You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Does apps_script mode have a hard limit on the number of Deployment IDs? #1207
@sparkling-firefly — no hard-coded limit on script_ids count in apps_script mode. The client iterates whatever is in your config.json array.
Practical considerations even though no software limit:
Round-robin distribution: mhrv-rs cycles through all script_ids for each request. More IDs = more spread, but per-request latency is dominated by Apps Script edge RTT (~200-400ms), not script count.
Quota math: each script_id is owned by ONE Google account. Apps Script daily quota is per-account (~5K-20K UrlFetchApp calls depending on Google account age/region). 10 script_ids from 10 separate Google accounts = 10× quota. 10 script_ids from 1 Google account = same quota as 1 ID.
Marginal returns: past 3-5 healthy IDs from different accounts, adding more rarely helps. The bottleneck shifts from quota to Apps Script edge RTT and ISP network latency.
Probe overhead: at startup, mhrv-rs warms each ID with a probe (fetch_ips_from_api does some discovery work). 100 IDs = ~100 probe RTTs at startup = ~30-60s warm-up time.
Config file size: config.json parsing is bounded; 1000+ entries would work but isn't tested. Keep under 50 entries for sanity.
If ALL IDs in your config are throttled simultaneously (Apps Script regional throttle), the proxy fails until any one recovers — no graceful degradation back to direct.
Recommendation: 2-5 script_ids from 2-5 separate Google accounts is the practical sweet spot. More than that is diminishing returns.
میبندم — no software limit, but architectural ceiling is around 5-10 useful IDs.
Is there a hard-coded limit on the number of Deployment IDs the client will actually use in apps_script mode?
Thank you.