SchemaUnknownException: Schema 'sys' unknown

CrateDB
1

I’m having a weird problem.

I have a 3 nodes cluster running.

When I run crash inside one of the nodes, everything works fine and SELECT name, load['1'], os_info['jvm']['version'] FROM sys.nodes; has this output:

cr> SELECT name, load['1'], os_info['jvm']['version'] FROM sys.nodes;
+--------------------+-----------+---------------------------+
| name               | load['1'] | os_info['jvm']['version'] |
+--------------------+-----------+---------------------------+
| Schwarzberg        |      0.0  | 11.0.7                    |
| Muntejela de Senes |      0.0  | 11.0.7                    |
| Rochers de Naye    |      0.01 | 11.0.7                    |
+--------------------+-----------+---------------------------+

Which tells me the system can find the other two nodes.

But, when I connect to another node with crash, then I get this error:

./crash --verbose --host 10.0.0.195 -U defaultuser
Password:
+------------------------+--------------------+---------+-----------+---------+
| server_url             | node_name          | version | connected | message |
+------------------------+--------------------+---------+-----------+---------+
| http://10.0.0.195:4200 | Muntejela de Senes | 4.1.5   | TRUE      | OK      |
+------------------------+--------------------+---------+-----------+---------+
CONNECT OK
SQLActionException[SchemaUnknownException: Schema 'sys' unknown]

SQLActionException: 404 Not Found 4045 SchemaUnknownException: Schema 'sys' unknown

I have no clue what’s happening.

And the WebUI looks super weird too, and it’s not working.

Screen Shot 2020-05-29 at 12.41.00
Screen Shot 2020-05-29 at 12.41.001897×648 15.9 KB

I’m using exactly the same configuration that is working with other projects. So, I don’t know what can be wrong.

crate.yml

bootstrap.memory_lock : true
path.data: /data
gateway.expected_nodes: 3              # Total amount of nodes
gateway.recover_after_nodes: 2         # More than half of the nodes
network.host: _site_, _local_
discovery.seed_hosts:
    - 10.0.0.22:4300
    - 10.0.0.153:4300
    - 10.0.0.195:4300
cluster.initial_master_nodes:
    - 10.0.0.195
    - 10.0.0.22
    - 10.0.0.153

Maybe someone had the same problem.

2

It was a problem with the permissions.

The user crate had all privileges granted, while the defaultuser didn’t.

./crash
CONNECT OK
cr> GRANT ALL PRIVILEGES TO defaultuser;

solved the problem.