Admin Panel V2 stuck on blank screen / infinite loading after successful login (HTTP/2 PING Failed, API invalid request) #5686
AlexusTdV
started this conversation in
Bug باگ: please use issue
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Describe the bug
After a clean installation of Hiddify Manager on Ubuntu 22.04, the Admin Panel V2 does not load after entering the Admin ID. The login form accepts the credentials and redirects to /admin/v2/, but the UI remains completely blank (white/black screen). The issue reproduces in Chrome and Firefox, in both normal and incognito mode.
Environment:
Server: DLine Media (KVM), 4 GB RAM, 2 vCPU, 25 GB Disk.
OS: Ubuntu 22.04.
Hiddify: Latest stable version (installed via hiddify upgrade release).
Access: Admin ID from the link, password left blank (never set).
What I have already tried (without success):
Full Reset: rm -rf /opt/hiddify-manager/hiddifypanel/hiddifypanel.db followed by hiddify install (Reinstall). Done multiple times.
Server Reboot: reboot after reinstall, waited 3+ minutes.
Browser change: Chrome and Firefox (normal and incognito).
Disabling HTTP/2: Forcefully removed alpn h2 from /opt/hiddify-manager/generated/haproxy.cfg, leaving only http/1.1. Reloaded hiddify-haproxy.
Port check: haproxy listens on 80/443, granian (panel) listens on 127.0.0.1:9000.
Symptoms and logs:
Browser Console Error (Chrome):
Failed to load resource: net::ERR_HTTP2_PING_FAILED
(Even after disabling h2 in the haproxy config, the browser still throws this).
Panel Logs (journalctl -u hiddify-panel --no-pager -n 50):
No Python tracebacks appear in the journal. The logs only show systemd lifecycle messages: Scheduled restart job, Main process exited, code=killed, status=9/KILL, and Failed with result 'timeout'.
When running python -m hiddifypanel manually, it just outputs the Flask CLI help menu, meaning the Python environment works, but the granian process fails to stay alive under systemd.
Local API check:
curl -s http://127.0.0.1:9000/api/v2/admin/me -H "Host: 194.87.87.211"
Response: {"detail":{},"message":"invalid request","msg":"invalid request"}
(Expected due to missing proxy_path and UUID, but it confirms the API backend is responding).
Service Status: All hiddify-* services are active (running). granian is listening on 127.0.0.1:9000.
Question for developers:
Why does the Admin V2 bundle fail to load after a successful redirect from the login form? Is there a known compatibility issue with granian and the current frontend build, or is haproxy misrouting WebSocket/HTTP2 requests after the redirect to /admin/v2/, causing PING_FAILED and a timeout?
All reactions