All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
One year of service with many different providers: A postmortem.
Hello, everyone. I lurk LEB for good deals and have for some years. I'm a programmer and ran my own business doing web development and client site hosting for a long time (>6 years). I rarely see an objective, thorough review of services from many different providers, so I thought I'd buy services from about ten VPS providers advertised on LEB, run a few non-critical services on them for a year, and see how they hold up. I did this for myself, but since I'm sharing the results, I guess I'm also doing this for all of you.
Here's a disclaimer: As if it weren't apparent, I am only one person with one experience with these many providers over the last year, so just because a particular hosting provider completely hosed my VPS and then left me in the dark with bad support, doesn't necessarily mean it will happen to you. Likewise, just because a provider gave me the best service I've had in my entire life, doesn't necessarily mean you'll get the same thing. All of this being said, if you stick a thermometer in a bathtub full of water, you're going to get a pretty good idea of whether or not it's a good idea to jump in.
From January 30, 2013, to January 30, 2014, I was your thermometer. This review is limited to LEB-advertised VPS services. All of the services I purchased were under US $10/month in price, and generally those offers were VPS with under 1024GB of RAM. I didn't buy any 128MB of RAM VPS, meaning the VPS were all at least 256MB of RAM packages. These VPS were not the bare minimum packages offered, nor were they the best packages offered by providers. They were all middling offers of each of the providers, just that some providers offered more resources than others. This was by design. I'm not comparing providers to other providers, but comparing each provider against my own standards, and sharing the result with you. I ran non-critical services on top of CentOS 6.x 64-bit on all of the VPS purchased, by installing a few applications/scripts of my own on each instance, to give the things some work to do instead of just letting them idle.
I monitored uptime and response time via ping with multiple, physically-separate Nagios instances hooked into the Twilio API for SMS notification support, and a few system resources via the popular "bench.sh" script. All providers have some network downtime. That's a fact of life. But some providers had really disappointing uptime. As someone who's shopped around for rackspace at datacenters, I know some are better than others, and that's a choice the provider makes. I put the blame for network downtime on the shoulders of the provider.
Some VPS had better processors than others, but I accounted for this by being kind to the results. Therefore, I won't say "it had crappy speed" for a VPS with a core speed of only 700MHz, because it's a core with 700MHz and that's not fair. The only thing I care about is whether or not the VPS is sluggish for that VPS provider. Some providers were always sluggish, even with a high core speed rating.
As for support, I contacted support at least one time during the year for each of the providers to get an idea of what the support team was like. For some providers I was hard-pressed to come up with a reason to open a ticket, but I did so anyway just to get an idea.
Now for the meat of this sandwich. The providers I purchased LEB-advertised services with were, in no particular order: BlueVM, ChicagoVPS, FrontRangeHosting (now WireSix), Ramnode, HostFolks, Cloudshards, SecuredSpeed, EDIS GmbH, RansomIT, ServerHub, Webline Services (formerly YourDomainGoesHere).
See the top two comments for the actual reviews.
Comments
BlueVM: This VPS was fantastic (OpenVZ). It was almost always up and snappy. However, late in the year, there was a forced migration of the node from Kansas to Dallas. If I was using this VPS for critical/client services, that might have left a bad taste in my mouth, because they didn't offer any other locations and only gave a few day's notice. However, during the move, my VPS was hosed and the backup corrupted. They could not restore my VPS from before the move, despite assurances in the initial notification email that the move wouldn't affect services. After this happened, the first-line support people were pretty much read from a script, and said they wouldn't credit me for the downtime. However, the support issue was escalated and afterwards, the top-tier support guy really tried to help me, and gave me credit for the downtime. I was truly impressed by their top-tier support.
ChicagoVPS: I bought two VPS through these guys, one OpenVZ and one KVM. These nodes had pretty good uptime and were usually always responsive. I had a few notifications from Nagios but they never lasted long. In the summer of 2013, they were directly affected by the SolusVM hack, and they immediately pulled their SolusVM customer panel. For a while, you could only get things like root password changes (or CD media changes) by opening a support ticket. Their support personnel are not knowledgeable or helpful. They are the "it works for me" kind of support. It takes about two or three times as many ticket replies to get something accomplished as it does with another provider because of the ridiculous back-and-forth trying to coax their support into helping. The replacement they've created for SolusVM is nice for the OpenVZ node, but I ran into multiple errors constantly with managing the KVM instance using their new management interface, and ended up having to go through support (which is, again, a pretty major hassle). Despite the generally terrible support, they had good VPS instances, and I'm actually keeping one of these.
FrontRangeHosting (now WireSix): Oh, boy, where to start... I bought two OpenVZ instances on two different hardware nodes. One of the instances was blown away in a power outage. They said that all VPS on that node were corrupted, and they could not restore it except for giving me a completely new installation. Their (generally bad) support personnel repeatedly used the phrase "Act of God" in replies to my tickets during this process. Apparently their insurance company claimed it was an Act of God and therefore wouldn't approve their claim for damages, but either way that's the worst thing to hear in a support ticket: "It was an Act of God. Nothing to do." They did not credit me for the downtime. The other VPS in Atlanta was generally better, but always sluggish, even for four cores rated over 2GHz. There were times when I simply couldn't work on the node via SSH, and both of these VPS consistently triggered my Nagios/Twilio notifications for downtime. The downtime was never very long, but they were very numerous (at least once every other week). The VPS which was killed in the power outage was also force-migrated when FrontRangeHosting was purchased by WireSix. As we all know, getting a whole new set of IPs is pretty annoying when there's nothing you can do about it, especially when your free time is already at a minimum. After the WireSix purchase of FrontRangeHosting ("merger"), support got better, but they were still obviously only paying for $8/hour support personnel. This is a consistent problem with most of the providers, but I digress. Even after the move (one instance did not have to be moved), both instances were still very sluggish and notifying about flapping uptime states pretty frequently. I do not recommend this provider.
Ramnode: I have nothing but good things to say about them. The instance (OpenVZ) was always very quick, and it felt like each of the two I own have plenty of room on their host hardware nodes. I almost never had downtime, and when I did it didn't last long. I actually bought a second VPS through them late in the game. I only had to talk to support one time, for something administrative, and they were very responsive, very polite and professional, and knowledgeable. I have a feeling these support people get paid about twice what other providers pay their support. Meaning, Ramnode appears to pay for more qualified support personnel. I recommend them highly.
HostFolks: This provider is about on par with FrontRangeHosting/WireSix. I bought an OpenVZ VPS instance in their Falkenstein (Germany) datacenter. Support is really rude, and they literally ignored me late in the year, after a forced migration with almost zero notice to another datacenter. I got an email saying basically "If you want to keep your node in Germany, tell us now or we'll move it to the Netherlands". Naturally, I wanted to keep my node where I'd originally paid for it to be, and opened a ticket almost immediately to support requesting my node be kept in Germany. No response and two days later, my node was shut down and moved to the Netherlands and given completely different IPs. I actually had to open a support ticket to get my new IP addresses. That was shocking. This node was ALWAYS really slow and hard to work with. The "non-critical service" I had installed on this node was a secondary DNS server, and it gave me constant problems that I can only attribute to poor performance of the instance. The uptime of this VPS was also quite disappointing. I do not recommend this provider.
Cloudshards: This little OpenVZ VPS was a great workhorse. I had him configured as a DNS-based load-balancer after a couple of months because he was one of the instances that was always green across the board in Nagios. This instance only came with a single core, but this is what I was talking about above with being fair: It always performed really well, had great uptime, and was responsive. In my single support ticket, they responded quickly and inquisitively in regards to my needs. I was impressed with this provider and recommend them.
SecuredSpeed: This OpenVZ instance was about on par with the other good instances. I had to open a ticket for a question about a network problem that was actually some software I had accidentally misconfigured. They are by no means required to help with that kind of problem, and I knew what was wrong immediately when they pointed out the possibility, but the support guy I talked to was willing to work with me to solve the problem, which I thought was pretty cool. It seems this provider also pays its support personnel better (or at least they are more invested in the customer's needs than others). As far as the VPS itself, it had good response time, uptime, and never triggered Nagios/Twilio notification. I recommend this provider.
EDIS GmbH: I actually now use this OpenVZ VPS in my critical services network. Most of my infrastructure is distributed to account for downtime across any part of the rest of the network, but I feel like I can count on this instance if I needed to. They are a company based in Europe, and I have a VPS on one of their nodes based in Stockholm, Sweden. It's always been pleasant to work with, and support is lightning fast and rock solid. It feels like the support people are the owners of the company, or at least are paid very well. I had only a couple of conversations with them (purely to ask some specific information about the hardware/datacenter to get an idea of how best to fit this VPS into my critical services infrastructure), but I felt good about the transaction. I had very little downtime, and despite the hardware being so "far" away, it always felt like I had a great connection to it with fast SSH sessions, and it NEVER lagged with the workload I handed to it. I highly recommend this provider for any kind of VPS, not just as a good Euro location.
RansomIT: An Australian company. I think it's run by one guy, and he's awesome. I had a KVM VPS installed down under and never needed support. During my obligatory test support ticket, the guy was very quick to respond and friendly. Looking at my notes, there's actually three exclamation points because I opened the ticket on a Friday night (local time) and he replied within a few minutes. The instance had great uptime. I would recommend this provider to anybody looking for a good, solid, inexpensive VPS down under.
ServerHub: I bought two OpenVZ instances from ServerHub. They were pretty snappy instances with pretty good uptime. I had heard some negative things about these guys, but to be honest I thought the VPS I ordered were pretty good. Support was on the ball and friendly, and were willing to work with me on an invoicing issue. There was a single week in the late summer where I had a couple of flapping state notifications, but I didn't have any problems otherwise. I bought a third OpenVZ instance in the fall because I wanted those 6 IPv4 addresses in a datacenter physically close to me. I would recommend ServerHub as a provider.
Webline Services (formerly YourDomainGoesHere): I have had an OpenVZ VPS through YDGH for some years and was always blown away by the support (hey Corey!). Right at the beginning of my year, they were purchased by Webline Services, and I decided to purchase a VPS through the new company and put them to the test. I was pleasantly surprised. I was not expecting miracles because I thought the situation could only go downhill, but I was wrong. Their uptime was a little better for me than YDGH previously since being moved to their New York datacenter. Webline Services had really good support. I still have my VPS with them and will probably be keeping it. I highly recommend them as a provider.
Well, that's all folks. I hope you're able to get something from this review, and make a better, more informed decision about the provider you go with when purchasing your VPS. If this review helped you, please pay it forward in the real world by helping someone else make an informed decision about their VPS provider.
Wow. Thanks for the reviews. Gonna reply first then read
Thanks for the reviews.
I keep hearing people bashing chicagovps but they were the first vps company I have bought from and they have been great (almost on par with ramnode). I haven't opened support ticket yet because I never had a reason to, but the uptime, performance and price is unbeatable.
And ramnode is my favorite choice too, their servers feel really fast compared to other providers even on their lowest spec plan.
I've had that as well. It turned out to be a case of bad routing somewhere and they've been able to fix it for me.
It must have taken you awhile to write all of that up. Thanks for your opinions. Very valuable for me.
@eiYeK8dozai6KahyB
I have the feeling I know your password now. Random username much :-P
Good to get a mention now and again. Thanks for that.
Thanks for the mention. I am not sure who you are but I think you are awesome as well! :-)
Wow - great review. Really appreciate the time and it was fun to read.
It kind of amazes me that I have zero overlap with you, other than a RamNode VPS I bought last night on a semi-whim. A couple of yours I'd never heard of - and a couple I wouldn't touch with a 10m pole (ChicagoVPS).
Maybe we're in different geographies - you didn't say where you are.
Personally, I'm on:
LEBs: BuyVM, Catalyst, and picked up a RamNode last night.
Non-LEBs: 6Sync, WiredTree, Linux VMs in Microsoft Azure, AWS
Would Buy Again Without Hesitation: ServerDragon
Would Buy Again Though the Provider Needs to Fix his Self-Signed KVM VNC (lol): Hostigation
Non-LEBs I've Used in the Past and Like: Linode, KnownHost
@Oliver
pretty much the only "cheap" VPS provider in Aus..
I've been with a lot of LEB providers in the past few years. BuyVM, Hostigation, Prometeus, BlueVM, OVH, DigitalOcean just to name a few.
I would rate DigitalOcean and Prometeus (the new Dallas location) being the most stable providers I've used for the past few years.
I've had 0 problem with Prometeus, I/O always blazing fast, network is very very stable.
I have nothing but good things to say about DigitalOcean. Although I've encountered down times (very very few), they've handled the situation well. And I just love their interface and their support team (I've gotten support tickets answered within minutes plenty of times from them).
You know, after the last seven days... DO NOT BUY CHICAGOVPS. Their support guy "Mark S" is just... woefully, dreadfully asinine in every single interaction. As it turns out, I'm amending my review and dumping my services with ChicagoVPS. I just can't put up with this terrible support anymore, and now they've shut my VPS down for an invented reason (no logs whatsoever on my VPS indicating any problems, and they provided no logs from their end at all to support their claim).
DO NOT BUY CHICAGOVPS. They are as bad as FrontRangeHosting/WireSix and HostFolks. Perhaps even worse if you consider how much worse it is to waste your time with bunk "support" and terrible services than to actually just sinkhole your VPS and ignore you.
Only thing that could help a little bit on this if you use the @ to link to the providers here makes it easy to find there current offers. ALso if you dont mind putting in bold text some reating on support and perfomance and uptime. somethings i like support and perfomance can be out of 10 and then uptime as a % over the year.
I thought about doing that, but I deliberately chose not to assign values to anything because I wasn't comparing providers to one another, only to my standards (which are pretty laid back). I think the reviews I wrote are pretty descriptive of my experiences, and they do that fairly well without any quantified data. I could post data from all of my Nagios logs, but then the bad providers would probably just start making excuses. The hosting industry is full of shitty companies making excuses, and I wanted to avoid triggering that, while still posting some valuable personal experiences to help others make an informed decision.
I think its great but it just means when i was decided which providers to check out i had to reread it since i could not just remember a score insteed. i see what u mean well for example uptime is quanitive if u put a metric its a number backed by ur nagios. I guess the same cant be said about perfomance or support unless u do support by response time.
I answered every question you asked, in full.
Q1 You stated your service was offline, I fixed your VPS issue and brought it back online
Q2 You asked why it was offline, I informed you Nodewatch suspended it
Q3 You asked for logs, I supplied you with the requested logs.
Plenty of providers use NodeWatch, it is not an invented excuse to suspend your service. I'm not sure what I missed here, but I answered every question you asked in the ticket.
It took THREE tickets to coax an answer out of you over this one incident. This has been a consistent pattern for months now, Mark. You had your chance(s), and you blew it. Badly.
Funnily enough, I have a couple of services coming up to 1 year of use so I was considering doing something similar to this - we definitely need more longer-term reviews counter balance the 1-3 month reviews there seem to be plenty of.
An interesting read, especially the bits on support from some of the providers, thanks @L0j1k
Wow so nice history with all of them, pretty good
this is what we all need to hear from our customers, good things and bad things if there is the case of that hope not.