Github user mxm commented on a diff in the pull request: https://github.com/apache/flink/pull/2618#discussion_r83022754 --- Diff: flink-tests/src/test/java/org/apache/flink/test/checkpointing/ContinuousFileProcessingCheckpointITCase.java --- @@ -257,74 +190,158 @@ public void open(Configuration parameters) throws Exception { long failurePosMax = (long) (0.7 * LINES_PER_FILE); elementsToFailure = (new Random().nextLong() % (failurePosMax - failurePosMin)) + failurePosMin; - - if (elementCounter >= NO_OF_FILES * LINES_PER_FILE) { - finalCollectedContent = new HashMap<>(); - for (Map.Entry<Integer, Set<String>> result: collectedContent.entrySet()) { - finalCollectedContent.put(result.getKey(), new ArrayList<>(result.getValue())); - } - throw new SuccessException(); - } - } - - @Override - public void close() { - try { - super.close(); - } catch (Exception e) { - e.printStackTrace(); - } } @Override public void invoke(String value) throws Exception { - int fileIdx = Character.getNumericValue(value.charAt(0)); + int fileIdx = getFileIdx(value); - Set<String> content = collectedContent.get(fileIdx); + Set<String> content = actualContent.get(fileIdx); if (content == null) { content = new HashSet<>(); - collectedContent.put(fileIdx, content); + actualContent.put(fileIdx, content); } + // detect duplicate lines. if (!content.add(value + "\n")) { fail("Duplicate line: " + value); System.exit(0); } - elementCounter++; + + // this is termination if (elementCounter >= NO_OF_FILES * LINES_PER_FILE) { - finalCollectedContent = new HashMap<>(); - for (Map.Entry<Integer, Set<String>> result: collectedContent.entrySet()) { - finalCollectedContent.put(result.getKey(), new ArrayList<>(result.getValue())); - } + actualCollectedContent = actualContent; throw new SuccessException(); } - count++; - if (!hasFailed) { + // add some latency so that we have at least one checkpoint in + if (!hasFailed && successfulCheckpoints == 0) { Thread.sleep(2); --- End diff -- The waiting time seems arbitrary. Is it required for the next step (see below)?
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---