Session log: OwnCloud cache migration completed successfully

589G OwnCloud data moved from cache SSD to disk7 array (2h49m rsync).
Cache dropped from 82% to 34%. MariaDB + Discourse recovered and running
7h+ healthy. Share config changed to no-cache permanently.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

session-logs/2026-04-16-session.md

@@ -1163,3 +1163,55 @@ Ran full audit of Jupiter (172.16.3.20, Unraid 7.1.4):
 4. **UPS monitoring** — install apcupsd if UPS connected
 5. **Verify parity** — couldn't read parity status via CLI, check Unraid UI
 6. **All prior pending items from earlier this session** still open (Howard Gitea account, Len's deployment, etc.)
+
+---
+
+## Update: overnight — OwnCloud migration completed (verified 2026-04-17 morning)
+
+### Migration result
+
+rsync completed at 22:59 MST (2h 49m total for ~750G uncompressed data). Script executed all cleanup steps automatically:
+
+| Timestamp | Action |
+|---|---|
+| 20:10:22 | rsync started |
+| 22:59:21 | rsync exit code: 0 |
+| 22:59:21 | Removing cache copy |
+| 23:00:43 | Cache copy removed |
+| 23:00:43 | Share config updated (no-cache) |
+| 23:00:48 | OwnCloud VM started |
+| 23:02:04 | MIGRATION COMPLETE |
+
+### Post-migration state (verified 7h later)
+
+| Metric | Before | After |
+|---|---|---|
+| Cache usage | 756G / 932G (82%) | **311G / 932G (34%)** |
+| Cache free | 175G | **614G** |
+| MariaDB-Official | Crash-looping (143 restarts) | **Up 7h, healthy** |
+| Discourse (app) | Crash-looping (100+ restarts) | **Up 7h, healthy** |
+| OwnCloud VM | Shut off | **Running** |
+| OwnCloud data location | Cache SSD (589G compressed) | **disk7 array (~750G uncompressed)** |
+| OwnCloud share config | `shareUseCache="yes"` | `shareUseCache="no"` |
+
+### Key finding: BTRFS compression ratio
+
+OwnCloud data was 589G on BTRFS cache (with transparent compression) but ~750G+ uncompressed on XFS array. ~22% compression ratio on typical office files (PDFs, OneNote, images, videos).
+
+### Root cause chain (complete)
+
+1. OwnCloud data (589G) filled BTRFS cache data chunks to 99%
+2. Unraid's `shareFloor=195G` on appdata share rejected writes when free dropped below 195G
+3. MariaDB-Official couldn't write `ddl_recovery.log` -> crash loop
+4. Discourse depended on MariaDB -> cascade crash loop
+5. Mover couldn't fix it because OwnCloud VM held SMB locks on all files -> mover skipped silently
+6. Fix: stop VM, rsync to array, delete cache copy, change share to no-cache, restart
+
+### Credentials (referenced)
+
+- Jupiter: `root` / `Th1nk3r^99##` (vault: `infrastructure/jupiter-unraid-primary.sops.yaml`)
+
+### Migration script location
+
+- `/tmp/migrate-owncloud.sh` on Jupiter (ephemeral, lost on reboot)
+- `/tmp/owncloud-migration.log` on Jupiter (full rsync log, ~789 KB)