I can remember in the "old" days that every time the site was down for planned or even unplanned maintenance, we would find a message posted to that effect either before or after the fact.

Would it be too much to ask to have some kind of group email sent every time there is an S2KI.COM outage? Even if it's only after the site has come back, just letting us know it was down? Its become quite frustrating recently to have to ping S2KI every time my browser times out when I'm trying to connect.

Just my two cents

 

Actually, I wouldn't mind seeing a post to this forum after the site comes back from a significant down time. All that is needed is a message of the form:

"The s2ki site was off-line from <date/time> to <date/time> due to <cause>"

Just my own two cents. A warm-and-fuzzy feeling that someone is behind the curtain to make all things happen is appreciated.

 

yeah... I agree with both of you guys. It's ironic - this is the only site I pay to support, yet it is the only one that seems to go down (other than routine maintenance) fairly often

 

Good to hear that it isn't my T-1 connection.... It seems like lately the site goes down a few times a day...

A courtesy message would be cool!

 

We are trying very hard to fight a hardware problem. A new server has been ordered today and we are in the proccess of setting up a Load Balllencer to ballence s2ki over all 3 of our servers.

The current server hosting s2ki is locking up on us quite often right now and we have no way (presently) to redirect traffic when it goes out.

We are working very hard to resolve the reliabilty issues of the past couple weeks.

Your continued support is appreciated.

-Rylan

 

I feel for you guys... I had a Dell server giving me no end of problems a few years back. It was running as an NT 4 BDC/Exchange Server at the time and everything pointed to a hardware problem. Dell swapped out system memory, motherboards, CPUs, RAID cards, etc, etc. Nothing improved and it happened intermittently. They eventually pointed us to Microsoft and claimed it was a software problem.

So, we bit the bullet, pulled out the credit card, and called Microsoft. Weeks later, MS tells us that "this is the cleanest running Exchange server we have ever seen. It's a hardware problem." Got the execs onto the phone with the Dell sales reps and next thing you know they are giving us a brand new server to keep free of charge and sending a tech onsite to diagnose and fix the old server, which we'd also get to keep. Took a while, though.

Good luck and hopefully things will settle down for you once you get that new server in place!

 

Dear Members,

The main web server for s2ki.com has been having random and mysterious hardware issues for the past month. The system will lockup and crash unexpectedly many times a day and when we get it back online 10 billion people hit it causing it to crash again. Those of you with experience in running this type of equipment will relate to the utter frustration we feel.

Unfortunately we get no advance notice of this happening. If we did, we would surely let you all know in advance. However, what actually happens is that our cell phones ring and we get an SMS message that says "server is down". It's a bit hard to let everyone know of the problem at this point.

For the past 2 days Rylan and I have been working to resolve this issue for good. We have purchased 2 new Pentium 4 servers which will arrive and be installed tomorrow. We have also configured a load balancer system which will spread the load across all three of the servers. It will also take a server out of the loop should it stop responding, leaving at least 2 to continue serving the site.

Not only will these changes nearly triple our capacity but it will also mean that the failure of a single server will not disrupt the site as a whole.

The complete package will be installed and running by the end of this weekend. Until then we've had to reign back the capacity of the server we do have running resulting is very slow response from the site. You will also notice a few things breaking from time to time as we are working on the servers and making adjustments.

Please be patient, we are working full out to get things sorted in the shortest time possible.

Thanks,
The S2Ki Team,
cthree, krazik and HondaGal

 

We have had 2 web servers online since yesterday and the load balancer is working great! A web server in the cluster can crash or be taken offline without the slightest hint of an issue.

We have 2 more servers prepped and ready to go online tomorrow bringing the total to 4 in the load balanced cluster. Our capacity requires about 1.5 of these servers to run at maximum speed. In other words, we can lose half of them and you wouldn't even notice!

The S2Ki network now consists of 6 servers with 9 P3 and P4 CPUs and a total of 10GB of RAM. That's a lot of hardware!

We just dropped $5K this month on insurance and hardware. It would be cool if some of the regulars who haven't taken out a membership yet would do so soon.

Thanks for hanging in there while we make this transition. Our new configuration should provide ample capacity and reliability for at least the next year. We can now do maintenance and expand capacity further without any interruption to service.

Thanks again

 

Excellent! You guys rock! Happy Turkey Day-
Duncan

 

We now have 4 web servers online and running in the cluster.