Search results

I need some feedback from everyone doing anycast on OpenVZ :) Currently we're trying to make both venet0 & ha0 work together at the same time. Originally in our test beds this worked perfectly but I guess I didn't think about testing tons of different IP addresses all at once. This is proving...

You do otherwise you'll get dead zones. Having double up zones would get really messy and require an excessive amount of logic checks & wasteful IP usage. Francisco

  This is a soft launch so the few things like that can be worked out :D Put in a ticket about it and I'll fix it in a few hours once I get a bit more rest. I already noted an issue with OpenVZ that I've already got ready for Aldryic to fix up. 198.251.86.1 Francisco

Hello everyone, After much work, Anycast IP addresses & failover IP addresses are operational! Customers with service in all 3 of our locations can assign themselves anycast IP addresses and then enable them to be bound on each virtual server! We've also brought failover IP support, making it...

It was possible for a good while. The wide spread exploit that was sued was them passing $_GET['id'] or something like that right to a SETUID binary on the master. Francisco

Do you need a hard drink now? :P Francisco

Incorrect my good sir. The problem is that on the SolusVM panel side, they actually pass the users root password to CLI via something like the comm.php file takes into consideration if the VPS is hosted locally (master + users OpenVZ) or remotely. From there it either calls a SETUID root binary...

You'll have to do your own firewalls for now. Over time we'll likely get a managed firewall option in place globally. Francisco

It'll be an option in stallion once it's all committed :) I'm just doing up some loose ends in the code. Francisco

Backups are a pain in the ass. 'nuff said on that one. > :( Francisco

I should add, pooling is still about a week out, probably less. With that being said, we're going to have BW tracking disabled on anycast IP's for the time being. This doesn't mean you can rip gbit/sec on it and not worry, we're still going to keep a close eye on usage :-P I just need to do...

Haha :) I've been getting the node side changes in place. I tripped up while doing some of the bridges causing a couple minutes of network downtime on some nodes but for the most part we're almost all done on that front. Aldryic will commit the nodeside code itself in the morning and I'll...

Cost should never be an excuse for not being able to meet expectations of a product. If you're forced to deliver a subpar product just because you decided to compete on price, then you need to really re-evaluate your business model. Up until mid last month everything was kosher so it isn't a...

You're acting like it's 50 things and the box is catching fire. There's 2 issues, 1 is due to an update pushed by one of the platforms we use (which means others will suffer, or already are suffering, similar) the other is the drives are finally getting pushed hard enough for the hardware to...

User triggered ones, etc. You'd be surprised how many files some people have. We're just missing some processors, so I'm hoping to have those in hand by Tuesday/Wednesday. Once those ar there I can actually get the maintenance done. The initial debugging I did helped greatly but it's not going...

Hi! Thanks for the interest! We're just waiting on the last of the gear to arrive. Karen's finally back in town so I can get the missing parts ordered. Francisco

YOU BASTARDS. Francisco

Anycast is being rolled out Monday :) Most of the code will be committed on Sunday but the actual switch flipping will be on Monday. Francisco

M'ladies, Please see how your page load speeds are. The big maintenance wasn't performed but after a LOT of debugging I think I've figured out what was going on. From my own tests things are loading much much much better and iowait is good considering the work load in place (< 1.5% iowait...