Authored by: By Kieran Fox, Reveille
I spend most of my time opening the hood on Reveille for OnBase teams, and the most useful part of any session is never my slides — it’s the questions. People bring the exact problems they’re fighting in production, the ones that don’t show up as “the platform is down” but quietly cost a morning, a backlog, or a 7 a.m. phone call.
Here are six that come up again and again, and what Reveille actually does about each. None of them are dramatic outages. That’s the point — they’re the kind of thing a basic up/down check sails right past.
1. Unity Scheduler failures after maintenance
This is the classic. You take a maintenance window, everything comes back, every service shows green — but the although it looks like the Unity Scheduler is running It’s not processing work, scheduled jobs and workflow timers silently stop firing. Nobody notices until a process is a day or two behind.
Reveille catches this two ways. Its checks to confirm the OnBase services and scheduler genuinely came back up after the maintenance window, not just that the server is reachable. And its workflow-timer tests specifically watch for a timer that errored or one that simply never ran when it should have. If a self-healing action is configured, Reveille can restart the stalled service on its own. Either way, you know within minutes of the window closing — not when the work is already behind.
2. OnBase connectivity and “Invalid Data Source” errors
When the application server loses its line to the database — a datasource misconfiguration, a credential change, a SQL service that didn’t restart — users hit errors or can’t log in, and the first report is usually just “OnBase is broken.”
Reveille turns that into a located fault. Its connectivity tests run independently against the OnBase Application Server, Database Server, File Server, and line-of-business server, so you can see exactly which link dropped. Its SQL Server service and database checks confirm whether the problem is on the database side. And its synthetic Web Client and Unity Client logins reproduce the actual user experience, so instead of everyone guessing, you get “the app server lost its data source to SQL” — the real fault, already isolated.
3. Disk groups hitting 100% capacity
A disk group fills up, and OnBase can’t store new documents — imports and archival stall. By the time someone notices, capture is backing up behind it.
Reveille watches disk group fill levels and server free space continuously, and because it can set dynamic thresholds from a machine-learning baseline, it warns you when a disk group is trending toward full — days of runway, not a hard wall you hit at 2 a.m. The same trend data feeds capacity planning, so you provision ahead of the problem instead of reacting to it.
4. Database performance monitoring
OnBase “feels slow,” but the platform is up — and nine times out of ten the answer is in the SQL Server underneath.
Reveille tracks the database signals that actually explain OnBase slowness: deadlocks per second, memory pressure, batch requests, transaction-log backup health, and database growth. Dynamic thresholds mean a busy Monday morning doesn’t page anyone, but a genuine regression does. The practical payoff is that when OnBase is dragging, you walk into the conversation with the database evidence already in hand — instead of a bridge call where the OnBase team and the DBA point at each other.
5. Concurrent license cap and blocked logins
You hit your concurrent license ceiling, users start getting turned away at login, and you find out from the help desk — usually at the worst possible time of the month.
Reveille tracks OnBase license usage in real time and alerts as you approach the cap, before anyone is locked out. The same data gives you utilization trends for renewal timing and right-sizing, so the cap stops being a surprise. And if a blocked-login condition does occur, the synthetic login tests catch it immediately rather than waiting for tickets to pile up.
6. Full-text indexing queue backups
The full-text index queue backs up, newly imported documents don’t get indexed, and users start saying “I can’t find the document I just filed.” Search quality degrades quietly, with nothing technically “down.”
Reveille monitors queue depth and aging — documents waiting to be processed, items sitting in a queue past a set number of days — and flags an abnormal backlog early. That means you catch the indexing service falling behind while it’s still a small problem, before it shows up as a user complaint.
The thread running through all six
Notice what these have in common: not one is a platform-down event. They’re degradations, silent stalls, and capacity walls — exactly the failures a simple availability check misses, and exactly where an experienced admin’s instincts go when something feels off. Content Observability is really just that instinct, made continuous and automatic: watching the OnBase environment and everything around it — services, SQL, disk, licensing, queues, and the integrations between them — all the time, instead of whenever someone thinks to look.
Reveille does it agentless, with more than 80 OnBase-specific tests and 85 dashboard metrics out of the box, and it feeds whatever you already run — Splunk, ServiceNow, PagerDuty, Teams — rather than becoming one more console to live in. A large share of these issues it can resolve on its own through self-healing; the rest it hands you already diagnosed.
If you’d like to see Reveille open the hood on your own OnBase environment — your disk groups, your license usage, your scheduler — the i3 team can set that up.