RAMWolff - 03 June 2012 06:05 PM
Apparently they did some testing but as it seems not much at all before going live.
To be fair to the DAZ team - it is absolutely impossible to stress-test a test environment. A lot of the issues we’re seeing are caching and inter-server access issues. There are VERY few institutions that can afford to build a test environment that matches the production environment in all ways - paths to disk, network layout, cpu count, server count, memory, and all such similar items. And even if you get that - you can’t get a thousand testers to beat on it for two days.
I fault the team for such things as not having working links to the zen-desk system and the mantis web problem reporting system - those should both be trivial and should have been in -place before the cut. That was unprofessional and unacceptable.
I also fault management for not communicating. Back when I was working and had my back against the wall, my boss (tech services manager) was on-site with me for two reasons. One was moral support - I knew he had my back. The other was to keep the animals at bay. He would send out (in our case, hourly) status emails to the corporation. The first few hours usually amounted to short “no change, we’re still analyzing the problem” messages. But they satisfied the rest of the company by showing that someone was trying to do something.
I would like to see a new sticky from the manager of the web team updated daily, about an hour before the end of the day. It should list the top six items being worked on, and for each a status of analyzing, coding, testing fix, or fixed. For each item listed as ‘fixed’, add another to the list. (Seems to me I proposed this the last time the store cutover fell on it’s face).
Just let us know WHAT is being worked on and WHEN you think it’s been fixed. PLEASE.