Errors in log and net responding clients

[dima]

Mar 13 14:12:15 [2025-03-13 13:12:15.157][response][INFO] (websockets_hub) GET /notifications/hub?<data..> => 200 OK
Mar 13 14:12:17 => Healtheck error: Error: Timeout of 7000ms exceeded
Mar 13 14:12:27 => Healtheck error: Error: Timeout of 7000ms exceeded
Mar 13 14:12:37 => Healtheck error: Error: Timeout of 7000ms exceeded
Mar 13 14:12:47 => Healtheck error: Error: Timeout of 7000ms exceeded
Mar 13 14:12:57 => Healtheck error: Error: Timeout of 7000ms exceeded
Mar 13 14:13:07 => Healtheck error: Error: Timeout of 7000ms exceeded
Mar 13 14:13:14 [2025-03-13 13:13:14.017][vaultwarden::api::notifications][INFO] Closing WS connection from

and vaultwarden.log:
[2025-03-13 13:12:15.157][response][INFO] (websockets_hub) GET /notifications/hub?<data..> => 200 OK
[2025-03-13 13:13:14.017][vaultwarden::api::notifications][INFO] Closing WS connection from

there is a gap ..

[dima]

can you increase the timeout?

[Ixapack]

@nebulon Hello same as dima . i don't have any issue and suddently timeout

[joseph]

Can you install a fresh Vaultwarden and check if that timesout ? I am wondering if you guys enabled some feature in Vaultwarden that is now making the healthchecks fail. It's not a timeout issue afaict (7 secs is plenty enough for a simple healtchech call)

[dima]

you mention a 2nd instance without data?

[joseph]

@dima yes, correct, just a test instance without data to see if that works. Atleast, this will rule out networking/database etc

Do you happen to know/remember if you changed any setting in Vaultwarden config settings? BTW, the http port (if this is somwhere in the ui page) has to be 8000.

[Ixapack]

Hello

here below my config .
I enable yubico and the 2fa and have been powned
also the token on argo2

I try right now a new instance

{
"domain": "https://vault.ixapack.com",
"sends_allowed": true,
"hibp_api_key": "",
"incomplete_2fa_time_limit": 3,
"disable_icon_download": false,
"signups_allowed": true,
"signups_verify": false,
"signups_verify_resend_time": 3600,
"signups_verify_resend_limit": 6,
"signups_domains_whitelist": "ixapack.com",
"org_creation_users": "informatique@ixapack.com",
"invitations_allowed": true,
"emergency_access_allowed": true,
"email_change_allowed": false,
"password_iterations": 600000,
"password_hints_allowed": true,
"show_password_hint": false,
"admin_token": "$argon2id$",
"invitation_org_name": "Ixapass",
"ip_header": "X-Forwarded-For",
"icon_redirect_code": 302,
"icon_cache_ttl": 2592000,
"icon_cache_negttl": 259200,
"icon_download_timeout": 10,
"http_request_block_non_global_ips": true,
"disable_2fa_remember": false,
"authenticator_disable_time_drift": false,
"require_device_email": false,
"reload_templates": false,
"log_timestamp_format": "%Y-%m-%d %H:%M:%S.%3f",
"admin_session_lifetime": 20,
"increase_note_size_limit": false,
"_enable_yubico": true,
"yubico_client_id": "",
"yubico_secret_key": "",
"_enable_duo": false,
"_enable_smtp": true,
"use_sendmail": false,
"smtp_host": "mail",
"smtp_security": "off",
"smtp_port": 2525,
"smtp_from": "vault.app@ixapack.io",
"smtp_from_name": "noreply",
"smtp_username": "vault.app@ixapack.io",
"smtp_password": "",
"smtp_auth_mechanism": "Plain",
"smtp_timeout": 15,
"smtp_embed_images": true,
"smtp_accept_invalid_certs": true,
"smtp_accept_invalid_hostnames": true,
"_enable_email_2fa": true,
"email_token_size": 6,
"email_expiration_time": 600,
"email_attempts_limit": 3,
"email_2fa_enforce_on_verified_invite": true,
"email_2fa_auto_fallback": true
}

[Ixapack]

New instance i don't have any down
I try on my old one to disable one by one some function still the time out .
odd

[dima]

no errors in the test instance - healthcheck every 10 sec

[dima]

Any news?

Also without significant load, it isnt fine

10:00 healthmonitor App is down
23:31 healthmonitor App is back online
23:28 healthmonitor App is down
23:08 healthmonitor App is back online
21:48 healthmonitor App is down
20:06 healthmonitor App is back online
20:05 healthmonitor App is down

[joseph]

@dima @Ixapack please contact us on support@cloudron.io . It's unclear if the issue is with the app or the package or the platform

[girish]

@dima @Ixapack the issue seems to be that apache is getting busy because of many websocket connections. since the number of apache instances gets saturated, the healthcheck call stops working (since there is free apache process to respond).

I checked upstream . Looks like they have added a /alive route and we don't need the internal apache anymore . https://git.cloudron.io/packages/vaultwarden-app/-/merge_requests/14 . Think that should sort out the issue.

[Ixapack]

@girish Hello i confirm, the update fix the issue
thank you