Machine is getting no credits for 2.07 (GPU)
17 Apr 2020 21:05:41 UTC
Topic 221994
(moderation:
I have a rather bizarre problem with one of my 2 crunchers (it's ugh-pc - ID: 12815644). Since I (re)started at Einstein full time (April 2, 2020), ugh-pc has been averaging about 500K credits/day. Today, in 9 hours since ~10am UTC, it has added only 76K credits, about 1/2 what I would expect.
Looking at the tasks on the machine, and specifically those for 2.07 (GPU version), I find that (at this time) I have approx. 135 tasks completed consecutively just recently, with no validation. I checked the tasks, and some have wingmen that I assume are slower than me. BUT - some of them are listed as UNSENT to anyone, with a sent date of 1970 (!!!).
Can someone enlighten me as to what is going on? I thought tasks were sent out to users in pairs, so how come no wingman? And why just this machine. My other cruncher, big32 (ID: 12820411 has the more-or-less normal credit today of 290K for the 9 hours).
Any suggestions will be appreciated.
PS: I can't get "<!--break-->" to work at all, it just gets ignored with or without the "s, at least in preview.
Quote:BUT - some of them are
)
Ughh! Your PC is appropriately named. I got sent 100 of that same type with a return date of 1970, decades before they were ever sent. They were what I call "insta-expired" within a couple minute of them being sent. Instant errors.
The server glitched on the date time for that event. Happened to you also it seems.
That is normal. They send out
)
That is normal. They send out 1 work unit at a time with no wingman (unlike seti). They eventually send out the verifying work unit after sometime. The Jan 1, 1970 is the default setting for the software. It gets changed once a work unit is sent out to the actual time it was sent. I had over 500 work units of the new Gravity Wave VelaJr sitting there without a wingman (finally started to get some of them sent to other for verification.) Just going to have to wait.
Not in my case. I still have
)
Not in my case. I still have 25 of those errors hanging around that haven't been cleared yet. Take a look at the task and stderr.txt. Timed out within 15 minutes of being sent.
https://einsteinathome.org/workunit/447326155
https://einsteinathome.org/task/936769671
Take a look at the received date - 1 Jan 1970 or the beginning of Unix time.
Not in my case. I still have
)
Not in my case. I still have 25 of those errors hanging around that haven't been cleared yet. Take a look at the task and stderr.txt. Timed out within 15 minutes of being sent.
https://einsteinathome.org/workunit/447326155
https://einsteinathome.org/task/936769671
Take a look at the received date - 1 Jan 1970 or the beginning of Unix time.
Keith Myers wrote:Not in my
)
It never ran so there is no run time. Don't know what caused it to error immediately. Unfortunately I can't see when it says it was sent to you. All I can see is when it returned the work unit. My reply to C-A was in regard to his wingman having a sent date of Jan 1, 1970 and being unsent.
@Zalster: you said I had
)
@Zalster: you said
I had over 500 work units of the new Gravity Wave VelaJr sitting there without a wingman
I say: yes, that's the one that is killing me, too. Ugh, to coin a phrase. That's a lot of credit not being issued...of course, it's not really a concern, except I look at the credit fairly often to see if something is broken, like one of my video cards. Now I know. Thanks, I think. *sigh*
Zalster wrote:Keith Myers
)
I am still getting almost
)
I am still getting almost entirely no wingmen with my 2.07 (GPU) tasks, same as above. Why is this happening? Should I shut off my machine(s) and at least save on my electric bill? Is this project becoming (already!) as lazy as Seti was at times recently in not caring about their volunteers? Frankly, this sucks.
And are they fated to expire with no credit granted in 2 weeks or so? How long will they sit while waiting for validation? Or will they just hang for an indefinite length of time?
You are using more
)
You are using more application than must the GW gpu app. The other app tasks are mostly doing fine.
You do have a ton of aborted errors and that is going to be held against you for the project not sending you work till you validate some of your returned work.
The servers just haven't sent out the next replication on some of your tasks yet. Wait and see if the servers finally get around to noticing the tasks need to be resent.
Cruncher-American wrote:I am
)
My guess is that are not a lot of people crunching the GW work units. I don't think it will take 2 weeks before they are sent but my guess will be a few days. Without knowing how they decide when to send the second work unit out, it's hard to say. I would say, they probably check to see if the work unit is returned then send it out. Eventually they will send them out so I will continue to crunch them for the foreseeable future. Your aborts should not have any effect on them sending out the second work unit to other people.