All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
dediserve.com disaster, please give me advices
magentoadmin
Member
Hi everyone,
I have a VPS account at dediserve.com to host 3 ecommerce websites. Today, the server is dead. No warning from Dediserve about it. When i create support ticket, the staff let me know that [B]"they had an storage array issue on the HV and that corrupted the file system of the VM"[/B] and can only restore my account to the snapshot i [B]1 month ago[/B]. Yes, [B]1 month[/B]. My database backup is generated daily on that server by virtualmin. I cannot access them now. My snapshot feature on this account is enabled, it is supposed to backup daily too. I bought 40 GB disk space and i have 10GB free space for snapshot. I have only used like 5 GB for web data on that server and support staff said that because my total disk is 40GB > 10GB hence the snapshot feature failed, which really doesn't make sense to me. This snapshot feature is not working and their system doesn't have any kind of alert to notice users. I don't understand that Storage array issue but apparently this is not my fault at all and they can simply refuse to do anything about it if I don't store backup myself on local computer. Is that really what hosting company nowadays doing? They rely 100% on user backup, they can delete server at any time, break it for their own reasons and then refuse to do anything about it just because in the terms and conditions states that customer need to take full responsibility? I need some adivces please.
Comments
If you have an important data on your VPS, never rely on the provider to back it up. Always setup a backup to some external source.
No matter which provider you use, ALWAYS have your own backups.
@dediserve to serve some light.
You probably do not want to hear this but yes it is your responsibility, it is your data.
If their snapshot feature requires at least as much free space as you have data then again, that is your responsibility.
I am sure the snapshots were not silently failing, some sort of visual notification must have been available and if you did not check for a month, again that is your responsibility, 40 GB will not fit in to 10GB of space.
Snapshots work on your entire disk image usually, not how much you have used, if you bought a 40GB disk, the entire disk is 'snapshotted' so you need 40GB of free space.
dediserve offer a budget self managed service, it is not fully managed, you are expected to be able to manage and maintain your own systems, especially your own disaster recovery.
I am sure dediserve did not destroy the server on purpose, they probably got no warning either, if your ecommerce sites are 'commercial' then you should have daily offsite backups, don't put all your egg's in 1 basket.
They are probably refusing to do anything further because it is physically impossible for them to do anything else, if the data is gone and they do not have backups, you do not have backup's and you did not have enough free space for snapshots, what exactly do you expect them to do?
Hard lesson to learn, but you probably will never make that mistake again, hardware issues happen to the best of us, there is usually no way to predict it when its really bad.
Lol BBCode.
Yeah and a single post new account, more like they posted else where, and then searched for all locations dediserve has a presence and re-posted.
And I was correct : WHT
Whatever @AnthonySmith says: Dediserve had multiple issues with storage failing while their claims on redundancy are very broad.
@Anthony:
The mininum package they sell is 30GB SSD space hosting and the default snapshot space is configured at 10GB. Why don't they make it configured at 30GB or at least give some knowledge about it? It was there just to show customer that there is snapshot feature, but it won't work even with the minimum package? And does it means if i keep 2 snapshots in history i need to buy 60GB snapshot space???
Snapshots are compressed and usually take half of the disk space used on your VM. By buying 60 GB you would play it safe. All the recent fiascoes make me wonder if their HA (1000% uptime guarantee) even works or is it just another marketing trick.
Same here. Lost one complete site, have to start fron zero again. Thanks ds
I can only assume that you have become a better person and now keep backups.
Live and learn.
totally understand one should have backup somewhere else. I was wrong to rely on that snapshot feature. But there is definitely something wrong with hosting businesses really. I mean I agree customers don't have the right to request backup from provider but why don't provider do it for themselves? They broke the product they sold to customer, which is not customer' fault, and they don't have backup from their end to fix it? Is it too much to ask from customer point of view? I don't even get a sorry for that.
You should have not one backup but multiple backups.
Backups fail too.
Btw can you recommend me a tutorial how ti make auto backups with another box?
@emre22
Simplest way to do it yourself. Using rsync with cron:
http://www.howtogeek.com/135533/how-to-use-rsync-to-backup-your-data-on-linux/
The problem with the hosting business is that over the last 5 or so (probably a lot more) there has been a significant pressure to offer more for less, at some point something has to give.
As a host who offered unmanaged services and has had a disaster I can tell you that is is next to impossible to please everyone, expectations are very high while prices are very low.
The only backup they could have which would 'fix it' would be an entire server image or a redundant SAN or similar, these things can cost a lot of money, in most cases more than the entire profit margin.
Just taking 1 location as an example, if I kept up to date daily offsite backups of customer data I would need to move 40TB daily, that is simply impractical on cost, performance and management when people are paying $7 p/month for a 85GB disk image.
Instead I offer free backup space so customers can self manage their own backups of actual required data which is almost always 95-99% of the actual disk image total size.
I just had a look at dediserve, they offer 40GB of snapshot storage for only an extra €3.00 p/month, you could get an additional 10GB of NAS storage with them for only €1
They offer you every opportunity to make your data safe and redundant from what I can see.
Please dont take this as me having a go at you, but you really have no one to blame but yourself as your working on the assumption that unmanaged hosts should take care of disaster recovery on your behalf, that view is a bit unfair.
The fun acctually doesnt really start until you try to recover/re-implent a backup. Then things usually get interesting xD
Yes, exactly. Thus mentioning the "multiple backup" solution!!!
@Anthony:
Til this moment, I'm not sure why snapshot failed. If it is that i don't have enough space for new snapshot because of old snapshots then it still sucks anyway. Because in dediserve admin panel, i don't see feature to automatically delete old backup or "keep backup in x days". Which means i need to delete old backup manually for a scheduled task. They should at least send me an alert message when the snapshot failed.
The lower the price, the higher the expectations. The higher the price, the lower the expectations.
Don't ask me how that makes sense :P
It makes perfect sense. Those who are willing to pay a high dollar for services and know what they are paying for are usually those who have a better idea on how things work and are prepared. They also understand what does self-managed or unmanaged mean.
The majority of those who expect a superduperawesomekickassforthelord server for $6.00/yr expect to have Google-like service and backups because well... you know :P
You nailed it!
That goes for all different types of business...I see it everyday in my main business which is not related to technology.
Also setting up a raid may be useful for the future
Which is great for open source projects.
"What do you mean you won't drop everything and solve this specific problem which is causing me to lose money when I don't pay you any?"
I don't ask for any awesomekickass server. If i break my server, i will accept it. This is the provider who break my server. They sold me a product and they screw it up and you are telling me that they don't have any repsonsbility in it? Not even a sorry?
What's you are saying is not customer oritented at all. It's not just downtime. It's dead completely and you make it dead, not customer. Even if you have backup for customer, do you think customer will feel safe to stay with you anymore when it might be dead anytime even without any attack from hacker or bad actions from customers. Why don't any of you here think that at least you should not screw up the product you sell and have a backup for your own bad action?
If you think hackers are the only force that can take out your server where you would say "well atleast that would be smth id understand" you are bound to hit your next wall with the next host you will choose. That is if you continue to stick to low budget solutions. Either do proper backups and go with top of the line budge solutions (ramnode, do, vultr, time4vps..), have a project that doesnt rely on backups or is not that Important, or choose a managed solution and be happy (liquidweb, wiredtree).
I understand most people commented on this post are providers. My opinion is from customer point of view and it is simply this:
When you sell a hosting to customer, no matter it is cheap or expensive, if you break it, not customer, then you should not ask customer for backup. You should backup too for your own mistakes.
That's quite a shallow way of thinking. Disaster recovery is necessary for any line of business. While this does indeed raise questions on Dediserve's setup, let's be real here. Your data still remains your data. If it was important enough, you should have had your own backups. Remember the 321 rule? 3 copies, 2 mediums, 1 offsite. If you blindly trust the same provider to do backups for you, and rely on it ultimately without even checking if it works, ultimately the blame lies on you.
Yes you should. Thats because if the Service is UNMANAGED it is your responsibility to always have Backups. Most providers offer a % SLA but even that includes down time which includes but is not limited to hw failure..
@Ympker:
When i buy an unmanaged server, i expect to run, configure the server myself, i don't expect you to kill my server and ask for my backup.