January 28th Incident Report

Re:January 28th Incident Report by dice on Wednesday February 03, 2016 @06:53PM (#11030919)

>Redis is really reliable (in my experience at least)

Redis has been demonstrated[0][1] to lose data under network partitions. This is particularly concerning when discussing the type of partial failure that GitHub reported.

0: https://aphyr.com/posts/283-jepsen-redis

1: https://aphyr.com/posts/307-jepsen-redis-redux

[ Parent | on HN ]

Re:January 28th Incident Report by nemothekid on Wednesday February 03, 2016 @08:08PM (#11031259)

I meant Redis is really reliable as a single instance. If you reread my post I mentioned that Redis doesn't have a decent HA solution.

[ Parent | on HN ]

Re:January 28th Incident Report by obituary_latte on Wednesday February 03, 2016 @07:00PM (#11030954)

Not sure how your comment refutes the contention of reliability. Seems to me to be more a condemnation of failures that do happen (which is of course worthy of concern, but irrelevant in a conversation about stability).

[ Parent | on HN ]

Re:January 28th Incident Report by dice on Wednesday February 03, 2016 @07:34PM (#11031103)

I am using "reliability" in the sense of RAS[0]. An HA datastore which erroneously ACK's writes has lowered reliability, as there are known cases where it gives incorrect outputs.

0: https://en.wikipedia.org/wiki/Reliability,_availability_and_...

[ Parent | on HN ]

Re:January 28th Incident Report by obituary_latte on Wednesday February 03, 2016 @08:00PM (#11031221)

I see. Yeah, more concerning there being errors not based on an "event". Thanks for clarifying; sorry for any confusion.

[ Parent | on HN ]