![]() ![]() The botched spectre and meltdown zero day patches that MS pushed out all at once, in spite of having had the patches for a few weeks before the exploits were announced, interfered with dozens of virtual machines that were kept running through the patches, and caused provisioning problems for the machines that automatically shut down at night while they fought to get resources as millions of virtual machines and hosts reboot all at the same time. I have experienced two outages now where MS over-provisioned the capacity of their datacenter, leading to 2 and 4 core virtual machines not being able to boot without making the attempt a dozen or more times. This can’t be chalked up to “stuff happens.” “Stuff happens,” on rare occasions, like what happened in the AWS virginia datacenter a few years ago. This is above and beyond what I would expect, but that is the competition.Ģ. ![]() On multiple occasions, I had come into an AWS environment entirely blind to work a problem and not only did their support deliver, but also took the time to help me navigate an environment I had never seen before, show me how to use their available diagnostic tools and send me other tertiary documentation that may be of interest. I also wish I could say that support blowing their response time SLA, sometimes on a magnitude of days, was rare and almost unheard of, but I can’t. I wish I could say I had fewer than 5 total support tickets that ran for more than a month across the 3 clients I regularly work with Azure on, but I can’t. expensive support that is almost entirely useless. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |