Merge branch 'failure-mode' of https://github.com/smarunich/llm-d-inference-sim into failure-mode

Author: smarunich

README.md

@@ -118,15 +118,16 @@ For more details see the <a href="https://docs.vllm.ai/en/stable/getting_started
 - `min-tool-call-array-param-length`: the minimum possible length of array parameters in a tool call, optional, defaults to 1
 - `tool-call-not-required-param-probability`: the probability to add a parameter, that is not required, in a tool call, optional, defaults to 50
 - `object-tool-call-not-required-field-probability`: the probability to add a field, that is not required, in an object in a tool call, optional, defaults to 50
+<!-- 
 - `enable-kvcache`: if true, the KV cache support will be enabled in the simulator. In this case, the KV cache will be simulated, and ZQM events will be published when a KV cache block is added or evicted. 
 - `kv-cache-size`: the maximum number of token blocks in kv cache
 - `block-size`: token block size for contiguous chunks of tokens, possible values: 8,16,32,64,128
 - `tokenizers-cache-dir`: the directory for caching tokenizers
 - `hash-seed`: seed for hash generation (if not set, is read from PYTHONHASHSEED environment variable)
 - `zmq-endpoint`: ZMQ address to publish events
+- `event-batch-size`: the maximum number of kv-cache events to be sent together, defaults to 16
 - `failure-injection-rate`: probability (0-100) of injecting failures, optional, default is 0
 - `failure-types`: list of specific failure types to inject (rate_limit, invalid_api_key, context_length, server_error, invalid_request, model_not_found), optional, if empty all types are used
-- `event-batch-size`: the maximum number of kv-cache events to be sent together, defaults to 16
 -->
 In addition, as we are using klog, the following parameters are available:
 - `add_dir_header`: if true, adds the file directory to the header of the log messages

pkg/common/config.go

@@ -314,6 +314,7 @@ func (c *Configuration) validate() error {
 	if c.EventBatchSize < 1 {
 		return errors.New("event batch size cannot less than 1")
 	}
+
 
   if c.FailureInjectionRate < 0 || c.FailureInjectionRate > 100 {
 		return errors.New("failure injection rate should be between 0 and 100")
@@ -332,7 +333,6 @@ func (c *Configuration) validate() error {
 			return fmt.Errorf("invalid failure type '%s', valid types are: rate_limit, invalid_api_key, context_length, server_error, invalid_request, model_not_found", failureType)
 		}
 	}
-
 	return nil
 }
 
@@ -385,7 +385,7 @@ func ParseCommandParamsAndLoadConfig() (*Configuration, error) {
 	f.StringVar(&config.HashSeed, "hash-seed", config.HashSeed, "Seed for hash generation (if not set, is read from PYTHONHASHSEED environment variable)")
 	f.StringVar(&config.ZMQEndpoint, "zmq-endpoint", config.ZMQEndpoint, "ZMQ address to publish events")
 	f.IntVar(&config.EventBatchSize, "event-batch-size", config.EventBatchSize, "Maximum number of kv-cache events to be sent together")
-	
+
   f.IntVar(&config.FailureInjectionRate, "failure-injection-rate", config.FailureInjectionRate, "Probability (0-100) of injecting failures")
 
 	failureTypes := getParamValueFromArgs("failure-types")