01/13/2006 Server outage thismorning at 6:55 AM MST to possibly 8 AM

Greetings, our co-lo is installing a new battery so the server will completely shut down at 6:55 am. No services will be available during this window. The server will start back up automatically once their maintenance is complete.

csolutions wrote:

The CSolutions support team will be performing the following Data Center
Maintenance:

Date: 1/11/2005

Time 7:00 AM MST ? 8:00 AM MST

Reason: UPS Battery Replacement

This maintenance should only affect your service for about an hour and we
would anticipate the process will actually take less than an hour.

We appreciate your patience with us while we strive to always improve our
service.

That's good to know, my power has been known to go out from time to time. It will be nice not having to worry about that anymore...

I also noticed that the main web page isn't querying the UT server proper, showing:

There are currently Bad queryString players playing Bad queryString on map Bad queryString

Maybe they changed your IP address on the UT server when they brought their systems back up? Also clicking the links from that shows an empty list with Bad queryString all over the place.

FRAGaLOT wrote:

I also noticed that the main web page isn't querying the UT server proper, showing:

There are currently Bad queryString players playing Bad queryString on map Bad queryString

Maybe they changed your IP address on the UT server when they brought their systems back up? Also clicking the links from that shows an empty list with Bad queryString all over the place.

Not a chance of an IP change. I'll look into why it's not querying when I get home tonight.

I find the outage fairly interesting. For those of you who are not aware, I used to work at CompuServe, where I was responsible for the UN*X OS (BSD/OS) used on over 1100 servers, as well as for the USENET News servers. Since any downtime was unacceptable, we took numerous steps to prevent any outage. We started with two separate mains into each of our three facilities, with these feeds originating from separate substations, to reduce the chances we switched over to diesel. Then on top of all of that, our UPS units were huge (about the size of two refrigerators), and we had countless dozens of them all over our datacenters. Short of doing wiring directly on the UPS, we never had to shutdown. Each UPS had multiple AC feeds, and the batteries for the UPSes were hot-swappable. All we had to do is to pull a breaker to make sure no power/current was going through the batteries, then pull out the wrench and go at it, turning on the breaker when we were done. Given that the batteries were essentially large car batteries, you can imagine the work it took to do this.

Right along with all of this is the fact that I will be swapping out the batteries in our UPS at the radio observatory where I am the IT head (volunteer group, but hey...). We have a rack mountable APC UPSes, the SU3000RM, and it has hot swappable batteries, so all I have to do is to remove a "pretty" door on the UPS, remove a few screws from the battery cover, pull the batteries out and put the new ones back, all while the systems remain under load. Indeed, all of the APC units from their smallest (BE350U) to their largest units I believe are hot-swappable, so the outage makes me wonder why it was even necessary. If I were you Dru, I would be querying them in more details about this...

TheDruidXpawX wrote:

Greetings, our co-lo is installing a new battery so the server will completely shut down at 6:55 am. No services will be available during this window. The server will start back up automatically once their maintenance is complete.

A new battery? What, like throwing a new CMOS batt. in the server blades or something?

Edit: I'm a moron

Spacey wrote:

Dru, I would be querying them in more details about this...

I dont think I will. We're getting such a deal from them that the 4 - 15 minute outages and one 2 hour outage over the past year are more than acceptable.

a player on my server complaint aswel, since 30dec 2 HDD of the server i'm running on crasht, and since i was on vacation i could not even fix it when it was back up. (4 jan)
but because of long dow time, even from my hosts side, somebody said i should look for another host.. but where am i supposed to find one thats only around ?50-60 for 3 months

most ive seen are around like.. ?80 a month.. often less or even more..

just thought i might share this because of the crash.. making yours look little short, total my server was off more then a week because i could not make it run again due to not being home.