New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
People really like to complain very often, I'm not talking about specific cases, we have our problems too, but fortunately it's only a noisy minority.
Those with good experiences are unlikely to come here to write. But looking at the renewals of our offers, I can say that most people are with us.
I appreciate your thoughts, thanks for sharing them.
There are definitely things we can improve on, and we will.
@DeluxHost status on #UJN-217201 please?
I think you missed the point where the operation could be stabilized a while ago.
The DDoSers intends is to stop you from making sales. But, isn't it a great time to stop and start making automation tools to identify problem even before the customer noticed it? In the right hand, AI could ship working script within days, a week. For example, WHMCS script that extends service date? A host here made it in less than 36h.
Checking it in 1 hour
The DDoS attack issues resolved a month ago, so we fixed that and stabilized everything. Future automation is certainly possible and we'll work on it, but right now there are some critical issues that need to be resolved.
I think some people here simply have far too high expectations for a service they pay pennies a month for. I've never been one to say "you get what you pay for," but it's truly absurd how all the comments and major problems come from these people when business customers who pay double or triple the monthly amount don't make all this fuss and are more willing to communicate.
Offering the highest level of support service that many want costs money. We adapt and can certainly improve, and we will hire more people. But right now, I think it's simply unrealistic to attack a team of three people who serve thousands of customers, where, I can safely say, most of them are happy with what they have.
But, there's limit to patience. You're not trying to become @VirMach whose customer have to wait for 6-12 months until things gets fixed, right?
Serving thousands of customer with a team of three require heavy automation from monitoring, ticketing, and coordinating with remote hands. It's not optional
well this is worth quoting.
anyway, if you genuinely believe that people aren't paying enough to receive working services perhaps we'll see that disclaimer on future offer posts.
The DDoS problems when everyone thought we were dead seem to have been successfully solved, as well as many other problems, simply pointing the finger at the things not yet solved is easy, looking at the solved and optimized ones a little less sometimes, that's how it works, I'm not speaking against you and I understand what you and others say, but it takes still time to solve some issues.
support is definitely one of them.
We try to offer the best that we can at a rock-bottom price, but unfortunately people take everything for granted.
Our services, no matter how much they say, work and are not a scam.
I simply think that hosting critical or important services on 7€/y servers is wrong in my opinion.
If necessary, we will write it even more clearly in future posts.
We don't want to raise prices any further, despite all the unilateral increases and increases we've had, even at times making a loss on some things.
In fact, beyond the DDoS attacks, which are not included in the SLA and where we have provided compensation anyway, where have there been other problems?
the problems of a single customer or a group of customers, It is obviously not the problem of the entire infrastructure.
Saying that customers have "far too high expectations" when you have an explicit 99.9% SLA is bullshit. I don't care what stipulations you add
I feel like the thing I agree most with is the comments from folks about stopping promos and sale prices and whatnot until your house is in order and you’re caught up. Posting on let about how it’s a small team and support is expensive and people have too high of expectations while still actively running promos to bring in even more customers that will require more work to apply bonuses and whatnot to seems… self defeating. Take a pause, get caught up, re-evaluate the prices and start again, imo.
The thing is, he's responding to keywords instead of the content itself.
In fact, no need to change the status quo if he purchased the $200/m AI subscription. Things he should do:
1. Make LET thread ingestor. For example, to jsonl
2. Make WHMCS action parser (or use the existing database/excel that tracks LET user/service/upgrade) & utils such as invoice/bill/payment to service ID
3. Cross-examine the 1&2 for missing or customization gap that happened by email or LET DM
4. Let it ask the team for unexplainable upgrades. Purpose: keep edge case low
5. Make a script that checks the hypervisor for availability (storage/cpu/ram upgrade) given service ID.
6. Combine #5 with a WHMCS script that applies or reject upgrade -> say to customer that the upgrade is unavailable or have to be migrated elsewhere
7. Make another script that uses or modify #1, #2, #3 and create upgrade queue for automatic upgrade and deferred/backlog
8. Create a script that consume the queue created by #7 using #6, then put it on backlog queue if the upgrade is rejected
9. Make a dashboard that observe #7 and #8. Wire a method that triggers #8 to modify specific case from backlog e.g changed the upgrade/move it to another hypervisor/etc then implement the method on #8.
10. Ask the AI to deploy it on the WHMCS node so there's no latency. And if the WHMCS is down, it's also down so nothing is out of sync
They are currently in the process to build their own billing system. It's quite nice, looking forward to when it is being released for everyone.
The ticket system is congested largely due to last month's issues, but it is returning to normal quickly, as there have never been any such issues, we have not had a high workload, and the situation has remained stable.
We aim to be able to improve the situation even more, but I think what I said is simple honesty.
We listen to people who give the right advice, while those who start out prejudiced and attack us for no reason for the sake of it are a different matter.
A lot of things are already in the works as @NotFoundException rightly said in our new billing system.
Does it took human action, or it's fully automatic or with AI guidance?
IMO they heavily use AI, but are good enough in understanding how it works and working on their own to get a properly working product. AFAIK they don't have agents running completely on their own.
I guess that's the problem. Routine, bureaucratic actions should be driven by AI, while humans supervise other agentic AI and make decisions, focusing on Pareto-efficient outcomes.
To try to accommodate our customers, we've included a button that will work on every ticket that doesn't receive a response from us for 24 hours, then reusable every 12 hours.
This will notify an admin who will be with you as soon as possible.
This could be abused, if it does absolutely nothing if you don’t have the adequate resources because we’re still putting out fires from months ago.
It has limitations and those who abuse it will not be able to use it again in the future, I'm pretty sure it will help us a lot especially for tickets that have not been answered for a long time due to oversights or other various reasons.
How can this be the problem, when they use AI on a billing dashboard you didn't see and they didn't use in production yet? I don't expect anyone who values their time to code without AI anymore
In general, judging by all the responses from DeluxHost here, I can conclude that they are a newly born child who is still learning to walk. They make a lot of mistakes and try to fix them - they learn to walk, as every newborn does. Therefore, it makes no sense now to hire them to work as a courier in order to monitor their training for your own money. Perhaps it's worth waiting for some time, when they learn to walk, and only then it will be possible to deal with them. Later. Someday. But not now.
Why not simply just work from the oldest to the newest?
Why not simply add more resources who are not only competent but have experience and can deal with escalation and resolution?
Why make a button that doesn’t solve anything when your growth, the issues, and the systems(both infra and process) clearly isn’t working.
What good is a button that sends a ticket to the top, if you still don’t have the resources to answer?
I don’t think it will help you at all.
I really don't understand the point of complaining about an extra feature added that can only help and doesn't cause problems for anyone
Tickets are still replied to, but it could happen that someone isn't seen due to the high number, so let's give a tool to speed things up.
Our infrastructure is working anyway, since when is it not working? i can see all up, I think there is simply an exaggeration of the problems, which there are, but they are not catastrophic as some would have you believe.
yeah sure
YMI-429363
over 1 week btw.
if you're even in talks with PletX, at least let me know. Easiest is to just let me disable the DDoS protection completely for my ranges like other PletX customers allow me to (e.g. zap-hosting)
🙄
We aren’t humble either wow, I’m getting flashbacks to Dasabo.