A lot of work remains to be done in regards of the flaky server model, but at this very moment 0 servers are down.

We are hoping situation remains so at least until we get more hardware in to apply the fixes.

A bunch more of servers got the "janky" 120mm fix today, and the process is now honed in. Takes only roughly 30minutes per server to do when processing in larger batches.

More hardware should arrive from Germany by end of the week or early next week, and we will begin the work from servers with lowest uptime (=crashiest) at that point. Some of this server model is working perfectly fine, and the chipset temps are almost as low as on those with the cooling fixes.

Monday, August 13, 2018

«Tillbaka