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
I'm getting antsy here, he never replies to me. My previous ticket is unanswered since August...
Why bother if I don't come on LET and be loud about it, right. At this point, I will settle for a refund.
Have you received an email since November 23?
I understand that there is a significant problem.
I can understand that he can not answer all the tickets.
I do not understand and I am angry that he does not give news.
That email was my last one. But I think he is counting with our choice even we didn't get any answer back.
I was replied that the raid was resyncing for the first time then and it's done now.
I just made another test and the disk IO is 250MB/s now, much higher than before tests.
I guess he's just busy now.
Please be patient and wait for some more seconds.
Maybe it's more complex to operate on 23media since I found just now the Hetzner one seems newly built.
>
More expensive* - apparently @Falzo got the 23media offer because he had non-LET Specials. Anyone with LET specials only got the Hetzner offer, since 23media RAID 10 storage boxes aren't sustainable at our rate.
It seems that the PayPal account got unlocked, got an invoice refund confirmation with a paypal transaction ID (yet to show up in PayPal itself, though...).
me too, and i have 2 vps with him, some of you got provisioned after replied the e-mail. i replied and got ignored.
Sooooo, NO DATA, NO RESPONSE, NO NEW VPS... and someones keeps defending him.
I've just received a answer to my ticket by Ashley telling he was just working to my new vps.
Nobody is being forgotten, everyone who has requested a new VM and provided the OS will be provided with the VM and their renewal date updated.
The new VM's are Virtualizor and KVM so you have full control and access to templates so no longer manual reinstall as per CEPH + Proxmox.
However is still a manual process for me to re-setup VM's and delayed awaiting for OS's required.
If you haven't already please reply or create a new ticket with the OS you require and any other requirements you have. Will be setup as quickly as possible.
Can you give us a clear "yes" or "no" about the data on the Ceph cluster? Is there still a chance of rescuing it or is it definitely gone?
Don't ya get it! It's gone
Right now there is no bug fix for the problem so unable to start the OSD's to extract data.
So if you want a clear yes or not, I can't only 100% say it's gone, however there is still hope if they do realize a fix and the data is accessible to extract. But that is completely outside of my hands if when and even if it does happen. They have already delayed their next release full of past bug fixes by a few weeks.
The thing is, Ashley has been sidestepping this question the entire time, going no further than "we're working on it", and often no comment at all.
How am I supposed to decide how to proceed when I don't have any indication if the data is coming back tomorrow, next week, or not at all?
Edit after @AshleyUk's answer:
Okay, that's something to go by, at least. Thanks for the update.
I do see a problem there: while you're waiting for a fix from the Ceph devs, those drives are lying around costing money without generating any revenue… Are you going to swallow that, or eventually cut your losses and overwrite the drives for a new use case?
We will await to see if the next release has the fix, however our latest email that did go our stated that data has been lost and this is why we are setting up new VM's on local storage and asking for information required for this.
Before that point we was working to get the VM's back online as they are hence the working on it updates.
Either way we will be moving all services to local RAID, if any data on the CEPH does become avaliable we will look to help copy this to new VM's. However at no point will the old CEPH VM's come online as a working running VM.
ETA for refunds still January? Or should we expect some delays?
I have not received a response yet either. I just made simple questions regarding receiving the new VPS at Nov 23 (4 days ago). I am still confused and waiting for either vps or money but I need some answer before receiving the vps.... which still under no response. So... lets see few more days. Because he might be seriously busy. If 15 days can give him time to send a proper response to my ticket then I am ok. I will give him 15 days.
Thanks for your time and messages here @AshleyUk
I say "here" because it seems some of your emails got lost so it's good to have news. We now just have to cross fingers, pray for those believers, and wait :-)
@AshleyUk Does anyone who had the storage bonus previously offered by paying long term get those extras transferred over to the new local storage service, or is it just base spec (1c/2g/1tb) no matter payment term?
I just got the new VPS earlier today and it is only the base spec (1 TB instead of 2 TB, and no additional 100 GB root drive).
I just hope my data was not lost...
I wonder what customers would say if a firmware issue in a hardware raid controller flipped enough bits to trash filesystems. It happens.
Just learn your lesson and have redundancy for data you need 'available' even if one source of truth goes kaput.
I keep hearing ZFS (and maybe btrfs) takes measures to catch this type of thing. I wonder why they're not used more.
Neither of them are entirely production-ready on Linux. btrfs and ZFSonLinux have both been "stable" for a while now, but there are still occasional bugs — which seems to be the exact same situation as with Ceph. ZOL also requires custom kernels because ZFS can't be merged into Linux mainline due to license incompatibilities.
So if you want a tested, trustable checksumming snapshotting filesystem, you need to run FreeBSD or Solaris.
Also, checksumming can detect bit flips, but to correct them, you of course need some in-filesystem redundancy (i.e. btrfs-/ZFS-based RAID).
And is there a clear yes or no for the local storage servers? I have one of those non-storage specials in Frankfurt and it's also offline -> https://www.lowendtalk.com/discussion/69806/zxplay-cyber-monday-kvm-deals-free-domains/p1
I wonder why this was affected. Wasn't that before CEPH storage and all that? Ticket would be #465914.
I'd say yes, according to what Ashley posted here yesterday: "Either way we will be moving all services to local RAID."
Sounds like Ashley is going to setup the requested VMs on the good old Local Storage + Virtualizor plattform. My request (ticket #408705) should also be on the queue. Let's give Ashley a few days to process our requests.
Great news, I provided this info 5 days ago have heard nothing since. Will it be this year?
Me too. And I've open a ticket about this.
Since some of others have no new vps yet I think this problem would not have the first priority.
ZFS do have many features and good at data security.
However it has been never stable on Linux world yet as a migration from bsd world.
I'm in the same boat... no data, no VPS, no ETA, no communication.. :S