A DNS service is a simple server/cluster with a configuration. It’s easy. It’s a well-defined protocol. Yet M$’s cloud came crashing down because of a failure of DNS for hours. Was there no one awake to reboot the servers? I have run BIND 9 many times on my LAN and never had it fail. Maybe M$ should try Debian GNU/Linux instead of its own dogfood. I see three current bug reports on Debian’s bug tracking system: one for a fix when multiple instances are run, one when a misconfiguration prevents killing the process and one wishing for the newest version of Berkely DB to be used. That is, no one has any issues with the thing doing its job in the normal way. If M$ cannot use its software to run its systems flawlessly, why should you rely on them to run your systems? Is a discount going to make up for your loss of productivity?
We’ve been here a while – search
Tags
- 2015 - Year of the GNU/Linux Desktop
- Acer
- adoption
- android
- Apple
- ARM
- China
- cloud
- Debian
- Dell
- desktop
- education
- firearms
- FLOSS
- food
- garden
- GNU/Linux
- government
- horticulture
- HP
- hunting
- Intel
- LibreOffice
- Linux
- market share
- migration
- politics
- renewable energy
- Samsung
- security
- server
- small cheap computers
- smart phone
- software-patents
- Solo EV
- tablets
- that other OS
- thin client
- thin clients
- trolls
- Trumpism
- Ubuntu
- uptake
- weather
My Mission
My observations and opinions about IT are based on 40 years of use in science and technology and lately, in education. I like IT that is fast, cost-effective and reliable. My first use of GNU/Linux in 2001 was so remarkably better than what I had been using, I feel it is important work to share GNU/Linux with the world. Now that I'm retired I still use GNU/Linux on every computer in my home except the smartphones which run Android/Linux.Lately, I've been giving lots of thought to the world I inherited and which I will leave to my descendants. I'm planting grass, trees, flowers and vegetables in my large lot and I've ordered a Solo EV. I plan to charge my Solo by means of a tracking solar array. Life is good if you have a purpose. I do.
Outlinks
Posts
JairJy issue is google outage was something serous. Storage had gone off line in a way google had never seen before.
DNS failures are something you are meant to have automatic tests being run against on large clusters. So even if a human is not there back up servers and other cures should have been done by software.
Reason no matter how good the dns server is the hardware it is on can develop issues. Historically this has happened a lot.
Quality of service standards if you dig threw the ISO standards for it. Yes DNS is one of the things to get one of the ISO numbers you are ment to have.
Down because of why. DNS is a very bad answer. Now if DNS does not have automated monitoring what else is not monitored? Maybe not smart state of hard drives? Reason you know they are not working by the official standards to run a data center.
So, Microsoft is dommed because Hotmail where down by 3 hours. Ok. Like GMail hasn’t never have been down before.
I would argue, that in a corporate environment where you have more than one DNS server anyway, to make it heterogeneous.
Last year, we encountered an issue that affected windows dns servers – a default configuration that disallowed an old and common type of lookup. We skirted the issue with a registry hack, since you couldn’t configure it using any of the GUI tools.
My suggestion to the boss was to switch the secondary (if not the primary) DNS over to linux, (and leave the other one as MS) so that we could always have at least one server not affected by the latest shenanigans.
suggestion went no where. :/