oxidecomputer / oxidecomputer/omicron

[sim] "Not enough datasets" 503 on create disk

Open
#3,104 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

development
Dominant language
Rust
Stars
572
Forks
97
Avg merge
2d 12h
Merged PRs (30d)
96

Description

I don't know when this was introduced bcause I haven't tried to create a disk like this in a while. When I run sim sled agent locally (on Mac — possibly relevant) with `omicron-dev run-things generally work except for creating disks. When I create a disk with the CLI like this:

oxide disk create --project prod-online --json-body /dev/stdin <<EOF
    {
      "name": "nginx",
      "description": "The nginx disk.",
      "size": 1073741824,
      "disk_source": {
        "type": "blank",
        "block_size": 512
      }
    }
EOF

I get a 503 with this internal message in the logs:

saga error at node "datasets_and_regions": Not enough datasets to allocate disks

That error comes from here:

https://github.com/oxidecomputer/omicron/blob/857a9b95a83e7c833822fc29e3b89c4d1be41386/nexus/db-queries/src/db/queries/region_allocation.rs#L332-L373

I am not shocked to find there are no regions, zpools, or datasets in the DB:

~/oxide/omicron $ cockroach sql --insecure --url "postgresql://root@127.0.0.1:64043/omicron?sslmode=disable"                                    main
#
# Welcome to the CockroachDB SQL shell.
# All statements must be terminated by a semicolon.
# To exit, type: \q.
#
# Server version: CockroachDB CCL v22.1.9 (x86_64-apple-darwin19, built 2022/10/17 18:39:01, go1.17.11) (same version as client)
# Cluster ID: d494037f-10a5-425a-ae38-bed344cd7148
#
# Enter \? for a brief introduction.
#
root@127.0.0.1:64043/omicron> select * from zpool;
  id | time_created | time_modified | time_deleted | rcgen | sled_id | physical_disk_id | total_size
-----+--------------+---------------+--------------+-------+---------+------------------+-------------
(0 rows)


Time: 2ms total (execution 1ms / network 0ms)

root@127.0.0.1:64043/omicron> select * from region;
  id | time_created | time_modified | dataset_id | volume_id | block_size | blocks_per_extent | extent_count
-----+--------------+---------------+------------+-----------+------------+-------------------+---------------
(0 rows)


Time: 2ms total (execution 1ms / network 0ms)

root@127.0.0.1:64043/omicron> select * from dataset;
  id | time_created | time_modified | time_deleted | rcgen | pool_id | ip | port | kind | size_used
-----+--------------+---------------+--------------+-------+---------+----+------+------+------------
(0 rows)


Time: 4ms total (execution 3ms / network 0ms)

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with nexus/db-queries/src/db/queries/region_allocation.rs around the linked allocation logic, then reproduce the issue with the shown oxide disk create command under the sim sled agent and omicron-dev setup. Check the local database state for the missing regions, zpools, and datasets; done means disk creation no longer returns the 503.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust
Domain
backend, databases
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.