Renewed my 1-credit.
Deleted the Romania container.
Made new container in Canada, replacing the Gullo VPS there ($2/year for 256MB) that I'm canceling.
5 containers running, 192MB and 10GB unallocated.
Waiting for Sweden and Debian 12 and Loadbalancer editing.
Location wishlist for next year: Singapore, Chile/Brazil, USA-Texas, Iceland, Italy/Spain, Kenya/Ghana.
Sweden is now live although IPv6 isn’t yet routed.
Memory editing is now live in almost all locations.
We’ve also entirely replaced the Micronode worker on the nodes to prevent timeouts when provisioning, this is vastly more efficient however it means that instance creation is backgrounded and instances may take up to 5 minutes to come online after provisioning.
Not currently, we tried to get an offer together for yesterday but we still have a few things to tidy up first. We will have an offer soon.
@mee2 said: @natvps_uk How about Black Friday offer for UK NAT? We never had one like for FR NAT.
We’re no longer selling individual VPS’ the amount of support required for a £1 per year VPS simply made it not worthwhile. Micronode instances are currently available and we may even have an offer soon.> @cybertech said:
The Micronode panel is currently going through a rewrite to add some requested features.
It’s currently written in PHP which I have regretted since launch. I’m rewriting it in Python (django) and I’m at a bit of a loss on what to do with the existing VPS line of services. There’s a lot of logic in the panel around VPS’ and a fair amount of it isn’t reusable for instances. It’s also two things to maintain that effectively serve the same purpose.
Most clients have moved to instance credits but we still have around 200 VPS clients with active services.
This isn’t something that’s going to be finished overnight, in fact it’s likely several months away at this stage.
The idea I have currently is we allocate all active VPS Customers at least a single instance credit and convert their VPS’ to instances.
I feel this is pretty fair considering most individual VPS’ are around 128-256mb RAM and 6-20GB disk. Everyone will end up with more resources for this transition period.
We’d make sure the specs were matched so nobody would lose out.
The downside is when it comes to renewal of these services we would not be able to honour the VPS pricing. We would offer a discount but it’s likely to be around the £10 per year mark for 512MB RAM and 20GB disk.
I’m keen to hear people’s thoughts on this, nothing is set in stone and my mind can be changed.
@natvps_uk said:
The idea I have currently is we allocate all active VPS Customers at least a single instance credit and convert their VPS’ to instances.
There was a half-credit product last year.
VPS of less than 256MB and 10GB could be converted to half-credit, renews at £5 that is closer to VPS pricing.
They can open a ticket to upgrade to 1-credit for £10.
@natvps_uk said:
The idea I have currently is we allocate all active VPS Customers at least a single instance credit and convert their VPS’ to instances.
There was a half-credit product last year.
VPS of less than 256MB and 10GB could be converted to half-credit, renews at £5 that is closer to VPS pricing.
They can open a ticket to upgrade to 1-credit for £10.
That could be an option, the issue is a lot of the VPS’ are on the UK nodes which are HDD and a fair amount of the packages have a large amount of disk. Obviously this doesn’t scale well when all other locations are on SSDs.
I think the safer option would be to migrate everyone to a full credit with the option to downsize at renewal via a ticket.
@natvps_uk said:
a lot of the VPS’ are on the UK nodes which are HDD and a fair amount of the packages have a large amount of disk.
Make a UK-HDD location with double/triple/quadruple storage, and a UK-SSD location with normal amount of storage.
You can then convert an existing VPS to fewer credits.
As an Instance-only user, so far I avoided UK because it's labeled as HDD that I perceive to have less value.
If UK-HDD would have double storage, I may start using it.
If you got different Node sizes and Types, I would advertise a minimum disk size, depending on Location you might get more for example if its SSD.
So you would not have to bother with that.
Pick the Language you like or can code best in.
Ideally you pick the Language which is best for the Application, optional though.
Python or PHP isn't terrible, I use booth depending on use case.
I would suggest you try to refactor your existing panel, otherwise you have to deal with all the overhead migrating people again.
Not to speak of, writing everything again from scratch.
I might be able to help you, however can't make any guarantees.
Started adding billing and support tickets to the new panel. Here's a preview. The UI isn't finalised at all and a fair few changes are yet to be made. Its starting to come together though.
Not as such, the storage VPS lineup is likely going to be replaced when our new panel goes live and it will be available then.
This is likely to occur towards the end of January.
@natvps_uk said:
The worker on the AMS Loadbalancer crashed an hour or two ago, I've brought it back online now.
I feel I'm guilty of breaking the poor worker.
I deleted a loadbalancer and immediately re-created it with a different backend.
Changes didn't apply after an hour and I didn't read this forum, so I moved the webapp to another provider for now.
@natvps_uk said:
The worker on the AMS Loadbalancer crashed an hour or two ago, I've brought it back online now.
I feel I'm guilty of breaking the poor worker.
I deleted a loadbalancer and immediately re-created it with a different backend.
Changes didn't apply after an hour and I didn't read this forum, so I moved the webapp to another provider for now.
It was actually dead at the time you deleted the LB, certbot had hung and consumed 8GB RAM causing the worker to be OOM killed.
I'm glad that loadbalancer editing will come in the next panel version.
This would eliminate the need of deleting and immediately re-creating.
Going to try to add an LB in APAC and the US as well as a fair few clients are loadbalancing traffic from Canada and Korea over the LB located in Amsterdam.
In fact I might just grab an additional IPv4 address in each of our locations and stick an LB on every node. I’ll have to think it over before committing to it.
Comments
Renewed my 1-credit.
Deleted the Romania container.
Made new container in Canada, replacing the Gullo VPS there ($2/year for 256MB) that I'm canceling.
5 containers running, 192MB and 10GB unallocated.
Waiting for Sweden and Debian 12 and Loadbalancer editing.
Location wishlist for next year: Singapore, Chile/Brazil, USA-Texas, Iceland, Italy/Spain, Kenya/Ghana.
Accepting submissions for IPv6 less than /64 Hall of Incompetence.
@natvps_uk any plans for SG? I remember you saying SG and Poland were on the roadmap.
The Ultimate Speedtest Script | Get Instant Alerts on new LES/LET deals | Cheap VPS Deals | VirMach Flash Sales Notifier
FREE KVM VPS - FreeVPS.org | FREE LXC VPS - MicroLXC
Possibly, I would love to add new nodes in these locations but I’m not going to expand any further until we fill the current nodes up a little more.
Seems Amsterdam, Netherlands networks is not stable.
Yeah looks to be a DDOS at the upstream. Nothing we can do but wait it out.
Sweden is almost ready to go, hopefully should be live over the next few days.
We ended up using our own ipv6 range for this and we’re waiting on the provider to announce our IPs.
Sweden is also our first location to support Debian 12 with all other locations to follow in the near future.
Sweden is now live although IPv6 isn’t yet routed.
Memory editing is now live in almost all locations.
We’ve also entirely replaced the Micronode worker on the nodes to prevent timeouts when provisioning, this is vastly more efficient however it means that instance creation is backgrounded and instances may take up to 5 minutes to come online after provisioning.
Debian 12 is also now available in all locations.
Any BF offers for credits ?
Why?
@natvps_uk How about Black Friday offer for UK NAT? We never had one like for FR NAT.
MOARRRRRRRRRR
I bench YABS 24/7/365 unless it's a leap year.
Not currently, we tried to get an offer together for yesterday but we still have a few things to tidy up first. We will have an offer soon.
We’re no longer selling individual VPS’ the amount of support required for a £1 per year VPS simply made it not worthwhile. Micronode instances are currently available and we may even have an offer soon.> @cybertech said:
Soon hopefully!
🤟
Waiting for storage editing and loadbalancer editing.
Time to delete the unsightly You currently do not have any MicroNode Virtual Private Servers section?
Accepting submissions for IPv6 less than /64 Hall of Incompetence.
In time, we still have a fair few VPS customers. We're not stopping renewals we just won't be selling anymore.
There is now
The Micronode panel is currently going through a rewrite to add some requested features.
It’s currently written in PHP which I have regretted since launch. I’m rewriting it in Python (django) and I’m at a bit of a loss on what to do with the existing VPS line of services. There’s a lot of logic in the panel around VPS’ and a fair amount of it isn’t reusable for instances. It’s also two things to maintain that effectively serve the same purpose.
Most clients have moved to instance credits but we still have around 200 VPS clients with active services.
This isn’t something that’s going to be finished overnight, in fact it’s likely several months away at this stage.
The idea I have currently is we allocate all active VPS Customers at least a single instance credit and convert their VPS’ to instances.
I feel this is pretty fair considering most individual VPS’ are around 128-256mb RAM and 6-20GB disk. Everyone will end up with more resources for this transition period.
We’d make sure the specs were matched so nobody would lose out.
The downside is when it comes to renewal of these services we would not be able to honour the VPS pricing. We would offer a discount but it’s likely to be around the £10 per year mark for 512MB RAM and 20GB disk.
I’m keen to hear people’s thoughts on this, nothing is set in stone and my mind can be changed.
There was a half-credit product last year.
VPS of less than 256MB and 10GB could be converted to half-credit, renews at £5 that is closer to VPS pricing.
They can open a ticket to upgrade to 1-credit for £10.
Accepting submissions for IPv6 less than /64 Hall of Incompetence.
That could be an option, the issue is a lot of the VPS’ are on the UK nodes which are HDD and a fair amount of the packages have a large amount of disk. Obviously this doesn’t scale well when all other locations are on SSDs.
I think the safer option would be to migrate everyone to a full credit with the option to downsize at renewal via a ticket.
Make a UK-HDD location with double/triple/quadruple storage, and a UK-SSD location with normal amount of storage.
You can then convert an existing VPS to fewer credits.
As an Instance-only user, so far I avoided UK because it's labeled as HDD that I perceive to have less value.
If UK-HDD would have double storage, I may start using it.
Accepting submissions for IPv6 less than /64 Hall of Incompetence.
If you got different Node sizes and Types, I would advertise a minimum disk size, depending on Location you might get more for example if its SSD.
So you would not have to bother with that.
Pick the Language you like or can code best in.
Ideally you pick the Language which is best for the Application, optional though.
Python or PHP isn't terrible, I use booth depending on use case.
I would suggest you try to refactor your existing panel, otherwise you have to deal with all the overhead migrating people again.
Not to speak of, writing everything again from scratch.
I might be able to help you, however can't make any guarantees.
Free NAT KVM | Free NAT LXC | Bobr
ITS WEDNESDAY MY DUDES
Started adding billing and support tickets to the new panel. Here's a preview. The UI isn't finalised at all and a fair few changes are yet to be made. Its starting to come together though.
The worker on the AMS Loadbalancer crashed an hour or two ago, I've brought it back online now.
If anyone has created a LB in the last few hours you may need to delete it and recreate it.
We'll be adding further redundancy to Loadbalancers once we get the new panel live.
Hi there, will Debian 12 be available on Storage VPS instances in the near future?
Not as such, the storage VPS lineup is likely going to be replaced when our new panel goes live and it will be available then.
This is likely to occur towards the end of January.
Good to hear that. Great services, keep it up.
I feel I'm guilty of breaking the poor worker.
I deleted a loadbalancer and immediately re-created it with a different backend.
Changes didn't apply after an hour and I didn't read this forum, so I moved the webapp to another provider for now.
I'm glad that loadbalancer editing will come in the next panel version.
This would eliminate the need of deleting and immediately re-creating.
Accepting submissions for IPv6 less than /64 Hall of Incompetence.
It was actually dead at the time you deleted the LB, certbot had hung and consumed 8GB RAM causing the worker to be OOM killed.
Going to try to add an LB in APAC and the US as well as a fair few clients are loadbalancing traffic from Canada and Korea over the LB located in Amsterdam.
In fact I might just grab an additional IPv4 address in each of our locations and stick an LB on every node. I’ll have to think it over before committing to it.
Have you considered doing https passthrough?
Works via the SNI, which provides the domain/target in plain text.
However doesn't work with ESNI.
Free NAT KVM | Free NAT LXC | Bobr
ITS WEDNESDAY MY DUDES
I considered adding it as an option, we actually did this on the old proxy service back when we were on solusVM. It worked but had a few limitations.
certbot is too inefficient.
Nowadays I use acme.sh when I need a standalone ACME client.
HTTPS to HTTP (current loadbalancer, Flexible on Cloudflare):
HTTPS to self-signed HTTPS (Full on Cloudflare):
HTTPS passthrough (microLXC):
I'd like to see all three modes supported.
Most frequently I use the second mode.
Accepting submissions for IPv6 less than /64 Hall of Incompetence.
Is the Romanian node officially gone or it’s just being “Calin’s server”
Why?