I think that was separate, actually. Google also had an outage caused by them upgrading and not testing enough to notice that some config or command line parameter had been renamed, and they were hounding the bugtracker as if they had a 30 second SLA. Somehow them having an outage and not having the presence of mind to roll back the upgrade was supposed to be an emergency for an unpaid volunteer.
Service Level Agreement. It’s when you have a contract requiring a certain level of service, like minimum 99.999% uptime or (in my example) a maximum 30 second response time on support requests.
This is something you typically get from an enterprise vendor (although 30 seconds is exaggerated) in return for paying them a lot of money. It’s not something you can expect for free from volunteers.
I think that was separate, actually. Google also had an outage caused by them upgrading and not testing enough to notice that some config or command line parameter had been renamed, and they were hounding the bugtracker as if they had a 30 second SLA. Somehow them having an outage and not having the presence of mind to roll back the upgrade was supposed to be an emergency for an unpaid volunteer.
SLA?
Service Level Agreement. It’s when you have a contract requiring a certain level of service, like minimum 99.999% uptime or (in my example) a maximum 30 second response time on support requests.
This is something you typically get from an enterprise vendor (although 30 seconds is exaggerated) in return for paying them a lot of money. It’s not something you can expect for free from volunteers.
Whoa, that does sound expensive!