Limitless Hosting - suspending site without any warning, bad support, bad uptime

1

I wanted to share my experience with Limitless Hosting, and also wanted to know suggestions as well as the opinion of the HostedTalk community- as well as hear about experiences from other users of Limitless. I also got an account from another web host called MyW.pt hosting, and an account on DigitalOcean (through free credits, so I will probably not have this long-term). I did want to try out different hosts and have test websites running on all.

I got a reseller plan from Limitless Hosting. I figured a reseller plan would be better as it means I can create separate DirectAdmin accounts for separate websites.

So far I have only been doing test projects - so basically I am just testing, the site is not even complete, have not shared the links with anyone (except a few people for testing), but it is live. (This is also the case for MyW.pt and Digital Ocean).

It was working fine for a while. At some point I also put up uptime monitoring tools for all the hosts. My experience has been:

  • 100% uptime for DigitalOcean, never had to contact support
  • MyW.pt is not 100%, it does go down for like a minute or 2 once a week or so, and once it had a 10 minute downtime. I personally have not noticed the site being down, this is just according to uptime monitoring tool. Their support is OK, not very fast and I think they don’t respond on weekends. But I never had a major issue so the early response time was not crucial here, I would hope they would respond faster for true emergency issues.
  • Limitless has had the lowest uptime among the three, going down for a couple minutes every few days and an extended downtime for like 18 minutes once. It’s certainly the worse one among the three. I have also not noticed this myself, again relying on the uptime monitoring tool.

Now, the problem:
I woke up to seeing that there was a about 10 hours downtime that was still ongoing on a site hosted at Limitless. I contacted support through ticket, but they would only respond several hours later. The reply was simply that there was abuse so it is suspended and that I should fix the site. Now this site again is as low traffic as it gets. I went back and forth with the support before it was more than 18 hours when they finally took off the suspension.

I was never informed about the suspension, had to find it out from notification from an uptime monitor. When I asked, I was simply told that there was abuse, and on further questions they said I should improve security of the site.

I should also note that I found myself locked out of the DirectAdmin account on Limitless, I had to manually add the IP to the whitelist. Again, this problem is also not on other hosts, but Limitless apparently just likes to add things to blacklist immediately and just don’t care that the clients are affected.

So I think Limitless Host is simply not a good option for any production site - they have bad support who aren’t willing to help, bad uptime compared to others, and extreme suspension policy.

I do want to ask the community here what steps I could take here. I assume there was a DDoS attack? Sign up for CloudFlare and have the free DDoS protection from there? What else could be done? I wouldn’t want production sites to go down for such an extended period where the hosting provider does not seem concerned to have the site back online. How would I avoid this scenario?

1 Like
2

I think the first part is to figure out why your site went down for abuse in the first place. If support gave you any technical detail then have a look at it carefully and fix the issues. If you assume it went down due to DDoS and you move your site elsewhere, only to discover you had a vulnerability in a WordPress plugin, you’d have done a lot of work for nothing.

If you are experiencing DDoS attacks - or you think you could, Cloudflare could be a good fit. The “always online” functionality may also plug the gaps and keep your site online during the small downtime windows you mentioned across your other hosting providers.

In terms of a solution - you already use digital ocean and you seem happy with their uptime and product, so maybe just go with them? If you want to add other providers into the mix then perhaps try some of the other digital ocean competitors like Vultr.

5 Likes
3

I received this response:

I see that you had an issue with our hosting service, and I apologize for an inconvenience caused.
Looking into the issue in detail, I see that you have a reseller hosting from us, and a sub-account was suspended due to the reason of “Abuse”. The downtime was for only the website which was suspended, not for your whole account as we never suspend whole reseller account unless there’s an intentional abuse repeatedly. From past few days, we have been trying to find out the abuser which has been causing high load on our server. We suspended some of the sites that were causing this abuse, and thus found a CPU utilization hike from your domain too at the same time. Please try to understand that we have to take actions immediately to prevent other sites from slowing down, and for reseller sub-accounts, we usually do not inform users on promotional packages which comes with zero tolerance policy, yet we give chance to users to fix their issues and then un-suspend it, just like we did same with your account.

Anyways, you can request for the refund if you are not satisfied with our support or give us a chance to serve you better in future.

I think the first part is to figure out why your site went down for abuse in the first place. If support gave you any technical detail then have a look at it carefully and fix the issues. If you assume it went down due to DDoS and you move your site elsewhere, only to discover you had a vulnerability in a WordPress plugin, you’d have done a lot of work for nothing.

The site has been back up without any issues. I have made no changes to the site at all so far.

If you are experiencing DDoS attacks - or you think you could, Cloudflare could be a good fit. The “always online” functionality may also plug the gaps and keep your site online during the small downtime windows you mentioned across your other hosting providers.

I think I will give it a try. Is the “always online” functionality available for free users as well?

In terms of a solution - you already use digital ocean and you seem happy with their uptime and product, so maybe just go with them? If you want to add other providers into the mix then perhaps try some of the other digital ocean competitors like Vultr.

I tried DigitalOcean due to the generous free credits they offer so it’s not a long-term thing. They’re also quite expensive. Ideally I would want to make it work with a lower cost option than DigitalOcean, though of course if it doesn’t work out I will have to consider them.
Thank you for your thoughtful response.

4

WordPress? It could be a kind of one-off issue like a brute force attempt, or plugin auto-updates that ended up eating more CPU than your hosting service considers to be fair.

The reply you quoted from them does refer to spikes, which is why they tell you to fix your website, but I bet that at the same time they know it doesn’t take much to hit the resources cap.

I’d start by asking them the resource caps on your account, and if it extends to the subaccounts or if each subaccount has its own cap.

Please understand that shared hosting comes with limits, no way around that, otherwise it would have to be way more expensive than it is today… kind of pay peanuts get… kind of thing.

That said, they should never suspend your site without at least giving you a heads up.

1 Like
5

It sounds like they don’t know what accounts were responsible for an excessive amount of resource usage on the server, so they just whacked every account that they saw in “top” or something roughly equivalent until the problem stopped.

I can sympathize with how difficult it can be to tell the difference between the cause and the symptoms. A symptom of resource abuse is that users who aren’t actually the cause of the abuse can appear on the top of utilization simply because processes that do not complete as quickly as normal can stack on top of themselves. For example, wp-cron triggered by search engine crawlers on a struggling server might rise to the top, but not actually be the cause of the problem that caused them to rise to the top. This is why you can actually, sometimes, lower system load by not limiting CPU usage per user (and why I dislike Cloud Linux as it appears to solve a problem on the surface but often causes the problems it tried to solve, and even seasoned admins will overlook it).

But that’s their burden as a sysadmin, you have to be able to dig further than that.

1 Like