websocket connection closed

[2019jack]

2020-12-06T18:26:33.011+0800 WARN storageminer storage/wdpost_sched.go:108 window post scheduler notifs channel closed
2020-12-06T18:26:33.011+0800 ERROR miner miner/miner.go:176 failed to get best mining candidate: handler: websocket connection closed
2020-12-06T18:26:33.011+0800 WARN events events/events.go:97 listenHeadChanges quit
2020-12-06T18:26:33.011+0800 ERROR storageminer storage/wdpost_sched.go:96 ChainNotify error: handler: websocket connection closed
2020-12-06T18:26:33.011+0800 WARN events events/events.go:97 listenHeadChanges quit
2020-12-06T18:26:34.011+0800 INFO events events/events.go:104 restarting listenHeadChanges
2020-12-06T18:26:34.011+0800 INFO events events/events.go:104 restarting listenHeadChanges
2020-12-06T18:26:34.013+0800 WARN events events/events.go:132 tsc.add: adding current tipset failed: tipSetCache.add: expected new tipset height to be at least 298133, was 298132
2020-12-06T18:26:34.015+0800 WARN events events/events.go:132 tsc.add: adding current tipset failed: tipSetCache.add: expected new tipset height to be at least 298133, was 298132

This is the complete error message of the miner log

Daemon: 1.2.2+git.b13226bc2+api1.0.0
Local: lotus version 1.2.2+git.b13226bc2

@magik6k @dirkmc @Kubuxu

[askender]

same here. lotus 1.4.0

It also make filecoin wasted:
PublishStorageDeals is onchain and burned 0.25FIL, but still got falled:

Dec 24 20:44:42 bafyreihxmyk6bj5yhctdei2b3djescyfmn7pv7zvzp4qltq5k36vv2ptby 0 StorageDealError f3uvwjknloel3zcmjrgrbmgqwp3c32l7b7tyoqj7p7njb3c2xmex6dh2jzz45a6smzqvhw53nbng2zn2p7ciza 16GiB 0 FIL 1443101 PublishStorageDeal error: PublishStorageDeals errored: handler: websocket connection closed

[askender]

2021-01-13T10:55:07.169+0800    ERROR   rpc     go-jsonrpc@v0.1.2/websocket.go:654      Connection timeout      {"remote": "127.0.0.1:1234"}
2021-01-13T10:55:07.169+0800    WARN    storageminer    storage/wdpost_sched.go:108     window post scheduler notifs channel closed
2021-01-13T10:55:07.169+0800    WARN    events  events/events.go:100    listenHeadChanges quit
2021-01-13T10:55:07.169+0800    WARN    events  events/events.go:100    listenHeadChanges quit
2021-01-13T10:55:07.169+0800    ERROR   storageminer    storage/wdpost_sched.go:96      ChainNotify error: RPC client error: sendRequest failed: websocket routine exiting
2021-01-13T10:55:07.169+0800    WARN    rpc     go-jsonrpc@v0.1.2/websocket.go:289      failed to send request  {"method": "xrpc.cancel", "id": 28388, "error": "write tcp 127.0.0.1:56318->127.0.0.1:1234: use of closed network connection"}
2021-01-13T10:55:07.169+0800    ERROR   miner   miner/miner.go:230      mining block failed: failed to get mining base info:
    github.com/filecoin-project/lotus/miner.(*Miner).mineOne
        /home/zzz/lotus/miner/miner.go:365
  - handler: websocket connection closed
2021-01-13T10:55:07.169+0800    WARN    rpc     go-jsonrpc@v0.1.2/websocket.go:289      failed to send request  {"method": "xrpc.cancel", "id": 28387, "error": "write tcp 127.0.0.1:56318->127.0.0.1:1234: use of closed network connection"}
2021-01-13T10:55:07.169+0800    ERROR   rpc     go-jsonrpc@v0.1.2/websocket.go:497      sending ping message: write tcp 127.0.0.1:56318->127.0.0.1:1234: use of closed network connection
2021-01-13T10:55:08.169+0800    INFO    events  events/events.go:107    restarting listenHeadChanges
2021-01-13T10:55:08.169+0800    ERROR   miner   miner/miner.go:176      failed to get best mining candidate: RPC client error: sendRequest failed: websocket routine exiting
2021-01-13T10:55:08.169+0800    INFO    events  events/events.go:107    restarting listenHeadChanges
2021-01-13T10:55:08.169+0800    ERROR   events  events/events.go:98     listen head changes errored: listenHeadChanges Cha
inNotify call failed: RPC client error: sendRequest failed: websocket routine exiting

[askender]

2021-01-13T12:17:10.925+0800 ERROR rpc go-jsonrpc@v0.1.2/websocket.go:121 handle me:write tcp4 127.0.0.1:1234->127.0.0.1:35582: write: broken pipe Why broken pipe and timeout? daemon&miner are in same machine. Why not retry? One timeout will result to window post scheduler notifs channel closed. Is it really good design?