adjust filer startup

[zemul]

What problem are we solving?

During the master election, the filer port cannot be started.

How are we solving the problem?

Remove unnecessary block.

How is the PR tested?

Checks

• I have added unit tests if possible.
• I will add related wiki document changes and link to this PR after merging.

[chrislusf]

why this is a problem?

[zemul]

why this is a problem?

In 3.34 my master was not selected for 4 minutes.
meanwhile, my filer will not start successfully for a long time.

[zemul]

filer log

Log file created at: 2022/12/02 10:58:59
Running on machine: hn57
Binary: Built with gc go1.18.2 for linux/amd64
Log line format: [IWEF]mmdd hh:mm:ss threadid file:line] msg
W1202 10:58:59.997221 filer_server.go:142 skipping default store dir in ./filerldb2
I1202 10:59:00.006877 filer.go:130 existing filer.store.id = 1955893051
I1202 10:59:00.006906 configuration.go:28 configured filer store to mysql
I1202 10:59:35.954871 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 10:59:35.954973 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.59:9333: EOF
I1202 11:00:11.958809 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 11:00:11.958869 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.57:9333: EOF
I1202 11:00:47.964309 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 11:00:47.964370 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.58:9333: EOF
I1202 11:01:24.969882 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 11:01:24.969949 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.58:9333: EOF
I1202 11:02:00.974008 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 11:02:00.974065 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.59:9333: EOF
I1202 11:02:36.979647 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 11:02:36.979716 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.57:9333: EOF
I1202 11:03:13.983438 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 11:03:13.983509 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.57:9333: EOF
I1202 11:03:19.413518 masterclient.go:200 .filer masterClient failed to receive from 172.16.254.58:9333: rpc error: code = Unavailable desc = error reading from server: EOF
I1202 11:03:55.416575 masterclient.go:266 updateVidMap ignore short heartbeat: volume_location:{}
I1202 11:03:55.416637 masterclient.go:223 .filer masterClient failed to receive from 172.16.254.59:9333: EOF
I1202 11:04:19.421419 masterclient.go:208 master 172.16.254.58:9333 redirected to leader 172.16.254.57:9333
I1202 11:04:19.439025 master_client.go:20 the cluster has 2 filer
I1202 11:04:19.441596 meta_aggregator.go:97 loopSubscribeToOneFiler read 172.16.254.59:8888 start from 2022-12-02 11:03:19.439038516 +0800 CST 1669950199439038516
I1202 11:04:19.441655 meta_aggregator.go:97 loopSubscribeToOneFiler read 172.16.254.57:8888 start from 2022-12-02 11:03:19.439038516 +0800 CST 1669950199439038516
I1202 11:04:19.442078 meta_aggregator.go:108 subscribing remote 172.16.254.57:8888 meta change: connecting to peer filer 172.16.254.57:8888: rpc error: code = Unavailable desc = connection error: desc = "transport: Error while dialing dial tcp 172.16.254.57:18888: connect: connection refused"
I1202 11:04:19.446425 meta_aggregator.go:194 subscribing remote 172.16.254.59:8888 meta change: 2022-12-02 11:03:19.439038516 +0800 CST, clientId:1923813785
I1202 11:04:19.449388 filer.go:272 Start Seaweed Filer 30GB 3.34 at 172.16.254.57:8888

[zemul]

I found that this pull request only solves part of the problem.

seaweedfs/weed/server/filer_server.go

Line 165 in 6b4c033

existingNodes := fs.filer.ListExistingPeerUpdates()

There's also a blockage here