All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
TNAHosting Ryzen VPS rebuilt after PDU/NVMe failure; data lost, 1-month extension
On September 24, I found my TNAHosting Ryzen KVM VPS running a fresh Debian installation. The previous filesystem and configuration were gone. The new guest logs begin on September 22 at 19:57 UTC, when a new SSH host key was generated. I had not requested a reinstall.
In my support ticket, TNAHosting said a rack PDU failed on September 21. According to them, the physical host did not recover after power was restored, and both NVMe drives in its mirrored storage became inaccessible despite recovery attempts. They rebuilt the host and provisioned my VPS afresh.
TNAHosting said there was no recoverable remote backup for my VPS because that feature was opt-in at the time. They say automatic remote backups are now enabled by default. I received no direct notification during the incident; TNAHosting acknowledged that it posted status-page updates but did not directly notify every affected customer.
Some data that existed only on the VPS cannot be recovered. I restored what I could from separate local copies, but the outage and reconstruction disrupted workloads that other people rely on. When I requested compensation, TNAHosting extended my service by one month and declined compensation for lost data or rebuilding work. On a $24-per-year plan, that extension is worth roughly $2 in service time.
I recognize that I should have maintained independent offsite backups. I am addressing that now. Still, the reported failure of both mirrored drives, the fresh installation, and the lack of direct notice have seriously damaged my confidence in this provider.
I have tried to attach a cropped screenshot of TNAHosting's support reply. Account details, service identifiers, IP addresses, and private ticket information have been removed. TNAHosting is welcome to clarify any part of this account or provide a fuller incident timeline.
TNAHosting's relpy: https://ibb.co/v6Xc5Z99
Comments
Zero backup == Zero worth
From my understand and summary for your easy
1: you had not requested a reinstall
2: They rebuilt the host because the nvme failure
3: there was no recoverable remote backup for your VPS
4:you received no direct notification
5:TNAHosting acknowledged that it posted status-page updates
6: but you need(notify every affected customer.) didn't they already do by posting it public in their acknowledged as you say?
7: you restored what I could from separate local copies but still some only on vps
8: you requested compensation
9: TNAHosting extended your service by one month
10:declined compensation for lost data or rebuilding work (didn't they already do? or you want more?)
11: On a $24-per-year plan, that extension is worth roughly $2 in service time.(so if the same plan worth 10$ month you would be satisfied or you need another full year?)
so the only problem would worth all this that they didn't send a one by one email for each affected customer
If you review their Terms of Service they are responsible for no more than the cost of the service that wasn't delivered or at max, the value of the last payment for the services. Backups are also outlined in the ToS as your responsibility, along with any failures not being theirs. All and all, I think it was nice of them to compensate you for a month of service, which sounds like the amount of service time that was lost, as well as committed to providing a free backup for you going forward without having to opt-in, which they don't have to do.
https://tnahosting.net/tos/
I understand you are frustrated and they probably could have communicated better but it doesn't sound like they have been unreasonable and have compensated you for the lost service time. They also seem to be implementing a back-up service automatically going forward for free, which it sounds like you didn't opt-in to previously.
If your services are so important that downtime impacts you so much, maybe you should consider having a hot-spare (replicated server) available on another provider so that if something like this happens, you can switch over to your backup server to avoid downtime. You should build the redundancy into your production plan instead of getting upset with a provider for having unexpected issues they couldn't control.
My 2 cents.
Cheers!
Their response was completely reasonable, as was the 1 month extension. I don't view them negatively because of this and the only thing I'd criticize is the fact that they didn't notify you.
Keep backups if you care about the data.
What did you expect to get, honestly? $50?
A public apology. That's the hot new thing!