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.
Vultr Long Term Reviews
Considering Vultr as a replacement for DigitalOcean for a large Elasticsearch cluster due to their larger ram, and their large disk based storage tier.
Anyone been using Vultr, Netherlands PoP long term? Got any reviews in regards to stability & performance etc?
Additionally has anyone been using large amounts CPU/IO for long periods of time and had any issues? Are they like DigitalOcean and just throttle/schedule a bit or is a suspension more likely?
I couldn't see it in the API documentation, is anyone aware of any way to determine the Physical Node you are located on? Its actually fairly important for distributed systems with built in redundancy.
Comments
I can only answer 2.
They are really cool about large amounts of CPU/IO for long periods of time. I used to stress their servers a LOT, and they never complained.
I'm using Vultr from the start and they're stable not exactly %100 but ~%99.
CPU abusing is ok but I/O you'll be throttled. Probably they're not using even RAID and that's a reason why they're slow in terms of constant I/O.
Probably you should create a ticket and ask? Haven't used API myself, so don't know.
Smell benchmark:
Amsterdam is now multihomed.
That's NTT and Level3 at the minimum.
This isn't something the API covers however @DaveA has told me in the past it's unlikely a VM will be deployed on the same host node even when deployed in quick succession. You could probably submit a ticket after deploying and ask about that.
They don't disclose any RAID configurations, @DaveA has made off record comments that they're using some level of redundancy.
I'm dumping vultr at the moment. their API is broke.
You can forget getting a response from them on the issue either.
I'm done with those fools.
Vultr do have RAID setups.
Sauce?
Sources please? Vultr has (as far as I know) never released any information about their RAID setup. There was a topic about it a while ago, they tried everything possible to avoid answering questions about whether they use RAID or not.
They don't have any RAID, some level of redundancy is backups on host node. That's nothing in terms of speed and reliability!
I have a couple of servers in The Netherlands at Vultr, it's pretty good. Fast servers and 100% uptime as far as we have seen (they aren't production servers so we don't use Pingdom or anything).
As for suspension, etc., well, I have never experienced that myself; so I can't answer that.
Are you actually seeing any issues where the IO Vultr provides isn't sufficient for your use case or are you just spinning fanboi hype?
Vultr deleted account of my client twice and refused to reactive it nor to take backups after they got complains about spam (website hacked).
If your account is receiving multiple complaints regarding activities such as spam, then I wouldn't expect them to do anything else. Good on them.
vultr backup sucks sooo bad.
You can only go back 2 days.
I have been using service for a few months now. Rock solid in chicago can't comment on Netherlands.
Bad news is they have been degrading service. When I first started they advertised 3+ cores. They recently reduced this to 2.4. My current server with them is running 3.6. I did some unixbenchmark test with my running server vs new build. Performance was 2179.8 on old running server, clean server showed 1422.8. This was a 768m chicago server BTW
Today it also looks like we are being hit with price increase!
Needless to say I am looking for a replacement!
Dave said in IRC
@nightshade - when did those price increases take effect? They now seem to be the same price as DO and Linode for anything 1 GB or above. Except they have the 20% sale. This is the old grocery store pricing strategy - up your base price, then lower it with a sale. Then terminate sale down the road and hope the price increase sticks.
I think sometime today.
Thinking Choopa is wanting out of the vps business. They clearly don't want to compete anymore.
They're spending millions revamping their product right now, improved locations and new hardware etc. They're still working to bring the storage product to more locations.
You couldn't be more wrong.
I am sure they did the cpu drop to save money. Can't say I blame them but this was one of the main selling points that set Vultr apart from DigitalOcean. (Guessing this is vultr biggest completion) This is what finally got me to move my server off DO. If I was looking today I would pass up Vultr.
You top that with what a 30%+ price increase on at least the lower end servers (before discounts) not a good way to revamp brand. Not sure how much the sata storage plans when up that happened a couple of weeks ago.
@nightshade - it may also be that they were using up bulk of old servers which were higher speed than the ones being used now.
With you hanging around all day together with DaveA in a semi-private IRC channel, it seems as no wonder that you are his private spokesman now.
Or is it semi-public?
Whatever you call it.