Senior Java Engineer; all my links: https://code-with-kyryl.carrd.co/https://bsky.app/profile/code-with-kyryl.bsky.socialhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mrscjitj3w2q29 Jul 2026 16:05 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mrscjitj3w2qhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mrnlzcavdc2u27 Jul 2026 19:11 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mrnlzcavdc2uhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mrdk6gfwdc2a1/ You set Kafka to exactly-once, and a payment still got processed twice.23 Jul 2026 19:12 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mrdk6gfwdc2ahttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mr3wkrrotc2iYour Kafka retention.ms says 1 hour. The data is still on disk 6 hours later. Not a bug. Kafka deletes whole segments, never individual records.20 Jul 2026 18:32 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mr3wkrrotc2ihttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mqrwnhhdw22hAdding one consumer to a Kafka group can freeze every consumer for seconds. That is eager rebalancing, and it is still the default. Here is what actually happens, and the one config that fixes it.16 Jul 2026 19:07 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mqrwnhhdw22hhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mqee5xuadc2sCDC turns your database into an event source. And it is the cleanest way to share data without sharing a schema. No dual-write, no "remember to also emit an event" discipline spread across every service.11 Jul 2026 09:32 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mqee5xuadc2shttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mq3acdqz2o2nThe database went down for ten minutes. The restart storm it triggered lasted forty. One `/actuator/health` endpoint wired to both the liveness and readiness probe. That is the whole bug.07 Jul 2026 18:29 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mq3acdqz2o2nhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mq342gnbz42xCREATE INDEX without CONCURRENTLY locks out writes for the entire build. On a large live table that is not a migration. It is a multi-minute write outage for everything touching it.07 Jul 2026 17:13 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mq342gnbz42xhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpyn7ffldk2tA path variable in your metrics tag will quietly bankrupt your Prometheus backend. It passes code review. It compiles. It works, for a while.06 Jul 2026 17:42 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpyn7ffldk2thttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpuvok2k762h`@Transactional(readOnly = true)` gets slapped on every query method as a habit, treated as a marker for the next reader. It is not decorative. It changes Hibernate's flush mode.05 Jul 2026 06:03 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpuvok2k762hhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mptdmfaw6o2fCalled my own `@Async` method from inside the same class. No error. No async. Ran inline like a plain method call.04 Jul 2026 15:07 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mptdmfaw6o2fhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mptcpreqt52jRetry aspect and `@Transactional` aspect on the same method. No `@Order` on either. Spring is free to nest them any way it likes, and it does not ask you first.04 Jul 2026 14:51 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mptcpreqt52jhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpravdittg2ePrototype bean field-injected into a singleton. Created once. Spring resolves the constructor arg one time, the singleton holds that instance forever. State leaks between requests until someone finally traces it back.03 Jul 2026 19:13 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpravdittg2ehttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpr3w6ssoe2fA test with a 30-argument constructor call. Somewhere in there, the one field under test. Good luck spotting it without counting positions.03 Jul 2026 17:44 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpr3w6ssoe2fhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mplvsycisl2iGood enough integration tests were fine until AI started writing them. The authoring cost is gone, so the bar moves from "good enough" to exact.01 Jul 2026 16:11 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mplvsycisl2ihttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpjiryrqo52x`@Enumerated(ORDINAL)` is the JPA default, and the default silently corrupts data. It stores the enum's position: 0, 1, 2. Add a constant mid-list later and every old row decodes wrong. No error. No failed query.30 Jun 2026 17:13 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpjiryrqo52xhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpgzww6yk526Every team builds the same cost bot. A Lambda on a Cost Explorer cron, a CloudTrail lookup, a Slack ping for whoever tagged the runaway resource. It works. You just maintain every line of it yourself. AWS just shipped the managed version.29 Jun 2026 17:42 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpgzww6yk526https://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpgrxdjeyf2kEvery `POST /search` you have shipped is a safe, idempotent read wearing a write's costume. You used POST because GET will not carry a body and your filters were too big for the URL. HTTP finally has the right verb. It is QUERY.29 Jun 2026 15:19 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpgrxdjeyf2khttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mpbphcoase2xI created three Kafka quizzes and added them to my tech blog, one per level. A straightforward way to test your knowledge and see where the gaps actually are. 👇27 Jun 2026 14:51 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mpbphcoase2xhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mp7inq7kdm2l`UserService` + `UserServiceImpl`. One interface, one impl, forever. The reason this pattern exists died around Spring Boot 2.0, and most of us never updated the habit.26 Jun 2026 17:44 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mp7inq7kdm2lhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3mp4zdwqioh2fYour event fires. Your listener queries the DB for the entity just saved. It finds nothing. Add a log line — it starts working. Remove it — breaks again. That's not a race condition. That's `@EventListener`.25 Jun 2026 18:04 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3mp4zdwqioh2fhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3molg2zoib42qMapStruct's default config quietly ships bugs. Add a field to a DTO, forget to map it, it's null in prod. The compiler knew and said nothing, because the default reporting policy is WARN. Three things I put on every mapper.18 Jun 2026 18:04 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3molg2zoib42qhttps://bsky.app/profile/code-with-kyryl.bsky.social/post/3moa3v7e2gf26Confluence docs for your service are outdated the moment you push the code. Not a discipline problem. Structural. There's no mechanism keeping external docs in sync with code.14 Jun 2026 06:03 +0000at://did:plc:rxtinms6aewicsigmv6nr464/app.bsky.feed.post/3moa3v7e2gf26