Super interesting post. Would love to read more detail about their backup and restore infrastructure.
If Tom and/or Shlomi are reading this: you mention taking multiple logical backups per day. What benefit does this bring versus just having one per day and doing a point-in-time restore using binlogs? Is this just a tradeoff between time taken for a restore and storage you're willing to dedicate to backups?
We use it because it works well for us. We've put a lot of work into making MySQL scale for us to the point where it's a very well supported system and one of the main choices for a lot of storage decisions.
We even use MySQL as a queue for Facebook Messenger. More details about this:
In addition to what evanelias said, a logical dump also means we can load it into a MySQL instance running a different storage engine as well. In our case, it allows us to take a mysqldump from an InnoDB instance and load it into a MyRocks instance if we wish.
There's a few different ways to verify, a few of them involving stopping replication, like you pointed out. These can also sometimes be quite expensive, so depending on the type of verification required, the verification method can be tuned.
> Do you diff against the same base, or create an incremental chain? How many diffs do you take in between recapturing a full image? At $DAYJOB we always take full backups into a fast in-house deduplicating store.
We always diff against the same base and have 5 days in between subsequent full dumps. The number of days just comes from a trade off between space occupied by the backups and time it takes to generate them.
> Is there no better way to handling this than polling?
There's definitely different ways to approach this, we find polling works well for us. We also use the same database for crash recovery, so doing the assignments through it serves both purposes.
> Presumably you can only get this parallelism by disabling FK integrity. Is it re-enabled in the following VERIFY stage?
I'm not sure what you mean by parallelism through disabling FK integrity. Splitting the backup into its tables means we can restore a subset of tables instead of the entire backup. This allows us to load individual tables concurrently, but also not have to wait to load a massive database if all we need is a few small tables.
Those addresses are websocket servers that seem to publish wikipedia changes. The app is just subscribing to the changes by opening a websocket connection to the appropriate server.
> If you have not been explicitly informed by us in a separate communication that we detected suspicious activity involving your Slack account, we are very confident that there was no unauthorized access to any of your team data (such as messages or files)
Out of curiosity, what do you use as an alternative to Word if you do have to open something in a .doc/.docx format? Surely just saying "I don't have Word because I'm a developer" can't work in every situation?
Instead of pausing the feed, maybe a simple 'Load X new tweets' banner at the top of the feed would do a better job? Twitter actually does exactly this when loading new tweets on the feed.
The keypair AWS generates can only be downloaded once, at the time of instance creation. Beyond that, they expect you to be in possession of the keypair when launching another instance that uses the same keypair. If you happen to lose the file, you're basically out of luck.
So to directly address your concern, you can't download the keypair at any point in time, it's just a one time thing. To me that seems much more secure than emailing out a root password and enabling password authentication by default.
I am a Digital Ocean customer and the only password they've ever emailed me is the root password for the server I just bought. Arguably this isn't as safe as AWS' process of making you download a kaypair and only letting you login with that. However, VPS owners should get in the habit of logging on to any server they buy and immediately disabling password auth and root login via SSH, which helps negate the root password being sent over email issue to a certain extent.
Google would be defined as google.com (or the appropriate geo-specific domain) since this map is specifically for websites, not companies owning the most popular products .
As for what would happen if they did group by company? I'm guessing that Google would be #1 pretty much everywhere, save maybe Russia and China.
If Tom and/or Shlomi are reading this: you mention taking multiple logical backups per day. What benefit does this bring versus just having one per day and doing a point-in-time restore using binlogs? Is this just a tradeoff between time taken for a restore and storage you're willing to dedicate to backups?
Disclaimer: I work on Facebook's MySQL backup and restore system (https://code.facebook.com/posts/1007323976059780/continuous-...)