[TH2-1988] separate connections for publishing and subscribing #318
Conversation
lumber1000
requested review from
OptimumCode and
Nikita-Smirnov-Exactpro
and removed request for
OptimumCode
Test Results 31 files 31 suites 5s ⏱️ Results for commit 6b5c84a. ♻️ This comment has been updated with latest results. |
.../java/com/exactpro/th2/common/schema/message/impl/rabbitmq/connection/ConnectionManager.java
Outdated
Show resolved
Hide resolved
.../java/com/exactpro/th2/common/schema/message/impl/rabbitmq/connection/ConnectionManager.java
Outdated
Show resolved
Hide resolved
Nikita-Smirnov-Exactpro
changed the title
| private final Map<PinId, ChannelHolder> channelsByPin = new ConcurrentHashMap<>(); | ||
| private final AtomicReference<State> connectionIsClosed = new AtomicReference<>(State.OPEN); | ||
| private final AtomicReference<State> connectionState = new AtomicReference<>(State.OPEN); |
There was a problem hiding this comment.
In this implementation state is shared between to real connection. Maybe instead of correct current Connection Manager, create ConnectionRouter witch delegates publish / subscribe requests to two instance of origin Connection manager.
src/main/kotlin/com/exactpro/th2/common/metrics/CommonMetrics.kt
Outdated
Show resolved
Hide resolved
...lin/com/exactpro/th2/common/schema/message/impl/rabbitmq/connection/TestConnectionManager.kt
Outdated
Show resolved
Hide resolved
| assertEquals(blockAfter.toLong(), messagesCount - countDown.count) | ||
|
|
||
| // unblocks publishers | ||
| rabbit.executeInContainerWithLogging("rabbitmqctl", "set_vm_memory_high_watermark", "0.4") |
There was a problem hiding this comment.
Should we unlock the publisher in this test?
| retryTimeDeviationPercent = 0 | ||
| ) | ||
| ).use { manager -> | ||
| repeat(messagesCount) { index -> |
There was a problem hiding this comment.
messagesCount is greater than blockAfter, it means that publisher not blocking in this test because check logic is after this circle
There was a problem hiding this comment.
We send 3 messages, then block publishers, then send 7 more. Subscribe, receive 3, unblock, receive 7. This allows to ensure that publishers were actually blocked. We can simplify the scenario: just send a few messages, block publishers, try to read messages.
There was a problem hiding this comment.
then send 7 more -> If publisher blocked we could able to publish additional messages
…t simplified getMonitorName fixed
…ic rollback isPublishingBlocked property Awaitility used
|
@Nikita-Smirnov-Exactpro @lumber1000 Should we close this PR after #321 is merged? |
No description provided.