What this page is
status.webryk.ca publishes the observed state of the services Webryk operates: the marketing site, the client portal, sign-in and second factor, the partner area, the document vault, payments, email delivery, scheduled jobs and the public demo. Alongside them are the third-party vendors those services depend on.
It exists for one reason: when something is not working, you need to be able to check it somewhere other than in an email that may never be sent.
The measurements are automated. The details of the method, the thresholds, the cadences and the known limits are in the methodology. This document does not repeat what is there.
Who operates it
Webryk Studio, Sherbrooke, Quebec.
There is no intermediary: the people who operate the monitored services are the same people who operate this page.
The conflict of interest, named
Webryk measures Webryk. That is a real conflict of interest and it does not go away because it is awkward. Better to write it here than to let you find it.
What constrains it, concretely:
- Published states come from automated probes. Nobody types a state in by hand.
- History is not rewritten. No period is filled in after the fact, no past value is silently corrected.
- Incident opening will be triggered by the measurement, not by an editorial decision. That automation is not yet in service: see the methodology.
- Percentages are truncated toward zero at the second decimal, never rounded.
- No capability Webryk operates is removed from the page because it is behaving badly. The inventory is fixed.
- Corrections are appended and dated, and the original value stays visible.
This is not an independent audit and the page does not claim to be one. It is an automated measurement whose rules are published in advance, which is checkable from the outside: if the displayed behaviour departs from the written rules, it shows.
What is published and what is not
Rows are named as capabilities, as a visitor understands them. No API path, no internal route name, no scheduled job identifier, no database latency, no raw error message.
This is not coyness. Those details help nobody work out whether their work is blocked, and they usefully point someone looking for where to push.
Webryk also monitors client-owned systems. They are not here: their state belongs to their owner. Everything Webryk operates on its own account, on the other hand, is on this page, on good days and bad ones.
Infrastructure
This page is built on Cloudflare and uses none of the application components it monitors: no Vercel, no Supabase, no Resend. A total outage of any of those three leaves this page online and current.
The two remaining dependencies, a global Cloudflare outage and the DNS zone shared with webryk.ca, are described in the methodology.
Languages
French is the default language. The English version lives under /en.
Both versions say the same thing. A sentence is not published in one language until it exists in the other, including for updates to an ongoing incident. A half-translated status page during an outage is a half-usable status page.
Time zone and format
Timestamps are shown in UTC. An incident’s timestamps add the Montreal-time equivalent in parentheses. Durations are given in minutes, then in hours and days beyond that. The day that divides the 90-day bar begins at midnight, Eastern Time (America/Toronto). The uptime window is a rolling 90 days at minute granularity.
Getting notified
An Atom feed publishes incidents and planned maintenance: /en/feed.xml.
There is no email subscription. That would mean depending on a sending provider that is itself one of the monitored services, and so sending the outage notice through the channel most likely to be down at the same moment. The feed is served as a static file by the same infrastructure as this page.
Incidents and post-incident analyses
An incident will open automatically as soon as an outage is confirmed, with a pre-written sentence saying only what has been observed. Everything after that is written by a person. Until that automation is in service, every incident published is written in full by a person.
Post-incident analyses are always written by a person and published separately from the incident. They are not guaranteed for every interruption: they are written when there is something to say beyond “it went down, it came back”.
Contact and corrections
To report a discrepancy between what you observe and what is displayed: contact@webryk.ca.
Worth including: the timestamp in UTC, the capability concerned, what you observed and the network you were on. If our record is wrong, the correction is published on this page, not only in a private reply.
This address is not a support channel. Support requests go through the usual client portal routes.
Monitoring active since August 3, 2026.