[GH-ISSUE #1039] issue: Undocumented Env Variable: REDIS_SENTINEL_MAX_RETRY_COUNT #144

New issue

Closed

opened 2026-10-18 14:09:20 +00:00 by bsakaguchi · 1 comment

bsakaguchi commented

2026-10-18 14:09:20 +00:00

Owner

Originally created by @ctolon on GitHub (Jan 28, 2026).
Original GitHub issue: https://github.com/open-webui/docs/issues/1039

This environment variable not documented: https://docs.openwebui.com/getting-started/env-configuration/
In Sentinel deployments, a very important variable for HA (in terms of allowing time for automatic failover; if down-after-milliseconds is high for sentinel, this value is tuned accordingly, otherwise the service drops during deployments due to can't reconnect with provided scope) would be very helpful if it were added to the documentation in general.

Besides, I know this isn't the right place for this, but we can't specify a delay between retries, and the retries are happening too quickly. It would be great if we could add a delay as a configuration between retries (otherwise, because the retries are happening so fast, if the master drops in Sentinel mode, the service will drop again before a new master is selected).

Reference: https://github.com/open-webui/open-webui/blob/main/backend/open_webui/utils/redis.py L#48

Originally created by @ctolon on GitHub (Jan 28, 2026). Original GitHub issue: https://github.com/open-webui/docs/issues/1039 This environment variable not documented: https://docs.openwebui.com/getting-started/env-configuration/ In Sentinel deployments, a very important variable for HA (in terms of allowing time for automatic failover; if down-after-milliseconds is high for sentinel, this value is tuned accordingly, otherwise the service drops during deployments due to can't reconnect with provided scope) would be very helpful if it were added to the documentation in general. Besides, I know this isn't the right place for this, but we can't specify a delay between retries, and the retries are happening too quickly. It would be great if we could add a delay as a configuration between retries (otherwise, because the retries are happening so fast, if the master drops in Sentinel mode, the service will drop again before a new master is selected). Reference: https://github.com/open-webui/open-webui/blob/main/backend/open_webui/utils/redis.py L#48 <img width="952" height="925" alt="Image" src="https://github.com/user-attachments/assets/120ce3b9-5f92-4adb-b043-f3efd0cbd762" />

bsakaguchi closed this issue

2026-10-18 14:09:21 +00:00

bsakaguchi commented

2026-10-18 14:09:21 +00:00

Author

Owner

@Classic298 commented on GitHub (Jan 29, 2026):

fixed in dev for now - will be merged to main and live on docs soon

THANKS!

@Classic298 commented on GitHub (Jan 29, 2026): fixed in dev for now - will be merged to main and live on docs soon THANKS!

bsakaguchi referenced this issue

2026-10-18 14:10:45 +00:00

[PR #144] [MERGED] Doc for custom OpenID claims #278