Snapshotted backups which allow me to restore to any day in the past 3 months, any week in the past 6 months, or any month in the past 1.5 years; in a location 1000+ miles away; and yes, I have restored backups recently.
Oh, and my backups are encrypted, so someone with access to them won't be able to steal my data; my backup client code doesn't trust the backup server, so someone who compromises the backup server won't be able to leverage that into compromising the backed-up systems; and my backed-up systems don't have the necessary cryptographic keys stored on them to be able to delete data so even if someone compromises said systems, they won't be able to damage their backups.
Of all the people here you are the last one on the list of 'worrying whether he's got a backup' :)
I may become a tarsnap customer at some point in the near future, I'll let you know. My backup needs are outgrowing what I can store on individual backup servers (30T and counting) so I'll need something that scales better.
I'm trying to work out what the update rate is so I'll be able to figure out what it will cost me. The fact that you seem to be in for the long haul really helps.
Do you have some kind of policy in place about what will happen if/when tarsnap gets acquired ? Do you have a survival clause in your contract?
My backup needs are outgrowing what I can store on individual backup servers (30T and counting)
I'm not trying to discourage you from giving me money, but at $9k/month, Tarsnap is probably not the most cost-effective place to store 30 TB of data.
so I'll need something that scales better.
Erm, yes. Scaling. Please let me know before you start uploading 30 TB to Tarsnap... I'll want to watch a few things to make sure that they scale the way they're supposed to.
The fact that you seem to be in for the long haul really helps
Wow, someone who thinks that me not trying to build Tarsnap and sell it off as fast as possible is a good thing. After reading so many people espousing the virtues of "take lots of VC, get big fast, then exit", I was starting to think I was the only one.
In all seriousness, if someone offered me $10M for Tarsnap tomorrow, I'd probably sell; but given a choice, I'd much prefer to make money from Tarsnap being profitable in the long term, rather than by selling it.
Do you have some kind of policy in place about what will happen if/when tarsnap gets acquired ? Do you have a survival clause in your contract?
I'm not sure what you're asking here. There's no way that I would sell Tarsnap unless I was satisfied that it would continue to serve its users as well or better than they would be served otherwise -- I've invested a lot of my personal reputation in Tarsnap, and I have no intention of ruining that by turning around and screwing my customers.
> Wow, someone who thinks that me not trying to build Tarsnap and sell it off as fast as possible is a good thing.
That would pretty much be the deciding factor.
And I find it a real pity that not more people think like this.
I've had some business partners in the past of the 'get-rich-quick', 'retire-at-30' variety and I'm really not impressed with that mindset.
Especially not because most people that try that fail in spectacular ways, it's like gambling or playing the lottery.
I'm going a lot more slowly on my own projects, even though my investment profile is a lot more high risk than that.
$9K is a bit out of the budget though, I'm well under $300 for my current setup (which is 20T on one box and 5x3T on 5 others, but those could be condensed a bit).
7 comments
[ 4.2 ms ] story [ 20.4 ms ] threadOur backup server(s) are currently in pieces on the floor..... (long long story).
Coincidence? I think not :)
It's a pity you're not in the same country or I'd hop over to help.
Oh, and my backups are encrypted, so someone with access to them won't be able to steal my data; my backup client code doesn't trust the backup server, so someone who compromises the backup server won't be able to leverage that into compromising the backed-up systems; and my backed-up systems don't have the necessary cryptographic keys stored on them to be able to delete data so even if someone compromises said systems, they won't be able to damage their backups.
And yes, I am using Tarsnap. :-)
I may become a tarsnap customer at some point in the near future, I'll let you know. My backup needs are outgrowing what I can store on individual backup servers (30T and counting) so I'll need something that scales better.
I'm trying to work out what the update rate is so I'll be able to figure out what it will cost me. The fact that you seem to be in for the long haul really helps.
Do you have some kind of policy in place about what will happen if/when tarsnap gets acquired ? Do you have a survival clause in your contract?
I'm not trying to discourage you from giving me money, but at $9k/month, Tarsnap is probably not the most cost-effective place to store 30 TB of data.
so I'll need something that scales better.
Erm, yes. Scaling. Please let me know before you start uploading 30 TB to Tarsnap... I'll want to watch a few things to make sure that they scale the way they're supposed to.
The fact that you seem to be in for the long haul really helps
Wow, someone who thinks that me not trying to build Tarsnap and sell it off as fast as possible is a good thing. After reading so many people espousing the virtues of "take lots of VC, get big fast, then exit", I was starting to think I was the only one.
In all seriousness, if someone offered me $10M for Tarsnap tomorrow, I'd probably sell; but given a choice, I'd much prefer to make money from Tarsnap being profitable in the long term, rather than by selling it.
Do you have some kind of policy in place about what will happen if/when tarsnap gets acquired ? Do you have a survival clause in your contract?
I'm not sure what you're asking here. There's no way that I would sell Tarsnap unless I was satisfied that it would continue to serve its users as well or better than they would be served otherwise -- I've invested a lot of my personal reputation in Tarsnap, and I have no intention of ruining that by turning around and screwing my customers.
That would pretty much be the deciding factor.
And I find it a real pity that not more people think like this.
I've had some business partners in the past of the 'get-rich-quick', 'retire-at-30' variety and I'm really not impressed with that mindset.
Especially not because most people that try that fail in spectacular ways, it's like gambling or playing the lottery.
I'm going a lot more slowly on my own projects, even though my investment profile is a lot more high risk than that.
$9K is a bit out of the budget though, I'm well under $300 for my current setup (which is 20T on one box and 5x3T on 5 others, but those could be condensed a bit).
A > 20 fold increase is probably not very smart.