Cannot reseed a replica in the 'master-master' replication tutorial

[andreyaksenov]

1. The Inserting conflicting records, step 6 - startup_in_progress is shown for some time for instance001:

master_master:instance001> require('config'):info()
---
- status: startup_in_progress
  meta: []
  alerts:
  - type: warn
    message: 'box_cfg.apply: instance instance002 is unknown. Possibly instance_name
      is not set in database and UUID is not specified. Or instance have not joined
      yet.'
    timestamp: 2024-04-11T10:18:50.361880+0300
...

Then, the instance is in the ready state:

master_master:instance001> require('config'):info()
---
- status: ready
  meta: []
  alerts: []
...

2. The Reseeding a replica, step 5 - instance002 cannot run and shows a startup_in_progress status:

master_master:instance002> require('config'):info()
---
- status: startup_in_progress
  meta: []
  alerts: []
...

The ER_INSTANCE_NAME_DUPLICATE in logs of instance002:

2024-04-11 10:26:22.917 [5517] main/103/interactive I> Tarantool 3.0.1-0-g31c2ddb Darwin-arm64-Release
2024-04-11 10:26:22.917 [5517] main/103/interactive I> log level 5 (INFO)
2024-04-11 10:26:22.917 [5517] main/103/interactive I> wal/engine cleanup is paused
2024-04-11 10:26:22.917 [5517] main/103/interactive I> mapping 268435456 bytes for memtx tuple arena...
2024-04-11 10:26:22.917 [5517] main/103/interactive I> Actual slab_alloc_factor calculated on the basis of desired slab_alloc_factor = 1.044274
2024-04-11 10:26:22.917 [5517] main/103/interactive I> mapping 134217728 bytes for vinyl tuple arena...
2024-04-11 10:26:22.918 [5517] main/103/interactive I> update replication_synchro_quorum = 1
2024-04-11 10:26:22.918 [5517] main/103/interactive I> instance uuid 6df01da6-f29b-4574-91fd-3059f7b9c2ef
2024-04-11 10:26:22.918 [5517] main/103/interactive I> tx_binary: bound to 127.0.0.1:3302
2024-04-11 10:26:22.918 [5517] main/103/interactive I> connecting to 2 replicas
2024-04-11 10:26:22.918 [5517] main/111/applier/replicator@127.0.0.1:3301 I> remote master 8eba170d-f4a6-4087-a0fe-ee83692f3262 at 127.0.0.1:3301 running Tarantool 3.0.1
2024-04-11 10:26:22.918 [5517] main/112/applier/replicator@127.0.0.1:3302 I> remote master 6df01da6-f29b-4574-91fd-3059f7b9c2ef at 127.0.0.1:3302 running Tarantool 3.0.1
2024-04-11 10:26:22.919 [5517] main/103/interactive I> connected to 2 replicas
2024-04-11 10:26:22.919 [5517] main/111/applier/replicator@127.0.0.1:3301 I> authenticated
2024-04-11 10:26:22.919 [5517] main/103/interactive I> bootstrapping replica from 8eba170d-f4a6-4087-a0fe-ee83692f3262 at 127.0.0.1:3301
2024-04-11 10:26:22.920 [5517] main/111/applier/replicator@127.0.0.1:3301 I> can't read row
2024-04-11 10:26:22.920 [5517] main/111/applier/replicator@127.0.0.1:3301 box.cc:4444 E> ER_INSTANCE_NAME_DUPLICATE: Dupl

Reason

Starting with v3.0, Tarantool generates unique UUIDs that doesn't depend on the instance name. It is necessary to rebootstrap a second instance with the old UUID.