all: stop using math/rand.Seed

[callthingsoff]

Bug report criteria

• This bug report is not security related, security issues should be disclosed privately via security@etcd.io.
• This is not a support request, support requests should be raised in the etcd discussion forums.
• You have read the etcd bug reporting guidelines.
• Existing open issues along with etcd frequently asked questions have been checked and this is not a duplicate.

What happened?

It's updated to Go 1.20 minor release, however math/rand.Seed is marked "Deprecated":

// Prior to Go 1.20, the generator was seeded like Seed(1) at program startup.
// To force the old behavior, call Seed(1) at program startup.
// Alternately, set GODEBUG=randautoseed=0 in the environment
// before making any calls to functions in this package.
//
// Deprecated: As of Go 1.20 there is no reason to call Seed with
// a random value. Programs that call Seed with a known value to get
// a specific sequence of results should use New(NewSource(seed)) to
// obtain a local random generator.

What did you expect to happen?

Stop using math/rand.Seed.

How can we reproduce it (as minimally and precisely as possible)?

Use the new API as suggested.

Anything else we need to know?

No response

Etcd version (please run commands below)

Versions updated to Go 1.20.

Etcd configuration (command line flags or environment variables)

No response

Etcd debug information (please run commands below, feel free to obfuscate the IP address or FQDN in the output)

No response

Relevant log output

No response

[jmhbnz]

Thanks for raising this @gocurr. More context here for anyone interested: golang/go#54880

Looks like eight instances to consider updating:

 ➜ find . -name "*.go" -type f -print | xargs grep .Seed
./pkg/stringutil/rand.go:                                   rand.Seed(time.Now().UnixNano())
./pkg/adt/interval_tree_test.go:                            rand.Seed(time.Now().UnixNano())
./pkg/proxy/server.go:                                      mrand.Seed(int64(now.Nanosecond()))
./pkg/proxy/server_test.go:                                 rand.Seed(now)
./tests/integration/clientv3/snapshot/v3_snapshot_test.go:  rand.Seed(int64(time.Now().Nanosecond()))
./server/lease/lessor_bench_test.go:                        rand.Seed(time.Now().UTC().UnixNano())
./server/etcdserver/server.go:                              rand.Seed(time.Now().UnixNano())                                       

[callthingsoff]

I think perhaps you should check other deprecated APIs.

[jmhbnz]

I think perhaps you should check other deprecated APIs.

The only other formal deprecation I can see that impacts us for 1.20 is deprecation of math/rand.Read in favor of crypto/rand.Read. I'll raise a separate issue for that. Please let us know if you spot anything else that needs a cleanup.

[callthingsoff]

Anyone want to fix it?

[jmhbnz]

We welcome contributions so please feel free if you have capacity! Happy to assign it to you just let us know.

[callthingsoff]

Seems to me, there are two approaches.

1. Since "If Seed is not called, the generator is seeded randomly at program startup." just remove rand.Seed calls.
2. "Programs that call Seed with a known value to get a specific sequence of results should use New(NewSource(seed)) to obtain a local random generator." it depends on the situations.

[Aditya-Sood]

hi @gocurr @jmhbnz
is the fix in-progress or open for contribution?

[jmhbnz]

I'm not working on this currently. @gocurr did you make a start on this?

• Since "If Seed is not called, the generator is seeded randomly at program startup." just remove rand.Seed calls.
• "Programs that call Seed with a known value to get a specific sequence of results should use New(NewSource(seed)) to obtain a local random generator." it depends on the situations.

Ideally we should be just removing the .Seed calls.

[callthingsoff]

hi @gocurr @jmhbnz
is the fix in-progress or open for contribution?

A PR to this issue was just created.

[callthingsoff]

I'm not working on this currently. @gocurr did you make a start on this?

• Since "If Seed is not called, the generator is seeded randomly at program startup." just remove rand.Seed calls.
• "Programs that call Seed with a known value to get a specific sequence of results should use New(NewSource(seed)) to obtain a local random generator." it depends on the situations.

Ideally we should be just removing the .Seed calls.

Please take a look.