caivega / ipfslib

Other
2 stars 1 forks source link

S2 - 关闭某个节点的网络,可能导致其他节点丢失网卡ip信息并崩溃 #116

Closed foreso-GitHub closed 4 years ago

foreso-GitHub commented 4 years ago

关闭了百度节点的udp7001端口,刚才发现阿里、腾讯、天翼3个节点自己重启,应该是出错崩溃了。 根据log的内容分析,应该是触发了某段代码,将网卡的ip信息(网卡信息或者节点网络IP之类)都丢失了。

E1026 16:30:10.185562   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:10.212412   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:10.232643   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:10.259909   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:10.273157   52954 consensus.go:614] exists block C5F6CEEC1F1C0E67D76ECEFEF9D1A8215E6D$
E1026 16:30:10.273246   52954 consensus.go:743] drop block C5F6CEEC1F1C0E67D76ECEFEF9D1A8215E6DBB$
E1026 16:30:10.273252   52954 network.go:542] drop block C5F6CEEC1F1C0E67D76ECEFEF9D1A8215E6DBB0F$
E1026 16:30:10.281591   52954 consensus.go:614] exists block C5F6CEEC1F1C0E67D76ECEFEF9D1A8215E6D$
E1026 16:30:10.281672   52954 consensus.go:743] drop block C5F6CEEC1F1C0E67D76ECEFEF9D1A8215E6DBB$
E1026 16:30:10.281682   52954 network.go:542] drop block C5F6CEEC1F1C0E67D76ECEFEF9D1A8215E6DBB0F$
E1026 16:30:15.005004   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.136761   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.140846   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.141252   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.155741   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.176853   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.202837   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.223796   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.249938   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:15.269498   52954 consensus.go:614] exists block 497355ADBB5F3BED661D6C7AC0E4A91696E7$
E1026 16:30:15.269553   52954 consensus.go:743] drop block 497355ADBB5F3BED661D6C7AC0E4A91696E713$
E1026 16:30:15.269560   52954 network.go:542] drop block 497355ADBB5F3BED661D6C7AC0E4A91696E713E5$
E1026 16:30:15.270429   52954 consensus.go:614] exists block 497355ADBB5F3BED661D6C7AC0E4A91696E7$
E1026 16:30:15.270453   52954 consensus.go:743] drop block 497355ADBB5F3BED661D6C7AC0E4A91696E713$
E1026 16:30:15.270459   52954 network.go:542] drop block 497355ADBB5F3BED661D6C7AC0E4A91696E713E5$
E1026 16:30:17.501036   52954 network.go:1032] >>> error connecting to network peer 12D3KooWBaR65$
E1026 16:30:17.639422   52954 peer.go:577] NO_ERROR: No recent network activity
E1026 16:30:17.639523   52954 peer.go:577] NO_ERROR: No recent network activity
E1026 16:30:17.639586   52954 peer.go:577] NO_ERROR: No recent network activity
E1026 16:30:17.639624   52954 peer.go:577] NO_ERROR: No recent network activity
E1026 16:30:17.639660   52954 peer.go:577] NO_ERROR: No recent network activity
E1026 16:30:17.639710   52954 peer.go:577] NO_ERROR: No recent network activity
E1026 16:30:20.000802   52954 network.go:1189] >>>>>> peer 12D3KooWBaR65cviWrbKKmrJuCNRVY3bYZhwXg$
E1026 16:30:20.001013   52954 protocol.go:830] Write on stream 16 canceled with error code 0
E1026 16:30:20.143673   52954 node.go:688] droped the unused request message, 1690291, 0
E1026 16:30:20.143683   52954 node.go:688] droped the unused request message, 1690291, 0
E1026 16:30:20.179586   52954 node.go:688] droped the unused request message, 1690291, 0
foreso-GitHub commented 4 years ago

补充:节点断网后再发送大量交易的情况下更容易重现这个问题。 相关issue: #108

foreso-GitHub commented 4 years ago

已经在20201027版本中修正。