Out of the Frying Pan (Feb 17 2010) |
![]() |
| log in |
Message boards : Technical News : Out of the Frying Pan (Feb 17 2010)
Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next
| Author | Message |
|---|---|
|
I'd just like to say "thank you" to Matt and all at SETI, I know they are just as keen as the most rabid poster on this thread to get everything back up and running. I would hate too have to go through all the tests and checks they will to ensure that the A/C fail do not do to much damage. | |
| ID: 971612 · | |
|
However annoying it (still) is, the most annoying I think is there's no messages on the front page of what's going on. | |
| ID: 971622 · | |
|
What bothers me most about the outages is that Matt and the other 3 SETI staff do not seem to understand that we contributors are spending our money (electricity, machine wear and tear, etc.) to support SETI even if we do not donate cash. Keeping contributors informed is a must for any non-profit. I have been running SETI for a long time, and the stability of the crunching has deteriorated since BONIC was utilized. The recent outages are the latest in a string of problems and issues that have never been properly resolved. I understand the "lack of funds", but SETI management needs to address the problems because the science is loosing out to infastructure and "slap a patch on it" system management/maintenance. Perhaps it would be best to stop processing data and really fix the problems! As an engineering manager who has had to deal with this type of IT instability, shuting SETI down and fixing the problem may be the only way to resolve these continuing problems. | |
| ID: 971624 · | |
|
I am running SETI and 4 other projects on a small cache (0.25 days). When a WU is finished it is uploaded and I get a new one. If one of the projects has a blackout, the other four are running. In the worst case I have 2 projects running and three idling, so the 2 cores of my Opteron 1210 running Linux are always busy 24/7.AQUA@home alone grabs both cores since it is multithreading but it switches every hour, like all others, since they made it checkpointing according to our requests. | |
| ID: 971638 · | |
What bothers me most about the outages is that Matt and the other 3 SETI staff do not seem to understand that we contributors are spending our money (electricity, machine wear and tear, etc.) to support SETI even if we do not donate cash. Keeping contributors informed is a must for any non-profit. I have been running SETI for a long time, and the stability of the crunching has deteriorated since BONIC was utilized. The recent outages are the latest in a string of problems and issues that have never been properly resolved. I understand the "lack of funds", but SETI management needs to address the problems because the science is loosing out to infastructure and "slap a patch on it" system management/maintenance. Perhaps it would be best to stop processing data and really fix the problems! As an engineering manager who has had to deal with this type of IT instability, shuting SETI down and fixing the problem may be the only way to resolve these continuing problems. How can you say that they don't understand that their contributors spend money on electricity? They know they are harnessing a waste product from their users. Taking that one step further, the waste in electricity would be minimal if people only donated the actual waste, and didn't "create" waste by buying extra computers/parts just to run SETI. Anything above and beyond what you'd normally have in your system for doing your daily tasks is more than they asked for, so any waste when there's downtime is really on the user, not on the Project Admins. In addition to that, the project has stated numerous times that there will moments of trouble or no work, and if you do not wish to "waste" your electric bill, to donate your spare CPU cycles to a back-up project. If you choose not to do this, the "wasted" electric bill can only be faulted on the user. With only 4 part-time people in the lab, would you rather they keep posting information to us, or would you rather they just fix the issue while we see the end result? We all know there's a problem due to the flurry of complaints on the subject. I'd rather see them focus on the problem and not on spending precious time updating us. ____________ | |
| ID: 971639 · | |
|
Patience Daniel- san. Posts crying for an update make me laugh. When its up, it will be up. Do like I did and pick another project to volunteer for or donate a buck or 2 to try to help out( did both mind). Posting your opinion about what they should or shouldnt do to inform you is silly. | |
| ID: 971640 · | |
|
It is obvious that Matt could spend ten minute break a day to keep the contributors up to date. Both my posts took well under 10 minutes to submit. Keeping the contributors informed does not slow down the process of "fixing" the problems SETI is encountering. As I said, shut it down and fix it. Tell the contributors that the system will be down and get on with it. Give the contributors an update (5 mins?) daily as to the best guess when the system will be up. This does not keep the staff from fixing the system. | |
| ID: 971647 · | |
|
I find it hard to argue with a request for more information. I'm a big fan of communication, something that most companies fall apart on internally. | |
| ID: 971649 · | |
|
Patience. You must learn patience. | |
| ID: 971652 · | |
|
Hmmm, let's see, an update? Well, for the last two hours they were chasing cat-5 from router A to switch B then for the next two they checked the connections on all the servers, then they sat down and scratched their a... uhh heads and started all over again. | |
| ID: 971654 · | |
|
I have been more than patient, as this is my 7th post since 1999! How long do the contributors need to wait for SETI management (not Matt) to fix the issues? | |
| ID: 971655 · | |
|
First I want to say...SETI is awesome. I went down Puerto Rico 2 years ago, and after doing the obligatory "let's lie on the beach and take in the culture" for my wife, I convinced her to go to Arecibo to see the radio telescope. It was incredible and an inspiration. A giant satellite dish in a mountain. | |
| ID: 971658 · | |
A lot of money has been spent to develop recievers based upon Allen's money. My understanding is that Allen's money goes to another project, and that project does not provide any support or data to this one. SETI funding has been discussed at length on the number crunching forum, it comes mostly from UC Berkely and volunteer donations. If you want to see what a difference money makes, try using World Community Grid for a bit. Mega bucks from IBM keeps things up and running. That's fine, if you don't mind Big Corporations deciding what is worth crunching and what isn't. For me, I will continue to run a mix: taking advantage of IBM's big bucks and the SETI staff's hard work and dedication. ____________ | |
| ID: 971660 · | |
|
So still no conformation that anyone at seti even knows there is a problem....Bummer. | |
| ID: 971667 · | |
|
It is 7:34 Am in Berkley, No response is what is expected. Give Matt the benefit of coming to work at a normal hourt. | |
| ID: 971669 · | |
Looks like the problem is solved for now for most people. Has there been a post-mortem and if so what was the result? Obviously I was wrong. My results are being uploaded after a struggle with back-offs but the reporting step is just rejected each time. So the ready-to-report count just keeps growing. If this is true for everyone, then seti should expect a huge, johnny-on-the-spot flush when they fix their problem. I bet we will experience the ripple for days. | |
| ID: 971671 · | |
|
| |
| ID: 971672 · | |
At least not from Europe/Germany (or my home.. ;-). It' your home... and everything else except for my laptop, which finally uploaded enough results to request new work units. Download seems to be no problem. My other two machines couldn't upload anything since don't know when. ____________ . | |
| ID: 971684 · | |
Give Matt the benefit of coming to work at a normal hour AFAIK Matt is always free on Fridays. He works Sunday - Thursday for Seti, if that's still valid. ____________ Jord - BOINC FAQ Service - BOINC User Wiki Real is just a matter of perception. | |
| ID: 971687 · | |
... SETI has had outages multiple times in the past. "All this has happened before and it will happen again." If you really feel upset by it, let it inspire you to donate cash as well as time to them. no thank you, i would be upset if seti went away, but i would get over it. Most of the people that post in the forums are addicted; science, competition, or just seeing what that new rig will do. It would free up my time. I guit last summer, and will quit this summer, perhaps seti will still be here next fall. i just ask people don't do climate perdiction... climate change...!!! ____________ | |
| ID: 971689 · | |
Message boards : Technical News : Out of the Frying Pan (Feb 17 2010)
| Copyright © 2013 University of California |