配置内容为 server_count :2,server_list:[{server_id:"10.4.136.137",container_ip:"10.4.136.137","device":
[
{“device_id”: “0”, “device_ip”: “10.4.8.50”, “rank_id”: “0”},
{“device_id”: “1”, “device_ip”: “10.4.8.51”, “rank_id”: “1”},
{“device_id”: “2”, “device_ip”: “10.4.8.52”, “rank_id”: “2”},
{“device_id”: “3”, “device_ip”: “10.4.8.53”, “rank_id”: “3”},
{“device_id”: “4”, “device_ip”: “10.4.8.54”, “rank_id”: “4”},
{“device_id”: “5”, “device_ip”: “10.4.8.55”, “rank_id”: “5”},
{“device_id”: “6”, “device_ip”: “10.4.8.56”, “rank_id”: “6”},
{“device_id”: “7”, “device_ip”: “10.4.8.57”, “rank_id”: “7”},
]
},
{ server_id:"10.4.136.136",container_ip:"10.4.136.136","device":
[
{“device_id”: “0”, “device_ip”: “10.4.8.40”, “rank_id”: “8”},
{“device_id”: “1”, “device_ip”: “10.4.8.41”, “rank_id”: “9”},
{“device_id”: “2”, “device_ip”: “10.4.8.42”, “rank_id”: “10”},
{“device_id”: “3”, “device_ip”: “10.4.8.43”, “rank_id”: “11”},
{“device_id”: “4”, “device_ip”: “10.4.8.44”, “rank_id”: “12”},
{“device_id”: “5”, “device_ip”: “10.4.8.45”, “rank_id”: “13”},
{“device_id”: “6”, “device_ip”: “10.4.8.46”, “rank_id”: “14”},
{“device_id”: “7”, “device_ip”: “10.4.8.47”, “rank_id”: “15”},
]}],
“status”: “completed”,
“version”: “1.0”,}。 已安装通讯正常,如果ranktable从0到15,则报错首先是有Invalid DP number per node:0错误,其次还有terminate called after throwing an instance of ´std::out_of_range´ what(): _Map_base::at,然后就kill了;1到16时报错,同样是有Invalid DP number per node:0错误,但是没有MAP那个错误,不过最后有rankid必须小于卡数错误(valueError:rank_id should be a number less than world size),请大佬帮我看一下是什么错误,什么原因导致的,如何解决
配置内容为 server_count :2,server_list:[{server_id:"10.4.136.137",container_ip:"10.4.136.137","device":
[
{“device_id”: “0”, “device_ip”: “10.4.8.50”, “rank_id”: “0”},
{“device_id”: “1”, “device_ip”: “10.4.8.51”, “rank_id”: “1”},
{“device_id”: “2”, “device_ip”: “10.4.8.52”, “rank_id”: “2”},
{“device_id”: “3”, “device_ip”: “10.4.8.53”, “rank_id”: “3”},
{“device_id”: “4”, “device_ip”: “10.4.8.54”, “rank_id”: “4”},
{“device_id”: “5”, “device_ip”: “10.4.8.55”, “rank_id”: “5”},
{“device_id”: “6”, “device_ip”: “10.4.8.56”, “rank_id”: “6”},
{“device_id”: “7”, “device_ip”: “10.4.8.57”, “rank_id”: “7”},
]
},
{ server_id:"10.4.136.136",container_ip:"10.4.136.136","device":
[
{“device_id”: “0”, “device_ip”: “10.4.8.40”, “rank_id”: “8”},
{“device_id”: “1”, “device_ip”: “10.4.8.41”, “rank_id”: “9”},
{“device_id”: “2”, “device_ip”: “10.4.8.42”, “rank_id”: “10”},
{“device_id”: “3”, “device_ip”: “10.4.8.43”, “rank_id”: “11”},
{“device_id”: “4”, “device_ip”: “10.4.8.44”, “rank_id”: “12”},
{“device_id”: “5”, “device_ip”: “10.4.8.45”, “rank_id”: “13”},
{“device_id”: “6”, “device_ip”: “10.4.8.46”, “rank_id”: “14”},
{“device_id”: “7”, “device_ip”: “10.4.8.47”, “rank_id”: “15”},
]}],
“status”: “completed”,
“version”: “1.0”,}。 已安装通讯正常,如果ranktable从0到15,则报错首先是有Invalid DP number per node:0错误,其次还有terminate called after throwing an instance of ´std::out_of_range´ what(): _Map_base::at,然后就kill了;1到16时报错,同样是有Invalid DP number per node:0错误,但是没有MAP那个错误,不过最后有rankid必须小于卡数错误(valueError:rank_id should be a number less than world size),请大佬帮我看一下是什么错误,什么原因导致的,如何解决