The people doing the wu's need to pay attention to what their machines are doing and take action when necessary. There's always that handy little "No New Work" button if the cache becomes to large. With the almost total lack of downtime on Einstein I've never seen any reason to carry a large cache, but, I still check my boxes on a daily basis to make sure everything is OK.
I keep 3 days to cover against ISP related outages on the weekend. Otherwise I'd agree.
PS and OT, are you the Jim Bailey I met in the barfly suites at stellarcon and libertycon this year?
A little update on host 198136: This person now has a total of 1786 WU's issued to him. There were 78 cases of "no reply" today. Last WU he returned was on 10/28/06. And, the number of WU's increases with each passing day.
Well, at least at this point the host is in the "penalty box" now, and won't get more than 4 result a day until the log jam of results onboard all expire and it starts getting some back on time again, or the owner realizes something went wrong and interceeds manually.
BTW, both of this persons hosts seem to be in this condition, and the other one has been in this state for almost a month. This would seem to indicate that option two above is unlikely to happen. ;-)
It will be interesting to see how this plays out, and if they finally get back to actually doing something on their own (and how long it takes). :-)
I suppose this might be an example of the "Out of sight... Out of mind." downside of a BOINC service install.
That's about all we can do, just wait and see. Just gets a little irritating at times. I returned on in Sept that was sent out the first time in July, I bet whoever returned the first result was wondering what happened on that one.
A little off track here: Have a P4 that is acting up, will only run for about 30 minutes at a time. Checked everything I can think of and haven't found anything wrong with it. Question is, can the WU's from that machine be transfered to another 'puter? There's a half dozen or so of them.
Never mind, just found the problem! Power supply was going belly up in the pond!
I've had a few that exceeded 2 months between my result and my first getting credit myself.
This is anecdotal, but both times in the recent past when I started up a new box I initially got mostly 'leftovers' from several different datasets. Most of the leftovers where _2 or _3's. Considering that new hosts are most likely to drop early I think the scheduler should avoid giving them rework until they've proved thier reliability. In fact, one of the machines was shut down a few days later. (It was my parents old PC I was trying to troubleshoot while they used my old one, my old PC then smoked it's mobo, forcing me to return thiers to them.)
RE: The people doing the
)
I keep 3 days to cover against ISP related outages on the weekend. Otherwise I'd agree.
PS and OT, are you the Jim Bailey I met in the barfly suites at stellarcon and libertycon this year?
RE: I keep 3 days to cover
)
That's about what I keep in the way of a cache. Never needed more with Einstein!
PS and OT
Wasn't me, I run a small family farm in NW Arkansas. To much work to ever get away from here.
A little update on host
)
A little update on host 198136: This person now has a total of 1786 WU's issued to him. There were 78 cases of "no reply" today. Last WU he returned was on 10/28/06. And, the number of WU's increases with each passing day.
I don't think there's
)
I don't think there's anything that can be done about it before he notices the problem himself, is there?
Nothing that I can think of!
)
Nothing that I can think of!
Well, at least at this point
)
Well, at least at this point the host is in the "penalty box" now, and won't get more than 4 result a day until the log jam of results onboard all expire and it starts getting some back on time again, or the owner realizes something went wrong and interceeds manually.
BTW, both of this persons hosts seem to be in this condition, and the other one has been in this state for almost a month. This would seem to indicate that option two above is unlikely to happen. ;-)
It will be interesting to see how this plays out, and if they finally get back to actually doing something on their own (and how long it takes). :-)
I suppose this might be an example of the "Out of sight... Out of mind." downside of a BOINC service install.
Alinator
That's about all we can do,
)
That's about all we can do, just wait and see. Just gets a little irritating at times. I returned on in Sept that was sent out the first time in July, I bet whoever returned the first result was wondering what happened on that one.
A little off track here: Have a P4 that is acting up, will only run for about 30 minutes at a time. Checked everything I can think of and haven't found anything wrong with it. Question is, can the WU's from that machine be transfered to another 'puter? There's a half dozen or so of them.
Never mind, just found the problem! Power supply was going belly up in the pond!
I've had a few that exceeded
)
I've had a few that exceeded 2 months between my result and my first getting credit myself.
This is anecdotal, but both times in the recent past when I started up a new box I initially got mostly 'leftovers' from several different datasets. Most of the leftovers where _2 or _3's. Considering that new hosts are most likely to drop early I think the scheduler should avoid giving them rework until they've proved thier reliability. In fact, one of the machines was shut down a few days later. (It was my parents old PC I was trying to troubleshoot while they used my old one, my old PC then smoked it's mobo, forcing me to return thiers to them.)
Checked to see how many went
)
Checked to see how many went red today, 124 is the count. Many of the ones I shared have already been re-issued, completed, and returned. :)
Good that most other
)
Good that most other crunchers return their WUs a bit faster ;-)