llm, add a few more ollama env variables as well as set the keep alive to -1 (infinite).

tasks/workstation/linux/software/llm.yml

@@ -207,6 +207,8 @@
             owner: "{{ user }}"
             mode: "0755"
 
+        # https://docs.ollama.com/faq#setting-environment-variables-on-linux
+        # https://modelpiper.com/blog/ollama-environment-variables
         - name: Workstation | Linux | Software | LLMs | runit | ollama | Script | Run
           ansible.builtin.copy:
             dest: "{{ ollama_runit_script }}"
@@ -223,7 +225,7 @@
               export OLLAMA_REQUEST_TIMEOUT=3600 # default 300 (unconfirmed, not in the list of --help vars)
 
               ### Resource Usage ###
-              export OLLAMA_KEEP_ALIVE=2h # default is 5m
+              export OLLAMA_KEEP_ALIVE=-1 # default is 5m
               export OLLAMA_MAX_LOADED_MODELS={{ llm_concurrency }} # default is 3 if using CPUs, or 3 * NumGPUs
               export OLLAMA_NUM_PARALLEL={{ llm_parallel }} # default is 1 and applies per-model (unconfirmed)
               export OLLAMA_CONTEXT_LENGTH={{ llm_context }} # default is 4096, https://docs.ollama.com/faq#how-can-i-specify-the-context-window-size
@@ -239,6 +241,12 @@
               ### Tokens ###
               {{ ("export HF_TOKEN=" ~ hf_token) if hf_token else '' }}
 
+              ### Logging ###
+              export OLLAMA_DEBUG=1
+
+              ### Other ###
+              export OLLAMA_ORIGINS=*
+
               exec chpst -u {{ user }} ollama serve
             mode: "0755"