New servers and a new IT team...

[NadMat]

Way to go!!!!!

Glad the move is done and going well

And soon I will be red

And I am an 'official' Texas Swamp Diver as of today also :luxhello:

And got first dive in on new BC :bang:

All in all, is shaping up to be one fine weekend

 

[Don Wray]

HOOORAH!!!! Give the boys (and girls) a big hand!!!

 

[chrisc]

It seems the site since the switch has been up and down more in teh past 2 weeks than ever before. Are these updates being applied?

 

[Doubles]

It seems the site since the switch has been up and down more in teh past 2 weeks than ever before. Are these updates being applied?

I disagree, before the move we were down every couple of hours. It has been a lot more stable in the past 2 weeks. Yesterday afternoon was the exception and may be totally unrelated

 

[Splitlip]

I disagree, before the move we were down every couple of hours. It has been a lot more stable in the past 2 weeks. Yesterday afternoon was the exception and may be totally unrelated

I don't know. I have been getting kicked off with pretty much the same frequency as before. I am able to get back on more quickly however. Yesterday was the exception. An hour or so before I got back on.

 

[The Chairman]

We are doing FAR better than before. We will encounter another jump in performance this weekend, that will probably eliminate our "every other day" outage. Sure beats the heck out of our "every other hour" outage.

 

[letterboy]

it sure does doc!!! Thanks for you tireless efforts to keep the ship upright

 

[mike_s]

We are doing FAR better than before. We will encounter another jump in performance this weekend, that will probably eliminate our "every other day" outage. Sure beats the heck out of our "every other hour" outage.

Pete, why not do "load balancing" between mutitple servers to "spread the load" and add fault redundancy?

You can still attach RAID-5 to all servers with great speed via SAN using something like Brocade fibre-channel switches.

That way if one server goes down, the board as whole doesn't die. The worst that could happen is that the last thread someone is reading or submitted doesn't complete. When they refresh, a next server should answer their request.

In theory, with the reduced load, you'd never have one go down and you could take one server out of the rotation for maintenance purposes without loss of service.

Nee more horsepower as scubaboard grows later on? just add another server to the cluser.

simple eh? [sarcasm implied as I know your hands are already full]

 

[DandyDon]

it sure does doc!!! Thanks for you tireless efforts to keep the ship upright

I guess it's kinda rough when we overload the ferry, and it has to be rebuilt during crossings. Couldn't be in better hands, tho...

Pete, why not do "load balancing" between mutitple servers to "spread the load" and add fault redundancy?

You can still attach RAID-5 to all servers with great speed via SAN using something like Brocade fibre-channel switches.

That way if one server goes down, the board as whole doesn't die. The worst that could happen is that the last thread someone is reading or submitted doesn't complete. When they refresh, a next server should answer their request.

In theory, with the reduced load, you'd never have one go down and you could take one server out of the rotation for maintenance purposes without loss of service.

Nee more horsepower as scubaboard grows later on? just add another server to the cluser.

simple eh? [sarcasm implied as I know your hands are already full]

I'm sure that's code that was not meant for us mortals to understand, but for good reason was not sent in a PM...?

 

[mikerault]

Yes but RAID 10 would give better upload performance since it performs better for read and write activity. Make sure you balance IO needs with capacity. Just because you can get a terabyte disk doesn't mean you should...I've seen too many sites where the SA bought a few large disks instead a bunch of smaller ones and thus introduced an IO bottleneck. Each disk is only able to produce about 100-150 IO/second so plan accordingly!

The previous message was just to completely confuse DandyDon and should not be implied as any criticism of the board IT team...

Mike