invalid json will make cluster_mgr restart

[jd-zhang]

*Issue migrated from trac ticket # 855 www.kunlunbase.com *

component: cluster manager | priority: major

2022-06-27 16:52:45: zhangjindong@zettadb.com created the issue

---

info:

kunlun@kunlun-test2:~/kunlun_scripts/cluster$ curl -d '{"job_type": "get_meta_summary", "version":"1.0"}' -X POST http://192.168.0.111:58000/HttpService/Emit
{"attachment":{"list_meta":[{"hostaddr":"192.168.0.110","master":"true","port":"6001","status":"online"},{"hostaddr":"192.168.0.111","master":"false","port":"6002","status":"online"},{"hostaddr":"192.168.0.100","master":"false","port":"6003","status":"online"}]},"error_code":"0","error_info":"","job_id":"","job_type":"","status":"done","version":"1.0"}

kunlun@kunlun-test2:~/kunlun_scripts/cluster$ curl -d '{"job_type": "get_meta_summary", "version":"1.0"' -X POST http://192.168.0.111:58000/HttpService/Emit
curl: (52) Empty reply from server
kunlun@kunlun-test2:~/kunlun_scripts/cluster$ curl -d '{"job_type": "get_meta_summary", "version":"1.0"}' -X POST http://192.168.0.111:58000/HttpService/Emit
curl: (7) Failed to connect to 192.168.0.111 port 58000: Connection refused