Alerts
Storacle warns you before storage growth turns into an overrun, and tells you when its own engine needs attention. Here is what it alerts on, how an alert lives, and who hears about it.
What Storacle alerts on
Every org gets the alerts that say whether Storacle is working and whether its seats line up. The alerts about your storage, the ones that warn you before a limit arrives, come with a seat or the trial. See Pricing for what a seat costs.
Free in every org
Engine run failure
A nightly run failed outright, so that day may have no measurement. Every missed day is a day of history that can never be collected.
Engine run issue
A run finished but recorded a serious problem along the way. One run produces at most one alert of each kind, however many problems it logged.
Stale record-count statistics
The record counts Storacle reads have stopped refreshing, so the numbers collected can't be trusted. A Salesforce support case can get them moving again; it isn't something a Storacle purchase fixes.
Seat shortfall
More people hold a seat than the org is entitled to, which closes the paid features for everyone until it is corrected. The person who has to fix it hears about it, seat or not.
Upcoming seat shortfall
The coming renewal entitles the org to fewer seats than people already hold. This warns ahead of the date, before anyone loses access.
With a seat
Storage utilization, warning and critical
Data storage or file storage has passed a share of its limit. The two are judged separately, so each gets its own alert, and an org past the critical level carries one critical alert rather than a warning as well.
Growth rate
An object is gaining records far faster than its own history says is normal: the runaway integration, the duplicate load, or the misconfigured batch job. An object too young to have a history is judged against a curve suited to its size.
Forecast limit
The forecast projects data or file storage reaching its limit while there is still time to act cheaply. Only a projection Storacle trusts raises it; a low-confidence forecast raises nothing.
Renewal readiness
Ahead of your Salesforce contract renewal, the forecast says to buy storage at that renewal or sooner. It needs the renewal date entered in Storacle Configuration.
The utilization alerts raise at 75% of a limit for the warning and 90% for critical out of the box. Both levels are adjustable on the paid tier.
Where alerts show up
Every unresolved alert is listed on Storacle Home, most urgent first, and in the Storacle Alerts tab. New critical and high-urgency alerts also appear at the top of Home until someone acknowledges them.
Everyone who can open Storacle sees the free alerts. The storage alerts are shown only to people with a seat, or to everyone on the Storacle user list while the trial runs, so a colleague without a seat in a paying org is not shown what the seats bought.
How an alert lives
Alerts are built not to turn into daily noise.
One condition, one alert
A condition that stays true for a hundred days is one alert, not a hundred. Storacle keeps the open alert current as the numbers move.
Mailed when it first appears, and again only if it gets worse
A standing alert is not emailed every night. It is emailed again only when it crosses into a more serious band than the one it was last reported at.
A short spike does not become an alert
Some alerts wait for a condition to hold for several days before they raise, so a single busy day passes without one.
It clears itself
When the condition stops being true, the engine resolves the alert on its next run. The record is kept, so you can still see that it happened.
Acknowledge and resolve
Both actions are on the alert's record page. Alerts are raised by the engine, so they cannot be created or deleted by hand; what you record on one is its status and what you did about it.
Acknowledge
Says you have seen the alert and own it. It moves the alert off the top of Home and leaves it in the alert list and every open count, because acknowledging is triage, not closure. An acknowledged alert still clears itself when its condition does.
Resolve
Closes the alert with a required note saying what was done, and stamps when. If you resolve an alert while its condition is still true, Storacle takes that as a decision to stop being told and does not reopen it. If the condition later clears and then comes back, that is a new episode and raises a new alert.
An alert the engine resolved on its own has no note, which is itself the record that nobody had to step in. Resolved alerts are kept for a retention period and then cleaned up; open alerts are never removed.
Who gets the email
Email is the only channel alerts go out on. Each alert is sent to two groups combined, so adding a shared address never takes anyone off the mail.
People on the Storacle user list
Being on the list is what makes someone reachable. Every active person on it is emailed the free alerts with nothing for anyone to set up, and the storage alerts too if they have a seat or the trial is running. Someone not on the list is emailed nothing; a Storacle Administrator can add them.
Each person chooses for themselves, one alert type at a time. In Storacle Configuration, under Individual User Preferences, Alert Subscriptions has a switch for every type, and every switch starts on. Turning one off changes that person's inbox and nothing else: the alert is still raised, everyone else is still emailed, and it still shows in Storacle. Anyone on the list can do this for the alerts they receive, seat or not.
The org's notification addresses
A list of addresses that receives every alert, for routing alerts somewhere that is not a person: a shared inbox, a ticketing system, or a distribution list. It starts empty. It is set in the Notifications section of Storacle Configuration's org-wide settings, which a seat holder can change, alongside one org-wide switch for alert email. Turning that switch off stops the mail and nothing else; alerts are still raised and still shown in Storacle.
On a day that raises a great many alerts, Storacle emails the most urgent ones individually and lists the rest in one summary to each audience, so the org's shared daily email allowance survives. Every alert is still in Storacle in full.
If alerting itself fails during a run, Storacle sends a plain message saying so to the org's notification addresses, or to the person the engine runs as when there are none. That message goes out even with alert email switched off.
Tuning alerts
Every alert type has a definition in the Storacle Alert Definitions tab. An administrator can adjust a definition's threshold, how long a condition must hold before it raises, and its urgency, or switch the alert off for the whole org by deactivating it.
How unusual growth has to be, how long record-count statistics may go without refreshing before they raise an alert, and the notification settings are org-wide settings that a seat holder can change. Anyone can read them, and values set while paying are kept if the org stops. The How Storacle Works page covers settings in full.
Still stuck?
If an alert is not reaching the people this page says it should, contact support.