The Common Wholesale Platform (CWP), which is a non-profit member-owned organisation that helps alternative broadband networks to connect with the retail ISP market, appears to be suffering some sort of disruption today after their main website domain went down (CWPUK.org) and stopped some consumer switching between ISPs.
At the time of writing it’s unclear what has caused the outage (some suggest it may be a change of web host), although CWP are one of the Managed Access Providers (MAPs) that enable a number of smaller ISPs to harness the industry-led One Touch Switching Company’s (TOTSCo) messaging hub; this forms part of the One Touch Switching (OTS) process for easier and quicker UK consumer switching between broadband and phone providers.
The domain outage is currently known to be disrupting the ability of some CWP partner ISPs to switch consumers between providers. Quite why CWP’s OTS integration / portal is dependent upon a single publicly exposed domain being functional is also unclear, but we’d imagine they may want to find a solution to that in the future.
Advertisement
UPDATE 3:35pm
We’ve had it confirmed that the outage affecting the CWP platform and its support systems was indeed due to a domain name change to a new host. The change was unexpectedly triggered earlier today instead of during the planned maintenance window. The platform is expected to be back online again shortly.
Advertisement
We appologise profusely about this (first ever) outage with our platform to our members and the industry. As suggested this is caused by a domain name transfer to a new name server affecting switching platform and support systems. The transfer should have been planned out of hours as a planned outage. Unfortunately our backup domain also failed. We will review our process once we are back up. TOTSCo and customer have been informed. I will post a further update once resolved.
There is a whole set of CWP domains, such as cwpuk.co.uk, that appear to be available for registration, which seems extraordinary.
Why extraordinary? Who is going to build their API calls to a domain not listed in the technical manual?
@125us:
It facilitates attack vectors that can be surprisingly effective.
Nameservers appear to have moved GoDaddy/123-reg to Hostinger, certainly a choice for such infrastructure.
Since the change was unexpected, I wonder if they’re utilising some kind of managed migration service, it’s the only way I can imagine nameserver changes would be made without having the zone transferred and/or replacement infrastructure in place – currently they have no DNS records, not even email.
Used to work at a web host as a senior engineer, one of my responsibilities was migrations. Also picking up migrations after another engineer had cocked it up or there were unknown complications.
Wouldn’t surprise me if someone switched the nameservers without setting up the zone. Or if there was just a misunderstanding and an IT vendor responsible for the domain switched the nameservers too early etc.
Seen it happen a bunch of times.
The issue is now resolved by backing out of the DNS change and we will more carefully plan this change in the near future. Thank you all for the above feedback. A number of schoolboy errors were made today in this first platform outage since our launch in Dec-2022. All members and other ISPs via TOTSCo have been updated.