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:
xiguiw
2025-01-20 20:13:13 +08:00
committed by GitHub
parent 59722d2bc9
commit 2d5898244c
17 changed files with 65 additions and 24 deletions

View File

@@ -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