Summary
Single beta soak node on macOS: chunks.mdb is growing several GiB per hour with no cap in the running config (storage.db_size_gb: 0). Disk will fill if this continues.
Environment
- Client:
ant 0.3.4-beta.2
- Node: 0.17.2-beta.2 commit
8020eed
- OS: macOS, one node (
node1)
- Upgrade channel: beta,
stop_on_upgrade: true
- EVM: Arbitrum One
- Log forwarding: on, 0 failed batches
Startup line:
2026-08-21T06:50:40Z INFO ant_node: ant-node starting version="0.17.2-beta.2" commit="8020eed"
Config excerpt from the same log:
storage: StorageConfig { enabled: true, verify_on_read: true, db_size_gb: 0, disk_reserve_mb: 500 }
Observed sizes (Europe/London)
| Time (uptime) |
chunks.mdb |
Home volume free |
| ~15 min |
1.4 GiB |
257 GiB |
| ~1h 13m |
11 GiB |
244 GiB |
| ~2h 14m |
17 GiB (+6 GiB/h) |
238 GiB |
| ~3h 14m |
26 GiB (+9 GiB/h) |
229 GiB |
Node stays Running / Healthy. Forwarding 2615 batches, 0 failed. Only ERROR in the log is a one-off at startup:
2026-08-21T06:51:56Z ERROR saorsa_transport::high_level::endpoint: I/O error: relay recv stream closed
Not a crash loop. Growth is not slowing.
Ask
Is unbounded db_size_gb: 0 expected on beta, or should the node refuse/store-cap before filling the volume? Happy to add more log excerpts if useful.
Summary
Single beta soak node on macOS:
chunks.mdbis growing several GiB per hour with no cap in the running config (storage.db_size_gb: 0). Disk will fill if this continues.Environment
ant0.3.4-beta.28020eednode1)stop_on_upgrade: trueStartup line:
Config excerpt from the same log:
Observed sizes (Europe/London)
Node stays Running / Healthy. Forwarding 2615 batches, 0 failed. Only ERROR in the log is a one-off at startup:
Not a crash loop. Growth is not slowing.
Ask
Is unbounded
db_size_gb: 0expected on beta, or should the node refuse/store-cap before filling the volume? Happy to add more log excerpts if useful.