The most common is the implicit flow.

The most secure is the code flow, but it's more complex to implement as it requires more steps. As it is more complex, many providers end up suggesting the implicit flow.

/// note

It's common that each authentication provider names their flows in a different way, to make it part of their brand.

* 📝 Update `security/first-steps.md`. PR [#11674](https://github.com/tiangolo/fastapi/pull/11674) by [@alejsdev](https://github.com/alejsdev).
* 📝 Update `security/first-steps.md`. PR [#11673](https://github.com/tiangolo/fastapi/pull/11673) by [@alejsdev](https://github.com/alejsdev).

3. **Reliability**: 
   - Persistent handles survive 99% of network interruptions
   - Witness protocol achieves <5 second failover time
4. **Compatibility**: Works with Windows Server 2016+ and Azure Files

## Next Steps

1. **Immediate Actions**:
   - Set up Windows Server test environment
   - Create detailed technical specifications for Phase 1
   - Identify and evaluate Java RDMA libraries

2. **Phase 1 Kickoff**:

   *
   * TODO(b/218700094): Does this matter to prod users, or is it just a feature of our testing
   * environment? If the latter, maybe write a custom Executor that avoids failing the test when it
   * sees an Error?
   */
  @AndroidIncompatible
  public void testTaskThrowsError() throws Exception {
    class MyError extends Error {}
    CyclicBarrier barrier = new CyclicBarrier(2);

     */
    if (throwable instanceof Error) {
      /*
       * TODO(cpovirk): Do we really want to log this if we called setException(throwable) and it
       * returned true? This was intentional (CL 46470009), but it seems odd compared to how we
       * normally handle Error.
       *
       * Similarly, do we really want to log the same Error more than once?
       */
      log(throwable);
    }
  }

                    } else {
                        final long interval = fessConfig.getThumbnailGeneratorIntervalAsInteger().longValue();
                        if (interval > 0) {
                            ThreadUtil.sleep(interval);
                        }
                    }
                } else {
                    logger.warn("{} is not available.", generatorName);
                }

        log.info("Connection lost, waiting for witness notification before retry");
        
        // Wait briefly for witness notification before retrying
        try {
            Thread.sleep(2000);  // 2 second grace period
        } catch (InterruptedException e) {
            Thread.currentThread().interrupt();
        }
    }
    
    // Proceed with normal error handling/retry

   * elements.length) are all {@code null}.
   */
  @VisibleForTesting transient @Nullable Object @Nullable [] elements;

  /**
   * Keeps track of metadata like the number of hash table bits and modifications of this data
   * structure (to make it possible to throw ConcurrentModificationException in the iterator). Note

    failover.handleFailure(channel, error);
    
    // Verify channel marked as failed
    assertEquals(ChannelState.FAILED, channel.getState());
    
    // Verify recovery attempted
    Thread.sleep(2000);
    assertTrue(channel.getState() == ChannelState.ESTABLISHED 
        || manager.getChannels().isEmpty());
}
```

### 7.2 Integration Tests
```java
@Test

                this.readSize =
                        Math.min(th.getConfig().getReceiveBufferSize() - 70, th.areSignaturesActive() ? 0xFFFF - 70 : 0xFFFFFF - 70);
            }

            // there seems to be a bug with some servers that causes corruption if using signatures + CAP_LARGE_WRITE
            if (th.hasCapability(SmbConstants.CAP_LARGE_WRITEX) && !th.areSignaturesActive()) {