We are moving most of our servers to a colocation facility with better cooling, power, and network, which should improve uptime in general. This procedure will start on Monday, April 1st at 08:00 (PDT) and should take 3 days to complete, during which our entire project will be unavailable.

---

-- BOINC/SETI@home network/web/science/development person
-- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude

Thanks for the update Matt, Good Luck.

Claggy

Monday, April 1st

I hope that's a more auspicious day than it sounds :P

Best of luck, and thanks for the heads-up.

Great news! I hope all goes well with the planning and execution.

---

Hi Matt,

I understand that the BOINC server is going to be moved out of the SSL to some place down the hill as well. Is that happening around the same time?

---

Down the hill and rise the limit, that´s a good ideia.

Good luck!

---

Glad to hear the server move is going ahead.
Look to catch them on the other side.

I hope it's not a April Fish....

---

Sounds great! Hope it's someplace convenient for you guys.

Better network too?

Thanks for the update Matt.
Good news this relocation :)
Good luck to all the roster for the move !

---

Will you or others have physical access over weekends to kick kit if it becomes necessary like you have now?

As Matt said when he brought up this possible move a while ago, one major advantage of it is that it will be among other servers in a place where someone is always on duty to kick them right away. Not necessarily one of the Seti project staff, but someone who knows what he's doing with computers of this kind.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

Thanks for the news, Matt.

Hope all goes as planned and moving day does not bring too many surprises.
Might have to print out rebooting order instructions and such for the colo staff, eh?

Will we still have a trusty Cricket graph to monitor traffic from the new location? Or will it be buried in with the other traffic?

I better make sure my furnace is still ready to rock, 3 days without Seti and this spring has not been very warm so far this year...LOL. Must keep the kitties warm.

---

"Time is simply the mechanism that keeps everything from happening all at once."

Thanks for the news, Matt.

Hope all goes as planned and moving day does not bring too many surprises.
Might have to print out rebooting order instructions and such for the colo staff, eh?

Will we still have a trusty Cricket graph to monitor traffic from the new location? Or will it be buried in with the other traffic?

I better make sure my furnace is still ready to rock, 3 days without Seti and this spring has not been very warm so far this year...LOL. Must keep the kitties warm.

Let Einstein have a burst. Just put it on NNT after one work request.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

And on a related note, will the servers be able to keep their old IP addresses in the new facility?

I know we should be using DNS and it should all 'just work' - but human beings have a habit of fiddling with things, and then forgetting to 'un-fiddle' with them again. We recently had a couple of cases of people trying (and failing) to download work from .18, six months after you retired it - the legacy of some long-forgotten episode when whichever elderly server had .13 at the time was throwing a hissy fit.

Let Einstein have a burst. Just put it on NNT after one work request.

I just set it to a resource share of zero. Then it only grabs work units when SAH doesn't have anything to do for either CPU or GPU. Sometimes E@H is a little too generous on the CPU side but the GPU side units will only download one at a time like this, and never if SAH has one ready.

---

Great news!

---

Will this move improve system responsiveness? No endless download attempts for AP and standard 6.03 WU's anymore?
I understood that the overall bandwidth avail is OK, however, the servers sending out the WU's are overwhelmed constantly.

GS
Germany

Will this move improve system responsiveness? No endless download attempts for AP and standard 6.03 WU's anymore?
I understood that the overall bandwidth avail is OK, however, the servers sending out the WU's are overwhelmed constantly.

GS
Germany

If I remember correctly the original announcement about the possibility of making this move, it would allow for more bandwidth for downloads (I could be wrong). The problem is partly bandwidth, partly the servers.

Also, you should go to the Number Crunching forum and read the TCP fix thread. It will help with your downloads.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

If I remember correctly the original announcement about the possibility of making this move, it would allow for more bandwidth for downloads (I could be wrong).

I hope so... there ends up being even more traffic if there isn't enough bandwidth due to all the retries (ie uploads don't resume.) That would be a great improvement and well worth the outage.

---

Yea, Matt ! ! !

I am sooooo looking forward this.

Thanks for doing whatever it was you had to do to get this approved.

I wish you the best of luck seeing it through to completion with a minimum of surprises.

It would be "fun" to know what's being moved, exactly.