Enchance health check in GenAIExample docker-compose (#1410)
Fix service launch issue 1. Update Gaudi TGI image from 2.0.6 to 2.3.1 2. Change the hpu-gaudi TGI health check condition. Signed-off-by: Wang, Xigui <xigui.wang@intel.com>
This commit is contained in:
@@ -91,7 +91,7 @@ services:
|
||||
RETRIEVER_COMPONENT_NAME: "OPEA_RETRIEVER_REDIS"
|
||||
restart: unless-stopped
|
||||
tgi-gaudi:
|
||||
image: ghcr.io/huggingface/tgi-gaudi:2.0.6
|
||||
image: ghcr.io/huggingface/tgi-gaudi:2.3.1
|
||||
container_name: tgi-llava-gaudi-server
|
||||
ports:
|
||||
- "8399:80"
|
||||
@@ -110,6 +110,11 @@ services:
|
||||
LIMIT_HPU_GRAPH: true
|
||||
USE_FLASH_ATTENTION: true
|
||||
FLASH_ATTENTION_RECOMPUTE: true
|
||||
healthcheck:
|
||||
test: ["CMD-SHELL", "curl -f http://${host_ip}:8399/health || exit 1"]
|
||||
interval: 10s
|
||||
timeout: 10s
|
||||
retries: 100
|
||||
runtime: habana
|
||||
cap_add:
|
||||
- SYS_NICE
|
||||
|
||||
Reference in New Issue
Block a user