Data Source: Ollama Server

Description

Ollama server logs (HTTP access logs via GIN framework and system logs including GPU/CPU utilization, model loading, memory allocation, errors, and warnings) via Splunk TA-ollama add-on by configuring file monitoring inputs to your log directories (sourcetype: ollama:server), or enable HEC for real-time API telemetry and prompt analytics (sourcetypes: ollama:api, ollama:prompts). TA available in Splunkbase

Details

Property Value
Source server.log
Sourcetype ollama:server
Name ▲▼ Technique ▲▼ Type ▲▼
Ollama Abnormal Network Connectivity Non-Standard Port Anomaly
Ollama Abnormal Service Crash Availability Attack Service Stop Anomaly
Ollama Excessive API Requests Network Denial of Service Anomaly
Ollama Possible API Endpoint Scan Reconnaissance Active Scanning Anomaly
Ollama Possible Memory Exhaustion Resource Abuse Endpoint Denial of Service Anomaly
Ollama Possible Model Exfiltration Data Leakage Exfiltration Over Alternative Protocol Anomaly
Ollama Possible RCE via Model Loading Exploit Public-Facing Application Anomaly
Ollama Suspicious Prompt Injection Jailbreak Exploit Public-Facing Application, Command and Scripting Interpreter Anomaly

Supported Apps

Event Fields

+ Fields
  <span class="pill kill-chain">CPU_0_AVX</span>
  
  <span class="pill kill-chain">CPU_0_AVX2</span>
  
  <span class="pill kill-chain">CPU_0_AVX_VNNI</span>
  
  <span class="pill kill-chain">CPU_0_BMI2</span>
  
  <span class="pill kill-chain">CPU_0_F16C</span>
  
  <span class="pill kill-chain">CPU_0_FMA</span>
  
  <span class="pill kill-chain">CPU_0_LLAMAFILE</span>
  
  <span class="pill kill-chain">CPU_0_SSE3</span>
  
  <span class="pill kill-chain">CPU_0_SSSE3</span>
  
  <span class="pill kill-chain">CPU_1_LLAMAFILE</span>
  
  <span class="pill kill-chain">CUDA_0_ARCHS</span>
  
  <span class="pill kill-chain">CUDA_0_PEER_MAX_BATCH_SIZE</span>
  
  <span class="pill kill-chain">CUDA_0_USE_GRAPHS</span>
  
  <span class="pill kill-chain">LOG</span>
  
  <span class="pill kill-chain">OS</span>
  
  <span class="pill kill-chain">app</span>
  
  <span class="pill kill-chain">args</span>
  
  <span class="pill kill-chain">available</span>
  
  <span class="pill kill-chain">bundle</span>
  
  <span class="pill kill-chain">cmd</span>
  
  <span class="pill kill-chain">compiler</span>
  
  <span class="pill kill-chain">compute</span>
  
  <span class="pill kill-chain">cores</span>
  
  <span class="pill kill-chain">count</span>
  
  <span class="pill kill-chain">date_hour</span>
  
  <span class="pill kill-chain">date_mday</span>
  
  <span class="pill kill-chain">date_minute</span>
  
  <span class="pill kill-chain">date_month</span>
  
  <span class="pill kill-chain">date_second</span>
  
  <span class="pill kill-chain">date_wday</span>
  
  <span class="pill kill-chain">date_year</span>
  
  <span class="pill kill-chain">date_zone</span>
  
  <span class="pill kill-chain">dest</span>
  
  <span class="pill kill-chain">driver</span>
  
  <span class="pill kill-chain">efficiency</span>
  
  <span class="pill kill-chain">env</span>
  
  <span class="pill kill-chain">eventtype</span>
  
  <span class="pill kill-chain">free</span>
  
  <span class="pill kill-chain">free_swap</span>
  
  <span class="pill kill-chain">gpus</span>
  
  <span class="pill kill-chain">host</span>
  
  <span class="pill kill-chain">http_d</span>
  
  <span class="pill kill-chain">http_method</span>
  
  <span class="pill kill-chain">http_path</span>
  
  <span class="pill kill-chain">http_pattern</span>
  
  <span class="pill kill-chain">http_response_code</span>
  
  <span class="pill kill-chain">http_status</span>
  
  <span class="pill kill-chain">id</span>
  
  <span class="pill kill-chain">index</span>
  
  <span class="pill kill-chain">installer</span>
  
  <span class="pill kill-chain">interval</span>
  
  <span class="pill kill-chain">layers_model</span>
  
  <span class="pill kill-chain">layers_offload</span>
  
  <span class="pill kill-chain">layers_requested</span>
  
  <span class="pill kill-chain">layers_split</span>
  
  <span class="pill kill-chain">level</span>
  
  <span class="pill kill-chain">library</span>
  
  <span class="pill kill-chain">linecount</span>
  
  <span class="pill kill-chain">maxEfficiencyClass</span>
  
  <span class="pill kill-chain">memory_available</span>
  
  <span class="pill kill-chain">memory_gpu_overhead</span>
  
  <span class="pill kill-chain">memory_graph_full</span>
  
  <span class="pill kill-chain">memory_graph_partial</span>
  
  <span class="pill kill-chain">memory_required_allocations</span>
  
  <span class="pill kill-chain">memory_required_full</span>
  
  <span class="pill kill-chain">memory_required_kv</span>
  
  <span class="pill kill-chain">memory_required_partial</span>
  
  <span class="pill kill-chain">memory_weights_nonrepeating</span>
  
  <span class="pill kill-chain">memory_weights_repeating</span>
  
  <span class="pill kill-chain">memory_weights_total</span>
  
  <span class="pill kill-chain">model</span>
  
  <span class="pill kill-chain">msg</span>
  
  <span class="pill kill-chain">name</span>
  
  <span class="pill kill-chain">overhead</span>
  
  <span class="pill kill-chain">package</span>
  
  <span class="pill kill-chain">parallel</span>
  
  <span class="pill kill-chain">port</span>
  
  <span class="pill kill-chain">punct</span>
  
  <span class="pill kill-chain">request</span>
  
  <span class="pill kill-chain">request_id</span>
  
  <span class="pill kill-chain">required</span>
  
  <span class="pill kill-chain">response_time_ms</span>
  
  <span class="pill kill-chain">source</span>
  
  <span class="pill kill-chain">sourcetype</span>
  
  <span class="pill kill-chain">splunk_server</span>
  
  <span class="pill kill-chain">status</span>
  
  <span class="pill kill-chain">threads</span>
  
  <span class="pill kill-chain">threshold</span>
  
  <span class="pill kill-chain">time</span>
  
  <span class="pill kill-chain">timeendpos</span>
  
  <span class="pill kill-chain">timestartpos</span>
  
  <span class="pill kill-chain">tool_count</span>
  
  <span class="pill kill-chain">total</span>
  
  <span class="pill kill-chain">variant</span>
  
  <span class="pill kill-chain">vendor_product</span>
  
  <span class="pill kill-chain">version</span>
  
</div>

Example Log

1time=2025-10-02T14:46:19.789-04:00 level=INFO source=server.go:544 msg=offload library=cuda layers.requested=-1 layers.model=29 layers.offload=29 layers.split=[29] memory.available="[6.9 GiB]" memory.gpu_overhead="0 B" memory.required.full="3.1 GiB" memory.required.partial="3.1 GiB" memory.required.kv="448.0 MiB" memory.required.allocations="[3.1 GiB]" memory.weights.total="1.9 GiB" memory.weights.repeating="1.6 GiB" memory.weights.nonrepeating="308.2 MiB" memory.graph.full="256.5 MiB" memory.graph.partial="570.7 MiB"

Source: GitHub | Version: 2