Page 3 of 3
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 1:38 am
by Punchy
EVGA has dozens of users completing/failing tens to hundreds of WU per update. How can they get out of that state, other than waiting for a fix from the Stanford end? It does appear to be having a significant impact on overall points.
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 1:49 am
by Jesse_V
Punchy wrote:EVGA has dozens of users completing/failing tens to hundreds of WU per update. How can they get out of that state, other than waiting for a fix from the Stanford end? It does appear to be having a significant impact on overall points.
And it's not being helpful with their 80% success qualification for bonus points.
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 2:05 am
by bruce
Punchy wrote:EVGA has dozens of users completing/failing tens to hundreds of WU per update. How can they get out of that state, other than waiting for a fix from the Stanford end? It does appear to be having a significant impact on overall points.
That depends on what you mean by "significant impact on overall points" For the most part, FAH is designed around the multiple server approach. The failure of a single server, for whatever reason, creates a problem for each client/slot that is processing a WU from that server, but that should never be more than one WU per client/slot. While that's a significant number of WUs, everybody should have received WUs from other servers and should be busy folding them while the WUs for this particular server will remain on your system until the server can accept them.
No server is 100% reliable, no matter how many redundances it contains, and other work servers plus the Assignment Server increase the total reliability.
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 3:06 am
by Punchy
By significant I'd guess EVGA will be down 10-20% in the last 24 hours vs the previous two days. Not being entirely bigadv-heavy like other teams, that variability is not quite normal. It could be that the WU mix has shifted to less productive projects; it could also be that due to the AS issues, some clients that had not previously seen A4 cores are now running them, and failing with "bad" settings like smp 7.
Since there are multiple simultaneous problems at the moment, it's hard to tell people what to do to fix their issues. Is it possible to get 400+ completed WUs in the stats, with 0 points, due to the "server has no record" issue?
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 3:47 am
by Joe_H
Punchy wrote:Since there are multiple simultaneous problems at the moment, it's hard to tell people what to do to fix their issues. Is it possible to get 400+ completed WUs in the stats, with 0 points, due to the "server has no record" issue?
From what I was seeing in the stats as I looked into this problem, the answer to your question is Yes. Folders with a number of WU's waiting to upload could get quite a few 0 pt WU's added. But, they are not getting added in a way that appears to be effecting the 80% completion rate.
As for what to do in the meantime, I can offer
bruce's suggestion posted here - viewtopic.php?f=20&t=22628&p=225807#p225807.
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 8:17 am
by filu
What will the WU, which is still awaiting the results sent to the server? I have 14 such WU. I would like to remind that the free lending power of my computers, and I wish myself that the results of my work were taken seriously. I mean, I want the results of calculations has been received by the servers.
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 1:40 pm
by jaynick65
128.143.199.96 appears to be accepting connections again this morning.
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 1:58 pm
by orion
jaynick65 wrote:128.143.199.96 appears to be accepting connections again this morning.
Yep, mine uploaded this morning and got credited.
Re: 128.143.199.96 REJECT
Posted: Sun Oct 07, 2012 6:54 pm
by kasson
We brought the server back up early this morning. Apologies for the longer than usual delay.