I’m wondering if WPMU DEV can shift toward pro-active reporting of hosting versus re-active.
In the image posted to the first comment of this thread we see fluctuating transaction time reported by Uptime. This is on a site that hasn’t even been announced yet.
I’m hoping Dev will aggregate performance numbers from sites, note Hits/TransactionTime as a metric, and publish some information about what happened or what will be done about low numbers. For example, it’s easy to recognize 1hit/1200ms is better than 1hit/2400ms. And 100hits/2500ms would be better than 100hits/4000ms. When averages are detected that are “worthy of concern” (numbers to be determined), a notice should be triggered to Dev staff that something is up with the infrastructure. If the issue continues, I think a pro-active note to affected clients would be well-appreciated “Hey, we noticed the systems are running a little slow, don’t worry, we’re on it!” (add new Hero icon here…:wink:.
Note: I’m not saying the performance shown in the image is bad, just wondering who is monitoring it in case it does get bad.
I’m also wondering what the new protocol is when a hosted site is reported as being down by Uptime. I mean, who do we blame now? LOL I got notices that a hosted site was down for 3-5 minutes last night (depending on the report viewed). OK, what happened? What is Dev doing about it?
Also please note, I got four emails in a period of three minutes that this one site is down. I can imagine the flood of emails that someone would get if every one of their sites was down to some infrastructure failure – perhaps enough to get WPMU DEV flagged for spam by some email servers. Please try to tune that notification processor.
Thanks!