no rollback operation and may leave a failed node for manual recovery. The lab also tested reinstalling the prior minor build, subject to release-note review and acceptance checks. Results are a method demonstration, not evidence of fleet-scale capacity or major-upgrade safety.

### Source excerpt

How do you change a database fleet without creating a fleet-wide incident? This PostgreSQL walkthrough demonstrates preflight checks, a representative canary, sequential maintenance waves, and a tested recovery path--with clear boundaries between ClusterControl jobs, custom scripts, and manual decisions. The post Safe database change at fleet scale: canaries, maintenance waves, and recovery gates appeared first on Severalnines.

## WooCommerce Analytics with DuckDB

DevFeed: [WooCommerce Analytics with DuckDB](<https://devfeed.tech/articles/woocommerce-analytics-with-duckdb-119603.md>)

Original publisher: [Read original article](<https://www.percona.com/blog/woocommerce-analytics-with-duckdb/>)

Author: Juan Arruti

Published: 2026-11-05T16:31:19Z

Content type: article

Language: en

Sources: [Percona](<https://devfeed.tech/sources/blog-percona.md>)

Topics: [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Percona Server for MySQL](<https://devfeed.tech/topics/percona-server-for-mysql.md>), [data type](<https://devfeed.tech/topics/data-type.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [Percona](<https://devfeed.tech/topics/percona.md>)

Tags: [analytics](<https://devfeed.tech/tags/analytics.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [columnar](<https://devfeed.tech/tags/columnar.md>), [columnar-databases](<https://devfeed.tech/tags/columnar-databases.md>), [database-trends](<https://devfeed.tech/tags/database-trends.md>), [databases](<https://devfeed.tech/tags/databases.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [innodb](<https://devfeed.tech/tags/innodb.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [percona](<https://devfeed.tech/tags/percona.md>), [percona-server-for-mysql](<https://devfeed.tech/tags/percona-server-for-mysql.md>), [storage-engine](<https://devfeed.tech/tags/storage-engine.md>), [woocommerce](<https://devfeed.tech/tags/woocommerce.md>)

### AI overview

Columnar DuckDB replicas can speed WooCommerce analytics while the primary MySQL InnoDB server handles transactions. In a test with 400,000 orders, asynchronous replication to AliSQL's DuckDB engine reduced database size from 7.3 GiB to 601 MiB. Built-in reports ran roughly 10-63 times faster, segmented reports 14-83 times faster, and all eleven reports from an unmodified customer analytics plugin were faster; a query against live order tables took about 4.1 seconds on DuckDB versus 4 minutes 16 seconds on InnoDB. The setup used a second copy of WooCommerce pointed at the replica because WordPress uses one database connection for analytics and store traffic. Results have caveats: some MySQL-specific SQL failed on DuckDB, text grouping can differ because DuckDB ignores column collations, and floating-point decimal handling caused sub-cent differences. DuckDB also exposes less monitoring detail. The article notes that Percona does not offer or support a production DuckDB build.

### Source excerpt

MySQL and DuckDB Unlike row-oriented databases, which store data for a single record sequentially on disk, favoring single-row lookups, columnar databases store all the values from a column contiguously in storage, allowing only the selected column to be read from disk. The default MySQL engine is InnoDB, which is ACID-compliant and great for OLTP (Online ... Continued The post WooCommerce Analytics with DuckDB appeared first on Percona.

## Designing a reliable unplanned failover

DevFeed: [Designing a reliable unplanned failover](<https://devfeed.tech/articles/designing-a-reliable-unplanned-failover-121334.md>)

Original publisher: [Read original article](<https://planetscale.com/blog/designing-a-reliable-unplanned-failover>)

Author: Shlomi Noach

Published: 2026-11-05T00:00:00Z

Content type: article

Language: en

Sources: [PlanetScale](<https://devfeed.tech/sources/blog-planetscale.md>)

Topics: [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Availability](<https://devfeed.tech/topics/availability.md>), [distributed algorithm](<https://devfeed.tech/topics/distributed-algorithm.md>), [control-plane](<https://devfeed.tech/topics/control-plane.md>)

Tags: [admin](<https://devfeed.tech/tags/admin.md>), [detection](<https://devfeed.tech/tags/detection.md>), [durability](<https://devfeed.tech/tags/durability.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [failover](<https://devfeed.tech/tags/failover.md>), [freeze](<https://devfeed.tech/tags/freeze.md>), [neki](<https://devfeed.tech/tags/neki.md>), [postgres](<https://devfeed.tech/tags/postgres.md>), [primary](<https://devfeed.tech/tags/primary.md>), [wal](<https://devfeed.tech/tags/wal.md>)

### AI overview

Neki avoids promoting a replica based on Admin's failed health check alone. It compares replica reports of physical replication: a reachable replica still streaming from the primary blocks failover, while a reachable report of broken replication with no contradiction supports recovery; if Admin cannot reach any replica, it waits. Once it decides the primary is dead, Neki disconnects reachable replicas by changing `primary_conninfo`, preventing reconnection to a returning primary and freezing their WAL positions. Admin prefers a promotable replica that has replayed through the highest received WAL position, waits briefly for catch-up, then selects by `receive_lsn` and `replay_lsn`. The new primary is promoted and the others repointed; a replica ahead of it may need rewind or rebuild. With synchronous `ANY 1`, commits require an acknowledging replica, but acknowledged WAL can still be lost if the primary and the replica holding it fail together. Cross-availability-zone acknowledgements reduce that risk at the cost of commit latency. Async durability offers no guarantee that an unreachable primary has stopped accepting writes. Unrecoverable cases include having no reachable candidate.

### Source excerpt

How Neki detects a failed Postgres primary, and chooses a new one from the WAL that survived.

## Laravel October Product Updates

DevFeed: [Laravel October Product Updates](<https://devfeed.tech/articles/laravel-october-product-updates-119478.md>)

Original publisher: [Read original article](<https://laravel.com/blog/laravel-october-product-updates>)

Author: Laravel Team

Published: 2026-11-04T23:42:00Z

Content type: release

Language: en

Sources: [Laravel Blog](<https://devfeed.tech/sources/laravel-blog.md>)

Topics: [Laravel](<https://devfeed.tech/topics/laravel.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [email](<https://devfeed.tech/topics/email.md>), [observability](<https://devfeed.tech/topics/observability.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [account](<https://devfeed.tech/tags/account.md>), [added](<https://devfeed.tech/tags/added.md>), [alert](<https://devfeed.tech/tags/alert.md>), [also](<https://devfeed.tech/tags/also.md>), [api](<https://devfeed.tech/tags/api.md>), [api-key](<https://devfeed.tech/tags/api-key.md>), [apps](<https://devfeed.tech/tags/apps.md>), [array](<https://devfeed.tech/tags/array.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [email](<https://devfeed.tech/tags/email.md>), [forge](<https://devfeed.tech/tags/forge.md>), [framework](<https://devfeed.tech/tags/framework.md>), [laravel](<https://devfeed.tech/tags/laravel.md>), [october](<https://devfeed.tech/tags/october.md>), [php](<https://devfeed.tech/tags/php.md>), [php-security](<https://devfeed.tech/tags/php-security.md>), [query](<https://devfeed.tech/tags/query.md>), [security](<https://devfeed.tech/tags/security.md>), [security-advisories](<https://devfeed.tech/tags/security-advisories.md>)

### AI overview

October updates add safeguards and integrations across Laravel's framework, Cloud, and Forge. In the framework, QUERY routes can accept request bodies and remain subject to CSRF checks in the web middleware; opt-in PreventStaleDatabaseReads sends a user's reads to the write connection for a couple of seconds after a write, while usingWritePdo provides the same guarantee for individual reads. alwaysOnOneServer enables onOneServer for eligible scheduled tasks. Cloud's Resend integration imports sending domains, tracks verification, and injects mail settings and an API key into environments. Its scheduler runs supported-language commands without cron or a scheduling process, and hourly domain checks flag DNS or SSL renewal problems. Forge flags servers running PHP versions affected by the October 2026 security release, with patched-version guidance and a recheck control; failed provisioning now triggers an alert and guidance, with PHP and Nginx installation retries.

### Source excerpt

Resend email on Laravel Cloud, QUERY routes in the Laravel framework, and PHP security advisories in Forge. Here's everything that shipped in October.

## Fencing a Postgres primary that might still be alive

DevFeed: [Fencing a Postgres primary that might still be alive](<https://devfeed.tech/articles/fencing-a-postgres-primary-that-might-still-be-alive-117629.md>)

Original publisher: [Read original article](<https://clickhouse.com/blog/fencing-a-postgres-primary-that-might-still-be-alive>)

Author: Furkan Sahin

Published: 2026-11-04T00:30:00Z

Content type: article

Language: en

Sources: [ClickHouse Blog](<https://devfeed.tech/sources/clickhouse-blog.md>)

Topics: [control-plane](<https://devfeed.tech/topics/control-plane.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [hosting](<https://devfeed.tech/topics/hosting.md>), [communications](<https://devfeed.tech/topics/communications.md>), [TTL in DNS](<https://devfeed.tech/topics/ttl-in-dns.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [async](<https://devfeed.tech/tags/async.md>), [availability](<https://devfeed.tech/tags/availability.md>), [aws](<https://devfeed.tech/tags/aws.md>), [bare-metal](<https://devfeed.tech/tags/bare-metal.md>), [clickhouse](<https://devfeed.tech/tags/clickhouse.md>), [control-plane](<https://devfeed.tech/tags/control-plane.md>), [dns](<https://devfeed.tech/tags/dns.md>), [downtime](<https://devfeed.tech/tags/downtime.md>), [ec2](<https://devfeed.tech/tags/ec2.md>), [egress](<https://devfeed.tech/tags/egress.md>), [fencing](<https://devfeed.tech/tags/fencing.md>), [gcp](<https://devfeed.tech/tags/gcp.md>), [postgres](<https://devfeed.tech/tags/postgres.md>), [primary](<https://devfeed.tech/tags/primary.md>), [standby](<https://devfeed.tech/tags/standby.md>)

### AI overview

Promoting a Postgres standby and refreshing DNS do not stop writes to an old primary that remains reachable, so fencing is needed before promotion to reduce split brain risk. The control plane tries fencing methods in parallel to limit downtime. With guest SSH access, it can stop Postgres or replace pg_hba.conf, though stopping may be undone by reboot and HBA rules affect only new connections. Bare metal fencing takes down the VM's host-side veth, cutting all network traffic; it depends on the host being reachable. AWS fencing force-detaches a dedicated user NIC, since the primary NIC cannot be detached while running and security groups leave tracked connections active. GCP fencing deletes a Dynamic NIC; measured last-packet delays were 0.7-2.5 seconds, with the API completing in 5-7 seconds. If every method fails, the standby still promotes: the design deliberately favors availability over waiting for a confirmed fence.

### Source excerpt

Promoting a standby doesn't stop the old Postgres primary from taking writes. Here's how the control plane behind ClickHouse Managed Postgres fences it on bare metal, AWS and GCP, and why the design is fail-open.

## Slow Job Alerts and Replica Commit IDs in Laravel 13.40

DevFeed: [Slow Job Alerts and Replica Commit IDs in Laravel 13.40](<https://devfeed.tech/articles/slow-job-alerts-and-replica-commit-ids-in-laravel-13-40-119076.md>)

Original publisher: [Read original article](<https://laravel-news.com/laravel-13-40-0>)

Author: Paul Redmond

Published: 2026-11-03T23:53:01Z

Content type: release

Language: en

Sources: [Laravel](<https://devfeed.tech/sources/laravel.md>)

Topics: [Laravel](<https://devfeed.tech/topics/laravel.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>)

Tags: [alerts](<https://devfeed.tech/tags/alerts.md>), [commit](<https://devfeed.tech/tags/commit.md>), [ids](<https://devfeed.tech/tags/ids.md>), [job](<https://devfeed.tech/tags/job.md>), [jobs](<https://devfeed.tech/tags/jobs.md>), [lag](<https://devfeed.tech/tags/lag.md>), [laravel](<https://devfeed.tech/tags/laravel.md>), [laravel-releases](<https://devfeed.tech/tags/laravel-releases.md>), [middleware](<https://devfeed.tech/tags/middleware.md>), [news](<https://devfeed.tech/tags/news.md>), [queue](<https://devfeed.tech/tags/queue.md>), [read-replica](<https://devfeed.tech/tags/read-replica.md>), [replica](<https://devfeed.tech/tags/replica.md>), [replica-lag](<https://devfeed.tech/tags/replica-lag.md>), [scheduled](<https://devfeed.tech/tags/scheduled.md>), [scheduled-tasks](<https://devfeed.tech/tags/scheduled-tasks.md>), [tasks](<https://devfeed.tech/tags/tasks.md>)

### AI overview

Laravel 13.40.0 adds runtime alerts for queue jobs and scheduled tasks: plain numeric thresholds are milliseconds for jobs and seconds for scheduled tasks, and callbacks run after job attempts or task completion/failure. Database commit IDs let applications check whether replicas have received writes, with optional waiting; support is specified for MySQL with GTIDs, MariaDB with binary logging, and PostgreSQL, though PostgreSQL waiting requires version 19 and other drivers throw. Middleware can now be inserted around built-in middleware, and strict request input casting returns the default for invalid values; strict booleans accept only listed common true/false forms. The release also secures process-driver concurrency results with signed frames, reconnects after transaction connection loss, and rolls back nested transactions to savepoints on lock timeouts or deadlocks.

### Source excerpt

Laravel 13.40 adds callbacks for slow queue jobs and scheduled tasks, commit IDs to check read replica lag, and insertBefore() and insertAfter() for middleware. The post Slow Job Alerts and Replica Commit IDs in Laravel 13.40 appeared first on Laravel News. Join the Laravel Newsletter to get Laravel articles like this directly in your inbox.

## Is Azure SQL Database Hyperscale "distributed SQL"?

DevFeed: [Is Azure SQL Database Hyperscale "distributed SQL"?](<https://devfeed.tech/articles/is-azure-sql-database-hyperscale-distributed-sql-109694.md>)

Original publisher: [Read original article](<https://devblogs.microsoft.com/azure-sql/is-azure-sql-database-hyperscale-distributed-sql/>)

Author: Anna Hoffman

Published: 2026-10-30T16:52:56Z

Content type: article

Language: en

Sources: [Azure SQL Dev Corner](<https://devfeed.tech/sources/azure-sql-dev-corner.md>)

Topics: [Azure SQL](<https://devfeed.tech/topics/azure-sql.md>), [sql-database](<https://devfeed.tech/topics/sql-database.md>), [sql-server](<https://devfeed.tech/topics/sql-server.md>), [Amazon Aurora](<https://devfeed.tech/topics/amazon-aurora.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [SQL](<https://devfeed.tech/topics/sql.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [amazon-aurora](<https://devfeed.tech/tags/amazon-aurora.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [aws](<https://devfeed.tech/tags/aws.md>), [azure](<https://devfeed.tech/tags/azure.md>), [azure-sql](<https://devfeed.tech/tags/azure-sql.md>), [azure-sql-database](<https://devfeed.tech/tags/azure-sql-database.md>), [backup](<https://devfeed.tech/tags/backup.md>), [backups](<https://devfeed.tech/tags/backups.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [commit](<https://devfeed.tech/tags/commit.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [hyperscale](<https://devfeed.tech/tags/hyperscale.md>), [sql](<https://devfeed.tech/tags/sql.md>), [t-sql](<https://devfeed.tech/tags/t-sql.md>)

### AI overview

Azure SQL Database Hyperscale is a distributed architecture with a single-writer SQL engine. Compute replicas process queries, a log service handles durability and distributes log records, page servers hold data slices, and Azure Storage keeps the long-term copy and snapshots. This design supports automatic storage growth up to 128 TB, near-instant snapshot backups, restores largely independent of database size, and replicas that can be added without copying data. It retains SQL Server transaction semantics and T-SQL compatibility, but only one primary accepts writes; workloads requiring multiple writers face a different trade-off. In a Microsoft-commissioned comparison using HammerDB TPROC-C on a 2.3 TB database, Hyperscale Premium-series at 16 vCores delivered up to 52% higher performance and up to 51% better performance per dollar than Aurora PostgreSQL I/O-Optimized, leading at each tested concurrency level. The author cautions that benchmark results may not predict an individual workload and recommends testing it.

### Source excerpt

In my last post, I broke down the Socrates paper, the research that became Azure SQL Database Hyperscale. Writing it made me want to tackle a question I think trips a lot of people up when they're choosing a database: is Azure SQL Database Hyperscale "distributed SQL"? It's a fair question. "Distributed SQL" has become [...] The post Is Azure SQL Database Hyperscale "distributed SQL"? appeared first on Azure SQL Dev Corner.

## Database Replication: Types, Benefits & How It Works

DevFeed: [Database Replication: Types, Benefits & How It Works](<https://devfeed.tech/articles/database-replication-types-benefits-how-it-works-105579.md>)

Original publisher: [Read original article](<https://talent500.com/blog/database-replication/>)

Author: Ekakshar

Published: 2026-10-28T10:51:10Z

Content type: article

Language: en

Sources: [Talent500](<https://devfeed.tech/sources/backend-archives-talent500-blog.md>)

Topics: [copy](<https://devfeed.tech/topics/copy.md>), [Database](<https://devfeed.tech/topics/database.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Disaster Recovery](<https://devfeed.tech/topics/disaster-recovery.md>), [database-consistency](<https://devfeed.tech/topics/database-consistency.md>), [database-scalability](<https://devfeed.tech/topics/database-scalability.md>), [transaction-log](<https://devfeed.tech/topics/transaction-log.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>), [Debezium Connector](<https://devfeed.tech/topics/debezium-connector.md>), [Apache Cassandra](<https://devfeed.tech/topics/cassandra.md>)

Tags: [availability](<https://devfeed.tech/tags/availability.md>), [backend](<https://devfeed.tech/tags/backend.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [data](<https://devfeed.tech/tags/data.md>), [database](<https://devfeed.tech/tags/database.md>), [database-replication](<https://devfeed.tech/tags/database-replication.md>), [disaster](<https://devfeed.tech/tags/disaster.md>), [disaster-recovery](<https://devfeed.tech/tags/disaster-recovery.md>), [failover](<https://devfeed.tech/tags/failover.md>), [first](<https://devfeed.tech/tags/first.md>), [helps](<https://devfeed.tech/tags/helps.md>), [multiple](<https://devfeed.tech/tags/multiple.md>), [practices](<https://devfeed.tech/tags/practices.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [replication](<https://devfeed.tech/tags/replication.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [servers](<https://devfeed.tech/tags/servers.md>), [synchronization](<https://devfeed.tech/tags/synchronization.md>), [types](<https://devfeed.tech/tags/types.md>), [works](<https://devfeed.tech/tags/works.md>)

### AI overview

Database replication copies and propagates data across database instances to support availability, disaster recovery, read scaling and regional access. The article distinguishes primary-replica setups, which concentrate writes on one primary, from multi-primary and peer-to-peer designs that distribute writes but require conflict handling; cascading replication can reduce direct primary connections while adding downstream lag. Synchronous replication waits for configured replica acknowledgements and can reduce loss of acknowledged transactions at the cost of latency or blocked writes. Asynchronous replication lowers write delays but permits stale reads and possible loss of recent changes during failover; semi-synchronous guarantees depend on what the database acknowledges. Implementation typically initializes replicas from a snapshot or backup, captures changes through logs or CDC, transfers and applies them, then monitors lag, errors and log retention. Replicas do not replace backups, since deletions or corruption may propagate. Reliable deployments also need routing and promotion procedures, tested recovery, consistency choices, access controls and cost planning. PostgreSQL, MySQL

### Source excerpt

Database Replication helps maintain data availability and reliability by copying data across multiple database servers. Explore replication types, synchronization methods, failover strategies, benefits, and challenges, along with best practices for improving scalability, fault tolerance, and disaster recovery. The post Database Replication: Types, Benefits & How It Works appeared first on Talent500 blog.

## How to Scale Your Database

DevFeed: [How to Scale Your Database](<https://devfeed.tech/articles/how-to-scale-your-database-99201.md>)

Original publisher: [Read original article](<https://blog.levelupcoding.com/p/how-to-scale-your-database>)

Author: Nikki Siapno

Published: 2026-10-25T13:22:59Z

Content type: article

Language: en

Sources: [Level Up Coding System Design Newsletter](<https://devfeed.tech/sources/level-up-coding-system-design-newsletter.md>)

Topics: [database-scalability](<https://devfeed.tech/topics/database-scalability.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [sharding](<https://devfeed.tech/topics/sharding.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Database](<https://devfeed.tech/topics/database.md>), [Amazon RDS](<https://devfeed.tech/topics/amazon-rds.md>), [Azure SQL](<https://devfeed.tech/topics/azure-sql.md>), [cloud-sql](<https://devfeed.tech/topics/cloud-sql.md>)

Tags: [database](<https://devfeed.tech/tags/database.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [scale](<https://devfeed.tech/tags/scale.md>), [sharding](<https://devfeed.tech/tags/sharding.md>)

### AI overview

Scale a database by identifying its actual bottleneck, then adding complexity only as needed: monitor resources and query latency; reduce unnecessary work with caching or asynchronous processing; optimize frequently used queries with indexes; distribute read traffic across replicas when reads overwhelm the primary; and consider sharding when one primary can no longer handle the workload. Caches trade load reduction for data freshness and require expiration or invalidation decisions. Indexes speed lookups but add storage and write overhead. Replicas can serve reads across machines or regions, but propagation delays may leave stale data, so immediate read-after-write operations may need the primary. Sharding distributes data and workload, but makes data placement, cross-shard queries, and data movement harder. Managed database services can automate some operations without removing the need to diagnose the bottleneck.

### Source excerpt

More replicas? Bigger machines? Sharding? Here's what to consider first.

## The paper that split the database in four: inside Azure SQL Database Hyperscale, aka Socrates

DevFeed: [The paper that split the database in four: inside Azure SQL Database Hyperscale, aka Socrates](<https://devfeed.tech/articles/the-paper-that-split-the-database-in-four-inside-azure-sql-database-hyperscale-aka-socrates-101493.md>)

Original publisher: [Read original article](<https://devblogs.microsoft.com/azure-sql/the-paper-that-split-the-database-in-four-inside-azure-sql-database-hyperscale-aka-socrates/>)

Author: Anna Hoffman

Published: 2026-10-25T11:23:29Z

Content type: article

Language: en

Sources: [Azure SQL Dev Corner](<https://devfeed.tech/sources/azure-sql-dev-corner.md>)

Topics: [Azure SQL](<https://devfeed.tech/topics/azure-sql.md>), [transaction-log](<https://devfeed.tech/topics/transaction-log.md>), [sql-server](<https://devfeed.tech/topics/sql-server.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [sql-database](<https://devfeed.tech/topics/sql-database.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [Azure](<https://devfeed.tech/topics/azure.md>)

Tags: [also](<https://devfeed.tech/tags/also.md>), [azure](<https://devfeed.tech/tags/azure.md>), [azure-sql](<https://devfeed.tech/tags/azure-sql.md>), [azure-sql-database](<https://devfeed.tech/tags/azure-sql-database.md>), [brand](<https://devfeed.tech/tags/brand.md>), [but](<https://devfeed.tech/tags/but.md>), [database](<https://devfeed.tech/tags/database.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [first](<https://devfeed.tech/tags/first.md>), [hood](<https://devfeed.tech/tags/hood.md>), [hyperscale](<https://devfeed.tech/tags/hyperscale.md>), [inside](<https://devfeed.tech/tags/inside.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-research](<https://devfeed.tech/tags/microsoft-research.md>), [microsoft-sql](<https://devfeed.tech/tags/microsoft-sql.md>), [paper](<https://devfeed.tech/tags/paper.md>), [partners](<https://devfeed.tech/tags/partners.md>), [split](<https://devfeed.tech/tags/split.md>), [sql](<https://devfeed.tech/tags/sql.md>), [sql-database](<https://devfeed.tech/tags/sql-database.md>)

### AI overview

Azure SQL Database Hyperscale, codenamed Socrates, avoids the full-replica architecture's data-sized limits by splitting query processing, transaction logging, data pages, and durable storage across compute nodes, XLOG, page servers, and XStore. Compute nodes cache pages in memory and RBPEX; on a cache miss, GetPage@LSN asks a page server for a page current to a specified log position. XLOG distributes committed log records, while XStore snapshots support backups. This design removes full data copying when adding compute or replicas and makes backup and restore less dependent on database size. The text reports scale operations changing from O(data) to O(1), about half the storage overhead, and higher update-heavy log throughput than the full-replica design. It also cites skewed workloads where small caches served a larger share of reads. Hyperscale retains T-SQL compatibility, with tradeoffs involving remote page fetches and reliance on caching; the excerpt gives no detailed operational comparison of those effects.

### Source excerpt

Under the Hood: The Research Behind Microsoft SQL, part 1 The Microsoft SQL team is always innovating. They do this in day-to-day efforts, but there are also brand new efforts that are constantly being explored, investigated, tested, and researched. Our engineering team partners closely with MSR (Microsoft Research) on some of these efforts. The team [...] The post The paper that split the database in four: inside Azure SQL Database Hyperscale, aka Socrates appeared first on Azure SQL Dev Corner.

## Laravel Read Replicas: Prevent Stale Reads After a Write

DevFeed: [Laravel Read Replicas: Prevent Stale Reads After a Write](<https://devfeed.tech/articles/laravel-read-replicas-prevent-stale-reads-after-a-write-97626.md>)

Original publisher: [Read original article](<https://laravel-news.com/laravel-prevent-stale-database-reads>)

Author: Paul Redmond

Published: 2026-10-22T23:20:21Z

Content type: tutorial

Language: en

Sources: [Laravel](<https://devfeed.tech/sources/laravel.md>)

Topics: [read-replica](<https://devfeed.tech/topics/read-replica.md>), [Laravel](<https://devfeed.tech/topics/laravel.md>), [middleware](<https://devfeed.tech/topics/middleware.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [Eloquent ORM](<https://devfeed.tech/topics/eloquent.md>)

Tags: [database](<https://devfeed.tech/tags/database.md>), [lag](<https://devfeed.tech/tags/lag.md>), [laravel](<https://devfeed.tech/tags/laravel.md>), [middleware](<https://devfeed.tech/tags/middleware.md>), [news](<https://devfeed.tech/tags/news.md>), [prevent](<https://devfeed.tech/tags/prevent.md>), [read-replicas](<https://devfeed.tech/tags/read-replicas.md>), [replica](<https://devfeed.tech/tags/replica.md>), [replica-lag](<https://devfeed.tech/tags/replica-lag.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

Laravel 13.38 adds `PreventStaleDatabaseReads` to address replica lag across separate requests: after a request modifies records, the middleware stores a session timestamp, then routes reads to the write server on subsequent requests for a configurable window (two seconds by default). Set the window to cover the longest lag seen in replica metrics; a shorter window can still expose stale data. This differs from the `sticky` connection option, which only keeps reads on the writer for the remainder of the same request. For code that needs fresh reads regardless of the user's recent writes, `DB::usingWritePdo()` routes all reads inside a callback to the writer, including relationship queries; the connection returns to its prior setting when the callback finishes or throws. Queries in a transaction already use the writer. The article also notes that middleware tests should avoid `RefreshDatabase`, whose transaction would make queries use the writer even without the middleware.

### Source excerpt

Laravel's PreventStaleDatabaseReads middleware keeps a user's reads on the write database after they save, so read replica lag doesn't show them old data. The post Laravel Read Replicas: Prevent Stale Reads After a Write appeared first on Laravel News. Join the Laravel Newsletter to get Laravel articles like this directly in your inbox.

## Troubleshooting Data Mismatches with tablediff on Amazon RDS for SQL Server

DevFeed: [Troubleshooting Data Mismatches with tablediff on Amazon RDS for SQL Server](<https://devfeed.tech/articles/troubleshooting-data-mismatches-with-tablediff-on-amazon-rds-for-sql-server-95993.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/database/troubleshooting-data-mismatches-with-tablediff-on-amazon-rds-for-sql-server/>)

Author: Kanchan Bhattacharyya

Published: 2026-10-22T16:33:04Z

Content type: article

Language: en

Sources: [AWS Database Blog](<https://devfeed.tech/sources/aws-database-blog.md>)

Topics: [RDS for SQL Server](<https://devfeed.tech/topics/rds-for-sql-server.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [sql-server](<https://devfeed.tech/topics/sql-server.md>), [Amazon RDS](<https://devfeed.tech/topics/amazon-rds.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [observability](<https://devfeed.tech/topics/observability.md>), [Disaster Recovery](<https://devfeed.tech/topics/disaster-recovery.md>), [audit](<https://devfeed.tech/topics/audit.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>)

Tags: [2022](<https://devfeed.tech/tags/2022.md>), [2025](<https://devfeed.tech/tags/2025.md>), [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [amazon-cloudwatch](<https://devfeed.tech/tags/amazon-cloudwatch.md>), [amazon-ec2](<https://devfeed.tech/tags/amazon-ec2.md>), [amazon-elastic-compute-cloud-amazon-ec2](<https://devfeed.tech/tags/amazon-elastic-compute-cloud-amazon-ec2.md>), [amazon-rds](<https://devfeed.tech/tags/amazon-rds.md>), [amazon-simple-storage-service-amazon-s3](<https://devfeed.tech/tags/amazon-simple-storage-service-amazon-s3.md>), [asynchronous](<https://devfeed.tech/tags/asynchronous.md>), [audit](<https://devfeed.tech/tags/audit.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [lag](<https://devfeed.tech/tags/lag.md>), [rds-for-sql-server](<https://devfeed.tech/tags/rds-for-sql-server.md>), [read-replica](<https://devfeed.tech/tags/read-replica.md>), [replication](<https://devfeed.tech/tags/replication.md>), [scripts](<https://devfeed.tech/tags/scripts.md>), [sql-server](<https://devfeed.tech/tags/sql-server.md>), [t-sql](<https://devfeed.tech/tags/t-sql.md>), [technical-how-to](<https://devfeed.tech/tags/technical-how-to.md>), [troubleshooting](<https://devfeed.tech/tags/troubleshooting.md>), [utility](<https://devfeed.tech/tags/utility.md>)

### AI overview

After an Amazon RDS for SQL Server read replica catches up, tablediff can provide independent row-level parity evidence that replication metrics alone do not: compare the primary and replica only after `ReplicaLag`, `redo_queue_size`, and `log_send_queue_size` are all zero. Run it from a Windows EC2 client with access to both endpoints; target tables need a clustered primary key or unique index. It reports source-only, destination-only, and column-mismatched rows, and can generate T-SQL fixes. The replica is read-only, so remediation requires promotion, which is irreversible; regenerate the fix script after promotion and wrap its bare DML statements in an explicit transaction with a validation gate. Re-run tablediff to confirm zero differences. A fresh rebuild may be preferable for extensive or structural divergence. Comparisons scan whole tables and can give misleading results during active writes or redo, so run during a quiet period after queues drain.

### Source excerpt

Learn how to use the SQL Server tablediff utility to validate an Amazon RDS for SQL Server read replica against its primary after a replication lag event, detect and classify row-level data mismatches, and apply auditable T-SQL fix scripts on a promoted replica instance.

## Before You Buy Another Server: Squeezing a Read-Heavy API for Max RPS

DevFeed: [Before You Buy Another Server: Squeezing a Read-Heavy API for Max RPS](<https://devfeed.tech/articles/before-you-buy-another-server-squeezing-a-read-heavy-api-for-max-rps-95276.md>)

Original publisher: [Read original article](<https://hackernoon.com/before-you-buy-another-server-squeezing-a-read-heavy-api-for-max-rps?source=rss>)

Author: navaprasad maram

Published: 2026-10-22T08:18:30Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [API](<https://devfeed.tech/topics/api.md>), [Streaming](<https://devfeed.tech/topics/streaming.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [api-latency-reduction](<https://devfeed.tech/tags/api-latency-reduction.md>), [api-load-testing](<https://devfeed.tech/tags/api-load-testing.md>), [api-performance-optimization](<https://devfeed.tech/tags/api-performance-optimization.md>), [api-scalability](<https://devfeed.tech/tags/api-scalability.md>), [async](<https://devfeed.tech/tags/async.md>), [backend-optimization](<https://devfeed.tech/tags/backend-optimization.md>), [pipelining](<https://devfeed.tech/tags/pipelining.md>), [redis](<https://devfeed.tech/tags/redis.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [scaling-a-news-api](<https://devfeed.tech/tags/scaling-a-news-api.md>), [serialization](<https://devfeed.tech/tags/serialization.md>), [server](<https://devfeed.tech/tags/server.md>), [streaming](<https://devfeed.tech/tags/streaming.md>), [system-architecture](<https://devfeed.tech/tags/system-architecture.md>)

### AI overview

For a read-heavy inventory API, the author's sequence for improving per-instance throughput was to eliminate request-time work, reduce Redis round trips, stream results, then tune the server and network stack. Redis stored pre-rendered JSON bytes, which the service concatenated without parsing or serialization; ingestion paid the rendering cost, and two response shapes required denormalized copies. Grouping SKU-location data in Redis hashes reduced lookups, while a thread-safe async Lettuce client used one multiplexed connection per node; shorter timeouts, replica reads and topology refresh addressed failure and cluster load. Async callbacks fed a BlockingQueue consumed by a request thread writing chunked responses as results arrived. Netty benchmarked fastest, but the team chose Tomcat with more threads because its near-benchmark capacity was easier for the team to operate. HTTP/2 remained optional for clients that negotiate it, and the team retained TLS on both network hops despite the measured performance cost. Results depend on workload and infrastructure; the article recommends measuring compute time separately from delivery time.

### Source excerpt

Learn how to squeeze maximum RPS from a read-heavy API before buying more servers--using zero-serialization, Redis pipelining, and response streaming.

[Next page](<https://devfeed.tech/topics/read-replica.md?cursor=WyIyMDI2LTEwLTIyVDA4OjE4OjMwKzAwOjAwIiwgImI2MjZjMmVmLTJmYzEtNDdhNy04ZWYzLTA2YmYyNWEyMDExYiJd>)