Backup monitoring for MSPs
In development
Failures send email. Dead jobs send nothing. BackupWatch knows each job's schedule and alerts you when the email never arrives.
Failures send email. Dead jobs send nothing. BackupWatch knows each job's schedule and alerts you when the email never arrives.
The one failure email scrolls past with the rest, and nobody notices until the client calls.
A job that stops running sends no email at all. Nothing looks wrong until a client needs a restore.
Inbox rules catch the first one. Nothing in your inbox catches the second, because a rule cannot fire on silence.
One process on one server, polling the mailbox your backup notifications already reach. No agents on client machines.
Reads unread mail from one Microsoft 365 mailbox every five minutes through the Graph API.
Pulls vendor and outcome out of Veeam, DropSuite and Synology notifications: success, warning, failure.
Routes every email to the right client and job by sender, subject, or a rule it learned from you.
Every job carries a schedule and a grace window. No success email by the deadline is a missed backup. This is the part an inbox rule cannot do.
Failures and misses email your team once per missed window, never on a loop, plus a daily digest.
A new client's email lands in the Review queue. Assign it once and BackupWatch derives the routing rule, from the sender domain or a distinctive subject token, so the rest route themselves.
It suggests the job it believes is right, so confirming takes one click. Learned rules stay visible and editable.
Every client page carries a 30-day history of backup outcomes, one column per day and one band per job. Click a day to read the emails behind it.
Export it as a client-facing report or CSV, so "are our backups actually running?" has a one-page answer at the quarterly review.
NOC mode turns a readonly account into a TV kiosk: a 30-day session, a refresh every 60 seconds, and no way to change a password from the screen.
That is the real competitor and it is free. Here is exactly where it gives out.
Nothing happens. A rule cannot fire on an email that was never sent.
A missed alert the moment the schedule and grace window pass.
Flagged, if somebody opens the folder.
Alerted, tracked, and acknowledged by a named person.
A spreadsheet, if anyone still maintains it.
A 30-day chart and an exportable report, automatically.
Build folder rules by hand, per sender format.
Assign one email and the engine writes the rule.
One rule set for every sender format you meet.
Veeam, DropSuite and Synology parsed, the rest queued for review.
A free tier is planned from day one. No key, no signup, no time limit: install it and run it.
One flat annual licence for unlimited jobs, priced before launch. Early access members get the number first, and a say in it.
Request early accessA fresh Ubuntu box plus the release ZIP is two commands.
Point it at your notification mailbox next. The Microsoft 365 guide covers the Azure app registration in about ten minutes, and updates run from inside the app after that.
Ubuntu 22.04, one process, SQLite, no build step
BackupWatch is close, not shipped. Tell us what you run and we will send you the build when your vendor mix is covered, plus the pricing before it goes public.
Sample notification emails are the fastest way onto the list. Every one you send becomes a parser.
Success, warning and failure out of the box. Synology means Active Backup for Google Workspace specifically. Anything else, you can define as a vendor rule in the UI.
Unknown mail waits in Review until a human routes it. Nothing is ever dropped quietly.
One mailbox and one Azure app registration. Send us a sample email from another vendor and we will add the parser.
No. The only outbound calls are Microsoft Graph for your mailbox, your own SMTP relay for alerts, and a daily licence check that carries no client names and no email content.
You get a 7-day grace period, fully functional, with a banner explaining it. After that, five or fewer active jobs quietly becomes the free tier. Above five, polling pauses until you renew, but failures already detected still alert and the daily digest keeps sending.
One small Ubuntu 22.04 box. A single Python process and a SQLite file, so there is no database server, no Docker requirement and no build pipeline. The installer handles systemd, nginx and HTTPS.
One Microsoft 365 mailbox through the Graph API, using an Azure app registration with application permissions. The documentation walks through the whole setup in about ten minutes.
Probably. You can define custom vendor rules in the UI with sender and subject identifiers plus status keywords, and anything unrecognised lands in Review rather than vanishing. Send us a sample notification and we will build the parser.
