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
They exist at least since 2022, possibly since 2018, have multiple employees (if the name in the support answers are real), and seem to have multiple racks. The disk sometimes is fast (up to 100k IOPS and 14 GB/s) and sometimes slow (100 IOPS). I'm not sure whether I'll keep the server, but tbh it simply is very cheap for even an IPv4. And the support is fast, you have a 14 day right of withdrawal (full refund), and they guarantee 99.93% uptime (although it's unclear whether per year or per month). I will YABS it at least daily in the next days and measure the uptime and steal, let's see.
The world is small
Just discovered that the IP (in 37.114.40.0/24) I got was in 2021 announced by AS41944 @dataforest GmbH (see RIPEstat).
Yes we rented the subnet years ago
Short Offtopic: do you own or rent the IPs you use for your vServers?
So far after migration, the VPS has been fine.
For me as well, only the IO latency (up to 20 seconds) is a bit weird:
For comparison my Strato server for 12€/year:
https://ibb.co/wwjV5fx
Mostly idle, only thing running is statping-ng. It crashes after a few minutes.
Steal goes up to 15% while geekbench is running, during idling it's around 0% to 1%.
At least the GB4 score is still good:
But the disk is weird, sometimes multiple GB/s and 50k IOPS, sometimes slow:
Yabs after migrating to an epyc system:
And it's a bit false advertising as they say that it's Gigabit, but it's only (in the panel and in benchmarks) 800 Mbit/s.
No, the 1gig uplink is shared, so 800Mbits is already pretty good.
Yes, but in the panel there should be 1 Gbit/s. Ticket is now open, maybe it's simply a mistake.
Edit: and I don't think that if the physical node would only be connected with 1 Gbit/s, benchmarks would consistently show 800 Mbit/s.
Yes, the performance does not look good, I also have some io wait spikes every ~2hrs for 2-3 minutes.
I wonder if this is some trashy node they put all the sale vpses on, or do they have these performance issues with regular ones as well. Normally I would apply for a refund a while ago, but with €8/year I'll just keep it.
But given all of the above I would not be thinking about getting anything more expensive from them.
Maybe things will be back to more stable once the constant yabs hammering will stop
Was changed to 1Gbit/s now:
So they have a trash node for common people, and a strong node for concerned people who raise tickets? How does one migrate their service?
Maybe they are watching this thread and are scared of bad reviews on search engines?
I opened a ticket and got migrated some minutes later. They told me that the xeon gold nodes are full, that's all I know
They said that there was something wrong with the template and they fixed it manually, I wasn't migrated to another node.
Maybe, some providers see such offers as so low end, that they don't care if you see a bunch of steal and i/o wait.
We have/had a bunch of providers here on LET doing it that way.
I am pretty sure they are monitoring their nodes, they could have reacted.
But they didn't so I guess its the one above.
However they seem to be willing to move you if you ask.
They said in one ticket that they are building stronger storage nodes, maybe they currently simply don't have enough nodes.
Then put it out of stock? Why do keep selling then?
No idea.
And maybe they aren't used to LET customers YABSing it multiple times daily? My server now is limited to 80 MB/s.
To be honest, it does not sound like a surprise
You've been yabsing the sh*t out of that little fellow
Mine did not have any iowait for over 12 hours during the evening and night in EU, there's a small spike again this morning:
It's a good deal considering it's an €8/year non-Flash deal with clean IPs and sensible specs. Not like those 1 Core 1 GB RAM but with 10 or 15GB Disk plans for $12-15.
Network and Support are fast as well. They are quick to migrate and load balance stuff. I received a reply with solution within 2 minutes.
Also, many people are probably still YABSing.
I fully agree!
I've personally done yabs twice and moved on, maybe the ones that are yabsing several times a day should stop and either apply for a refund, or move on and see what happens in the real life scenarios.
If it would be higher spec, higher price VPS I would have already started the refund process long time ago with these iowaits, but with this offer I don't really mind.
Hi, RLTX Staff here;
we have some customers (mostly from here tbh) that are flooding the nodes with io benchmarks; we are manually applying limits to certains vservers and are working on fine-tuning the auto. parameters (so that no regular use gets affected but such peaks are minimized) that we will apply soon; VPS Nodes all have a raid of nvmes or intel datacenter ssds - which lead to nearly no problems in the last years; CPU or RAM is not the problem - its simply people hammering VPS Nodes with I/O at a level which is far higher than it was in the last 4 years.
Else, as some stated above, we are migrating VServers (some automatically, some after a ticket) to new hosts and are adding new hosts as soon as we get the hardware delivered.
The "VPS in reinstall" bugs were all from faulty debian11/12/ubuntu22 templates that did not start the guest agent correctly, which lead to the panel not recognizing the VPS Setup as completed.
The file-not-found err is a bug in the panel that it does not automatically set the new node after a migration from the hypervisor scheduler, we have contacted the developer and are working internally on a fix.
In both cases, the VMs were nearly always up but only the management was unavailable.
This comment here should now attack anybody, just give a brief summary of the most common problems, its roots and solutions. If you have problems with your service directly, please open a ticket ;D
Once again sorry for the problems some of you had, such numbers are generalle not OK and we are working on improving performance.
Welcome to LET. Do take part in the Black Friday Megathread 2024 here, and offer some sweet deals. Good luck with sales.
Welcome to LET. Can I just submit a ticket to move my vps from E5-2690 v4 to EPYC 7351 node?
Thank you.
sure
If some people are benching servers 24/7, I can, understand all hosters implement some limits which aren't affecting 99 percent of all users. I'm happy with my server and this hoster and hope to see some more nice sales here or on their website.