All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Have you tested your backup restore time - on a new system?
Lets image that your server(s) goes down - hardware failure, and it will take time to get it fixed. Or your host just vanish. If you have HA, DNS failover, live backup servers etc, all of them also goes down. So you have to start from scratch. But of course, you are a smart guy so you have offsite backups.
Still, have you ever tested it in real life, how fast you can restore your data on a new system and get back online?
I did a test today, created a new VPS on VULTR, used my CentOS server setup script that also installs Virtualmin, and then restored my backups. I did not actually change A records, but I have a TTL of 900 seconds, and changing A records is done within minutes.
Totally time from start to finish: 37 minutes. That's 20 WordPress sites, with around 35 GB of data. I of course had an VULTR account with credit on.
If I had to upload my backups from home, it would have taken me around 116 minutes just to get the backup uploaded since I have a 500/50 internet connection. So that's worst case.
So have you tested your backup restore time, on a new system in this worst case scenario?
Of course, people with larger system will use more time. I have a very little operation.
So how much time would you use? And what do you have to do to get back online?
Comments
That's just common sense, I have scripts which will setup server from scratch in 5-10 minutes (depends on role) and have backups synced to a storage VPS. Picked up another one recently so that even the backup has backups..
we have tested my backup and it takes some hours since my backup server isnt in same dc and it data size 500 gb
I've done restores and they take about as long as expected. I don't have any live services that can't stand downtime so I haven't worried about it. If it mattered, I'd use some kind of HA setup.
For things that matter I use a hot spare. When things go down all traffic will automatically be redirected to the fallback server
Yea, I also use DNS failover and two live backup servers. But this worst case scenario included that all my current hosts went down, and all I had left was my server there I store backup files of my sites.
I should add, I have a few small websites on shared hosting and some hosted email accounts, so the host gets to deal with those servers if they crash. Other than that, most of my stuff isn't uptime-critical.
I do have one live server of my own that I ought to figure out a better recovery plan for. It's not much data, but the hosting requirements are a little bit weird so I can't easily migrate.
As long as I can get my backups/all my websites restored within 24 hours I'm fine with that. I can get my backups up and running relatively fast but if I'm setting up/moving to a different server I usually take my time and make sure I do things right instead of rushing.
I usually use warm backup and then compress it to cold backup, so i have ready to run backup
once MySQL DB is restored all I have to do is git clone into public dir :P
this is the only site I have with some serious stuff
& yes I'm too lazy to write bash for automate this task :P
For my main 2 servers (web hosting) I keep proxmox backups in a different server, vestacp backups in 2 other servers and some site backups (files and sql) also on double copy (crazy for backups!).
Once I had to restore backups because the main server collapsed, I completed the task (copy backup and setup the new server) in about an hour. Backup size was ~100GB.
As of DNS, I use cloudflare (only dns), so, ip change was working in a couple of minutes.