auxren

Version: 7.2.0 (kernel 6.12.54-Unraid) Array state: STARTED, 25 disks, no invalid/missing/disabled per mdcmd status emhttpd: not running Symptoms -------- - WebGUI stale for weeks (disks.ini mtime frozen at 2026-04-01) - POST /update.htm actions fail: nginx error.log shows connect() to unix:/var/run/emhttpd.socket failed (111: Connection refused) - Parity-cancel from Main doesn't register - Array, shares, dockers still function via shfs; only the GUI / emhttpd is down What I've verified over SSH --------------------------- 1. pidof emhttpd is empty. Manual restarts /etc/rc.d/rc.emhttpd restart) log: emhttpd: unclean shutdown detected emhttpd: shcmd (6): /usr/local/emhttp/webGui/scripts/update_access emhttpd: shcmd (7): modprobe zfs emhttpd: Lifetime key detected, GUID: ... emhttpd: get_diskFsSizes 8588 diskFsType.0 not found …and the process exits. 2. /boot/config/disk.cfg has no slot-0 keys at all — no diskId.0, diskSlot.0, diskFsType.0, nothing. Slots 1–23 are fully populated. 3. /var/local/emhttp/disks.ini still has the parity entry correctly (idx=0, name=parity, device=sdk, id=HUH721212ALE601_8HGYTV7Y, status DISK_OK, size 11718885324). So the md driver knows about parity; only the persisted config has lost slot 0. 4. /boot/config/forcesync has been present since 2026-02-28 14:34. Parity history /boot/config/parity-checks.log): 2026 Feb 28 3600s sbSyncExit=-4 (cancelled) 2026 Mar 21 764317s 15,700,473 errors sbSyncExit=0 i.e. parity has been unhealthy for months. 5. Earlier in this sequence, one of my SSH sessions caught this in syslog: put_config ... rename: /boot/config/ident.cfg.new /boot/config/ident.cfg → No such file or directory It's transient now (/boot is rw, touches succeed), but it matches the pattern of a botched put_config write that could have truncated disk.cfg too. What I've done -------------- - Backed up disk.cfg → /boot/config/disk.cfg.backup-20260418-001039 - No config edits, no mdcmd stop, no reboot. - Generated diagnostics (attached). Questions --------- 1. What's the correct slot-0 stanza for a parity disk in disk.cfg on 7.2.0? I can reconstruct from disks.ini but want the exact expected keys before writing. 2. Does rename: /boot/config/ident.cfg.new ... indicate flash corruption worth addressing before the next reboot? 3. Safe to clear /boot/config/forcesync before the first good emhttpd start, or leave it so the next boot runs a full parity check? Holding off on reboots and mdcmd stop until I have guidance. Happy to run specific commands and report back. levyserver-diagnostics-20260418-0010.zip

Amazing! Thank you so so much. That did it.

I updated to 7.0 yesterday and everything seems smooth until I tried to run a script from the terminal. I've run the script many times over the past few months with no issues. Now I get certificate issues that I can't seem to figure out. Below is the error. Anything I should look into? WARNING: Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'SSLError("Can't connect to HTTPS URL because the SSL module is not available.")': /simple/