All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Migrationbrr
HostBrr - Storage Server Migration
Dear Client,
You are receiving this notification because one or more of your servers will soon be migrated. Please see the affected server(s) below.
We are performing this migration on a short notice due to unforeseen issues with this server. A migration had been planned eventually due to the outdated and underperforming hardware on this server. However, we are now expediting this.
As this storage server contains large amounts of data it will take up to a month to migrate all the hosted VPS. Due to this, we cannot guarantee when your specific VPS will be migrated. Migration of each VPS can take several hours depending on the size of the disk.
When the migration has been completed you will receive a e-mail with a new IP. If you've been using IPv4 NAT then you will receive a dedicated IP and thus need to SSH by the default port 22.
Your VPS will receive a significant performance boost in part of this migration process, as it will be a system powered by AMD EPYC 7513 and 10 Gbps connectivity.
We apologize for any inconvenience this may cause and appreciate your understanding and patience during this essential process.
Maintenance Schedule:
The following servers will be migrated as part of the scheduled maintenance and may be unavailable for the entire duration or part of the maintenance window:
box9 - 192.168.200.25
If you have any questions or concerns, please do not hesitate to contact our customer support team.
- HostBrr
LowEndBox & LowEndTalk.
Comments
why is it ipv4?
"it will take up to a month"
"may be unavailable for the entire duration"
It is internal IP, not public.
Don't know.
We only use IPv6 on this server.
It stores backups and does not have qBittorrent or containers.
But still v4
oddly specific
which plan is this
Is this one of the Falkenstein servers by chance? I believe HostBrr has been shedding their Hetzner servers so this was going to happen eventually.
It's an uint32 identifier just like BGP router-id.
We don't use IPv4 on backup storage servers, so that we don't get hacked.
Kind of a cool, bonus upgrade
We all remember WebHorizon exits KVM NAT market.
It's waste of IPv4.
NAT service should convert to IPv6-only service.
Correct.
Yeah, in a properly planned migration you would presync whatever is possible without interrupting. fix it afterwards, so you'd only have to sync smaller deltas finally and keep the downtime low.
However, not everyone is into migrations and their specifics... and of course depends a lot on the guest os and setup.
The worst (non deadpool) migration in history was:
We'd rather prefer:
You just gave labze another hearth attack, you know he sweats bullets catching posts like this.
I hope he's doing well, personally at least.
Problem is if everyone at the node has the idea and attempts to copy the data. It would be unusable if the dedicated server has a 1 Gbps connection...
This procedure still guarantees the server is online for the time being, at least.
This unforeseen issues has been hinted at in earlier threads. Due to Hetzner changing how they handle torrenting it is no longer feasible to offer IPv4 NAT services with them. They now actively monitor if their IPs are present on https://iknowwhatyoudownload.com/ and will block IPs that are present for a longer duration. Given that there is little to no provided information in regards to the torrents there it is impossible to determine the offending VM. Abuse notices sent by takedown trolls at least give enough information to trace the VM.
Hetzner has warned that the server will be blocked but has been lenient and give me some time to migrate clients off it. While it has been known for a while, I've awaited Dataforest to be able to deliver a storage server for me on a rush.
Their patience isn't endless with me unfortunately, and this short notice had to be given immediately when the new server was ready.
I see how this can be misunderstood. We migrate one system at a time to reduce downtime. Depending on size a migration can take 30 minutes to 10 hour depending on size of the disk. In total, migrating 100 TB of data will take some time on a 1 Gbps uplink is not idle on a live system.
You can open a ticket to get the IPv4 removed if you wish.
i wonder cant you get temporarily 10gbps?
That would be chaos - and costly.
Upgrading to 10 Gbps results in a new IP address for the server. Since it is IPv4 NAT all VMs servers use the same main IP, thus all servers would need to get a new IP once. It would incur downtime on several hours. Then they'd have downtime once again when their server was migrated and then once more had to change IP.
Not really interested in adding additional work or costs.
Hetzner theoretically could provide 10 Gbps, but that has an extra cost, has limited traffic and the hardware for 10 Gbps may not be there at all... it may not be an option for him
That would be the time to pull a claw and throttle every service to 10 Mbps.
hehehe
Are you suggesting you cannot get hacked via v6?
I didn't think he was that naive.
It's not just the looser person with his self-conscious room that only requires IPv6, if you don't use it he'll look at you like you're nothing!!!
I actually do and always ask providers to enable it if it is not there
I LOVE IPv4
FOREVER!!!
Sorry @labze
I think it's better if you read every @yoursunny cmt with an /s at the end.
Isn't it always?