StorageGRID intermittent 503 errors due to nginx-gw stuck due to node weight
Applies to
NetApp StorageGRID
- Prior to 12.0.0.8
Issue
- Intermittent 503 "Service Unavailable" errors during S3 operations (PUT, GET, LIST, DELETE) against certain S3 endpoints
- Error message returned:
503 Service Unavailable: No available peers. Try again later- Following log entries can be oberved under
/var/local/log/nginx-gw/error.log
2026/07/29 17:40:32 [warn] 19645#19645: *230573771 [lua] sgbalancer.lua:1015: dynip_data_updated(): Slow start: restore weight for node 612164db-2b89-4f8a-b910-c2728fcc9b62 is less than 1 (0), node_weight=1, fraction=0.1, context: ngx.timer2026/07/29 17:40:32 [warn] 19645#19645: *230573771 [lua] sgbalancer.lua:1015: dynip_data_updated(): Slow start: restore weight for node a93e4095-65ec-4449-b8a1-a8b37a316dc5 is less than 1 (0), node_weight=1, fraction=0.1, context: ngx.timer2026/07/29 17:40:32 [warn] 19645#19645: *230573771 [lua] sgbalancer.lua:1015: dynip_data_updated(): Slow start: restore weight for node aafd18c2-1da0-432f-b662-18a415bf9288 is less than 1 (0), node_weight=1, fraction=0.1, context: ngx.timer