Snogs Site was down, it is now back up ... move along, nothing to see here

because cpanel kept raising their prices it forced me to not use any panel. now i do everything by the command line. in fact i don't like panels anymore. i had more server issues with one then without.

saves 60 usd a month
 
A backup isn't a backup if it's on the thing being backed up 😅
Amazon S3 is very cheap and easy to backup to via cPanel, I now have daily backups retained for 7 days and weekly backups retained for 4 weeks. Something I wish I could have done on CyPa but their instructions didn't work. Their support team also didn't believe me when I said all of their images in their docs were broken even though I showed screenshots and linked to a few embedded images that were still referencing a WordPress installation.

Their incompetence truly knows no bounds.
 
Are you hosting with a certain someone who shares a namesake with the default Windows Browser? 🤔
Oh no… 😬 That’s not a name I’ve thought about for a while.

Is there an SLA published anywhere for response times to outages, needing backups restored, etc?
There is not. Because our response times are the same as they’ve always been. Essentially business day support is guaranteed but many things are dealt with out of hours, due to our robust monitoring that notifies us when outages and ticket submissions occur. On the one hand that means that all of us are essentially “on call” permanently but in reality, things are so stable that it is extremely rare for anything to go wrong, let alone go wrong out of hours.

We guarantee 99% uptime as documented in the cloud services agreement. In reality, actual uptime is more than 99.9% on average and most of that 0.1% is down to data center outages, and most of that is planned maintenance which gets communicated ahead of time. The small portion of issues which are unexpected are resolved quickly. Such as last night when a portion of customers hosted in Los Angeles went down due to a brief network outage, which was resolved so quickly the sites were all back up before I booted my laptop up.

To give a concrete example, especially to underline how things are dealt with out of hours, a real scenario that happened last year. A customer found themselves in a situation where 60,000 user accounts had been deleted due to a wrongly configured add-on. That request came in at 8:02 PM. Initial reply was approx 20 minutes later. We gave some restore options to the customer who replied 20 minutes after that at which point work was immediately started and completed around 90 minutes later.

We have a lot of customers now and currently things are still very easy to manage. Naturally Cloud is now a crucial part of our business. Over time as the service matures and the number of customers increase, we might need to extend the team and perhaps make more concrete SLA commitments but for now everyone is happy and things get done in a more than reasonable amount of time.
 
1730748975813.webp


Generated by bing chat. Ignore the extra r
 
Amazon S3 is very cheap and easy to backup to via cPanel, I now have daily backups retained for 7 days and weekly backups retained for 4 weeks. Something I wish I could have done on CyPa but their instructions didn't work. Their support team also didn't believe me when I said all of their images in their docs were broken even though I showed screenshots and linked to a few embedded images that were still referencing a WordPress installation.

Their incompetence truly knows no bounds.
For my own site (forum and billing), each is on their own dedicated server, using jetbackup to do hourly incremental backups to 4 different offsite locations (2 x s3 compatible and 2 ssh storage).

Don’t forget to test your backups occasionally as well 😉
 
There is progress today.

Update:
All databases restored.

File copies
Ozzmodz.com domain - done
Snoggsite.com domain - in progress

Validation operational site:
Ozzmodz.com domain - pending
Snoggsite.com domain - pending

Make operational site live:
Ozzmodz.com domain - pending
Snoggsite.com domain - pending
 
because cpanel kept raising their prices it forced me to not use any panel. now i do everything by the command line. in fact i don't like panels anymore. i had more server issues with one then without.

saves 60 usd a month
Namecheap's hosting is great they use cpanel and for $100 you get your sites hosted for a year.
Some people will laugh but heck they'll be rushing on over after all of this.
 
So far looks like a month lost, but hopefully that will change.
If it's post history, and you have the latest (corrupt) database files, ChatGPT may be able to help. My DB got corrupted and the InnoDB .idb files were somewhat human readable. I used ChatGPT to export what it could (even some stuff that wasn't readable that helped as it was able to find the user that posted it when that got jumbled up) and recovered a lot more than I thought I would have (nothing)... about 40%, though it could've been upwards of 70%, but I ignored thread word games because they were so out of order and so many posts that I would have to manually enter them in... as opposed to just starting the forum game over.
 
Back
Top Bottom