DNS down - set your DNS servers manually

[#1]

It's not THAT unreasonable to confirm after a major outage that service is restored. Almost everyone has a phone with mobile data these days. I watched a programme on TV the other day where people living in the slums of India had iPhone15's. 

Does PlusNet plan to publish an incident report on the DNS failure and if not, why not?

[#2]

I did wonder what was going on yesterday afternoon as some of my devices worked but others were very limited.

Then I realised it was those that I'd reconfigured to use Google DNS that were OK. 🤣

I have a Hub one on which you can't override the DNS settings (or tell me if you know how!).

On the other hand you can give the different frequency bands different names, unlike the weird limitation on Hub2.

[#3]

---

@Townman wrote:

---

@seabasskarp wrote:

As a customer it should be in my inbox, along with an explanation. It is not. 

---

This is an unreasonable expectation.

In the event of an incident impacting one's ability to access the internet ... what is the point of a communication sent to one's mailbox? 

With A&A, they email customers to indicate their connection is down (including exact time and cause), then another email saying when it recovers and the duration of being offline.

[#4]

Different ISPs and communications providers have different approaches to communicating issues.

For the ones I use I get e-mails from some, and SMS from others, and there are status web pages I can look at. With my current ISP I'm kept informed of issues, future maintenance, and other things that may affect my connection (although they seldom do).

In my experience Plusnet are unique in being totally quiet about issues.

[#5]

Many years ago Plusnet used to have a page where you could enter your phone number to see whether there were any issues in your area - similar to the Zen Broadband Status page.

The Plusnet page used to work well, then for a couple of years the updates became less and less frequent, until eventually the page was so out of date and useless that they removed it completely,  rather than automate the status updates to work properly as clearly Zen are able to do.

[#6]

With A&A, they email customers to indicate their connection is down (including exact time and cause), then another email saying when it recovers and the duration of being offline.

---

If you're happy with the expensive service you can expect that kind of hand holding. At the cheap end of the market we have to put up with what we can get.

[#7]

What is the actual cost of sending an email to all affected customers ?

[#8]

---

@Rowdy wrote:

I totally agree with sleepygaz. Would plusnet please explain what caused the problem in detail ? If I do not get a reply I will move to a more reliable 'which recommended' supplier when upgrading to Full Fibre. That would be a shame as for 20 years as a plusnet customer I have always received good service and only a couple of outages. 

---

Get a grip!!

By your own words - 20 years and only a couple of outages.

Consider

• 20 years - that is 175,200 hours
• Say 10 outages - a good few more than you cite (to make things worse) as grounds for moving away
• Say 3 hours average per outage
• that's 30 hours down time in 20 years

In other words you have had 99.85% reliability!!!  You can do the calculations for a different number of incidents and average duration.  You would have needed to have experienced more than 10 times the above outage (350 hours) to reduce the reliability to 99.80%.  Yes 320 more hours nonavailability to reduce reliability by 0.05%

How much more reliability do you think is attainable? Some reality is required here!

You might also care to look at this - Choosing the best mobile, broadband and landline provider - Ofcom - Plusnet appear to have the highest satisfaction rating for broadband.

For the avoidance of doubt, the above does not consider the unreliability brought about by BT Openreach infrastructure as that is the same for all ISPs using their equipment.  Thus the consideration of moving to a "more reliable supplier" should rightly ignore that contribution to service disruption (lack of 'reliability').

I would not disagree that that there is room for improvement in making known the fact that there is an incident impacting a significant number of people - invariably that will be a matter of deciding where to put effort (fixing or communicating) against the expected restoration time.

[#9]

---

@Anonymous wrote:

What is the actual cost of sending an email to all affected customers ?

---

No significant cost - but more than updating a status page.

And much cheaper than answering lots of phone calls.

[#10]

---

@Anonymous wrote:

What is the actual cost of sending an email to all affected customers ?

---

It is not the cost of sending an email, it is the cost of implementing the means to do so and then the systems load of doing it.  If the resolution time is expected to be short, for the MAJORITY of users, the "it broke, it is now fixed" messages will be of no material benefit.

[#11]

---

@Townman wrote:

... invariably that will be a matter of deciding where to put effort (fixing or communicating) against the expected restoration time.

---

I don't think that Plusnet have much involvement of fixing issues - and the ones they are involved in fixing (DNS?) it will be quite different people doing the fixing than those that could be doing the communicating.

BTW when I worked in telecoms development we had a contractual 99.999% (known as five-nines) availability requirement. That's less than 5 hours in total over the 50 years expected lifetime of the equipment. Edit: I should add that rebooting or upgrading the system was included in the downtime - so we had to be able to do upgrades without rebooting the exchange.

[#12]

---

In other words you have had 99.85% reliability!!!  You can do the calculations for a different number of incidents and average duration.  You would have needed to have experienced more than 10 times the above outage (350 hours) to reduce the reliability to 99.80%.  Yes 320 more hours nonavailability to reduce reliability by 0.05%

That's nowhere near the five nines reliabitity of the POTS system years ago.

[#13]

For me this is not just about downtime. I would like to know the cause of the downtime that has caused a total DNS outage across supposedly redundant servers. Not just for my own interest but to understand if there are any potential security concerns. 

Since PlusNet like to keep quiet how do we know they are not covering up a malicious attack or similar?

In an attempt to better make my point I have included a ChatGPT story that better clarifies my position:

In the heart of the digital jungle, an adventurous iguana named Iggy explored the supposedly impenetrable fortress of Plusnet's DNS servers. With a keen eye, Iggy noticed a small crack in the digital armor that guarded the servers. Intrigued, he slipped through, only to discover outdated security measures.

Playfully, Iggy left a message for the online realm, "Even the mightiest fortress has a hidden doorway." The news spread, and the online community chuckled at the irony, revealing the servers' vulnerabilities.

The moral of the story? In the vast jungle of the Internet, communication is key. Just like the iguana exposed the crack in the fortress, effective communication can unveil hidden weaknesses and strengthen the digital defenses we rely on. So, let the iguana's adventure remind us to communicate openly and address vulnerabilities for a safer online ecosystem.

[#14]

---

@corringham wrote:

---

@Anonymous wrote:

What is the actual cost of sending an email to all affected customers ?

---

No significant cost - but more than updating a status page.

And much cheaper than answering lots of phone calls.

---

Having a sensible message available on the phone makes more sense than a webpage - how were we supposed to acces that?

[#15]

---

@Townman wrote:

It is not the cost of sending an email, it is the cost of implementing the means to do so and then the systems load of doing it.

---

The cost of implementing ?

They had a perfectly good system that reported Service Status AND emailed customers - until they switched it off mid 2018 !

Here is a screenshot of one of those emails -