that's it

Probably not until Dave gets back on site at the end of the weekend. He took a weekend off, the servers went boom. On site folks did something of a process restart, but the validator and new work are not flowing.

---

I wonder if he can teach some1 restart the server and minor budfix.

---

Indeed, a fair question and resonable expectation.

I wonder if he can teach some1 restart the server and minor budfix.

---

So, your whole operation is dependent on one person.Very dangerous!

So, your whole operation is dependent on one person.Very dangerous!

And 1 server.

---

Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

Perhaps you don't realize just how resource thin most BOINC projects are.

This weekend has become a dire one for folks in ATI GPU land. MW offline starting Friday, Collatze went offline 3 hours ago, and now, the last ATI GPU project in BOINC land standing (DNetC) just crashed out about an hour ago.

All of these projects are 'low resource' projects (MW is somewhat better resourced than the others). For Collatz, the primary issue is not so much the availability of their primary -- the project for him seems to be a true labor of love to which he is very dedicated -- rather, that project (like all of us in terms of connectivity) is dependent on the performance and reliability (or lack of reliability) of his ISP. That ISP seems to suffer connectivity issues of one form or another on an all too frequent basis. Some of those outages seem to be 'unacknowledged' short duration outages of 30 minutes to two hours, but others, like this one go for many hours (in some cases a full day). I am so glad that I've not had to cope with that sort of ISP underperformance over the years.

DNetC is based in Poland, it is a new project which sometimes has server issues of various forms, and sometimes has connectivity issues. It is also a thin resource project and one which tends to 'undercommunicate' status, though there is a clear high level of activity on the application development side.

The thing is, at this particular moment, we are in a zero for three mode.

So, your whole operation is dependent on one person.Very dangerous!

---

Over to Stanford ;)

As of 5 Jun 2010 22:26:51 UTC, Results ready to send = 1. I wonder who got it?

---

Ok I will fess up.. It was me who got the wu.. :)

DD,

---

I think, i did.
6/6/2010 12.44am.

Both Dnetc and Collatz returned later in the day -- Dnetc required some further fixes (or reboots), for Collatz, once the ISP returned them to life all was good.

The thing is, at this particular moment, we are in a zero for three mode.

---

Thats the main reason I've set my cache to 10 days.

Does'nt make a difference with MW, but with collatz and DNET, it does make me survive if all 3 projects go down for a few days!

Unfortunately some projects do not allow a large buffer.
I think, there should be an allowance of at least 24h.
As performance of computers/graphics cards varies a lot, the buffer should be based on a i7 920(for now a reasonable standard) and an ati xxxx (to be determined) as a reference.
Any comments?

Unfortunately some projects do not allow a large buffer.
I think, there should be an allowance of at least 24h.
As performance of computers/graphics cards varies a lot, the buffer should be based on a i7 920(for now a reasonable standard) and an ati xxxx (to be determined) as a reference.
Any comments?

Sadly, we can't up the WU cache for here any higher than it already is. Part of the reason is the nature of the work we're doing, and the other part is our poor server wouldn't be able to keep up.

On another note the server should be generating work again now that I'm back and have fixed things.

---

Thank You, hope the weekend was fun at least!

---

"There is no limit to the amount of good a
person can do if they do not care who gets credit for it."