Message boards :
Number crunching :
Panic Mode On (85) Server Problems?
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 . . . 23 · Next
Author | Message |
---|---|
kittyman Send message Joined: 9 Jul 00 Posts: 51469 Credit: 1,018,363,574 RAC: 1,004 |
There's a certain type of tasks that I haven't seen for a couple of days now. I refuse to whine and bitch though. Last I recall, my Boincs were set to grab everything, at any time. I have not looked at it in ages, since the server side limits have made most cache settings moot. A hundred tasks on a multi GPU rig do not make a 10 day cache. LOL, or not. Maybe a few hours. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
A hundred tasks on a multi GPU rig do not make a 10 day cache. LOL, or not. Maybe a few hours. A 100 MB WU cache not even least for 6 hrs in a 2x690 host... even less if they are shorties... |
kittyman Send message Joined: 9 Jul 00 Posts: 51469 Credit: 1,018,363,574 RAC: 1,004 |
A hundred tasks on a multi GPU rig do not make a 10 day cache. LOL, or not. Maybe a few hours. Yah. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
.clair. Send message Joined: 4 Nov 04 Posts: 1300 Credit: 55,390,408 RAC: 69 |
There's a certain type of tasks that I haven't seen for a couple of days now. I refuse to whine and bitch though. In that case i will have a whine and bitch about not geting any vlar for my 7970 rig for months, Oh, better not, you will think i have gone mad, did i forget to take my tablets today ? one, two, three, ten !! whatever . . . . |
betreger Send message Joined: 29 Jun 99 Posts: 11362 Credit: 29,581,041 RAC: 66 |
Much more important than bitching about Credit New, which we can do nothing about, is where are the APs? It has been days and we can do nothing about that either. |
bill Send message Joined: 16 Jun 99 Posts: 861 Credit: 29,352,955 RAC: 0 |
Doesn't the backup on the MBs have to be done first? |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
Doesn't the backup on the MBs have Indeed. It is unusual to have that many data sets (known as "tapes") loaded at one time, and since AP blasts through much faster than MB.. we will probably have to wait until there are only 10-15 tapes left before more get loaded. It might be a week or two. A lot of it also depends on how much storage space is available on the servers, too. I mean, 50gb per tape times...42 = 2150 GB. Sure, that may not sound like too terribly much, but then you have the WUs themselves that get split out of it, and those add up, as well. For every WU that has not validated yet, there is a file on disk for it until it validates and moves on to deletion. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
kittyman Send message Joined: 9 Jul 00 Posts: 51469 Credit: 1,018,363,574 RAC: 1,004 |
Doesn't the backup on the MBs have I have now about 161 APs in play. They are dropping down, but not too rapidly, as most are assinged to CPUs. And on CPUs, they take anywhere up to 20 hours to complete. I do them all, kids. Whatever it takes. The kitties do them all. No favorites here. The servers send it, we do it. That is what the project asks, that is what we do. I have not abandoned or otherwise rejected a task in more years than I can tell you. It's just something I do not do. Even during the vlar attack after the v7 changeover. I did not take out even one. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
ivan Send message Joined: 5 Mar 01 Posts: 783 Credit: 348,560,338 RAC: 223 |
There's a certain type of tasks that I haven't seen for a couple of days now. I refuse to whine and bitch though. Breakfast! :-) |
betreger Send message Joined: 29 Jun 99 Posts: 11362 Credit: 29,581,041 RAC: 66 |
As near as I can tell it has been a week with no APs being split. I hunger for an AP. |
bill Send message Joined: 16 Jun 99 Posts: 861 Credit: 29,352,955 RAC: 0 |
Probably Friday at the earliest. |
betreger Send message Joined: 29 Jun 99 Posts: 11362 Credit: 29,581,041 RAC: 66 |
OK, it's worse than I thought. |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
My cache is definitely slowly dwindling, but I'm hanging on with occasional re-sends. In about 24 hours, I will have run through the last of my _0 and _1 APs and start getting into the cache of re-sends. From there, if I get no more, 5 days or so. So realistically.. I'm empty and idle in about 7 days from now, but I'm sure I'll get a couple more re-sends before then, extending it out another day or two, depending on how many I get. No panic yet, and no complaining ever. It is my choice to go AP-only, and you have to take the good with the bad. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
Thomas Send message Joined: 9 Dec 11 Posts: 1499 Credit: 1,345,576 RAC: 0 |
Can't you crunching MB's instead of AP's during these lean times ? It's always better to crunching all WU's.... There's WU's for everyone and we would never fall down... Maybe ;) |
Speedy Send message Joined: 26 Jun 04 Posts: 1643 Credit: 12,921,799 RAC: 89 |
I'm not having a go at anyone. I thought the idea of this thread was to talk about server related problems? For example if download/upload speeds are slow. How come at some point in the thread does the topic of APs come up? This topic was also reasonably prominent in panic mode thread 84. Or am I missing something somewhere is this related to a server problem? If you would like to talk about that that is fine but can I suggest you start a thread dedicated to it? Thank you for taking the time to read this [/quote] |
arkayn Send message Joined: 14 May 99 Posts: 4438 Credit: 55,006,323 RAC: 0 |
I'm not having a go at anyone. I thought the idea of this thread was to talk about server related problems? For example if download/upload speeds are slow. How come at some point in the thread does the topic of APs come up? This topic was also reasonably prominent in panic mode thread 84. Or am I missing something somewhere is this related to a server problem?[/quote] Not really, it is more of a central Bitching & Moaning thread and can be used for other things as well. As such there is not all that much off topic in this thread except for politics. |
Dimly Lit Lightbulb 😀 Send message Joined: 30 Aug 08 Posts: 15399 Credit: 7,423,413 RAC: 1 |
Can't you crunching MB's instead of AP's during these lean times ? It's always better to crunching all WU's.... There's WU's for everyone and we would never fall down... Maybe ;) Personally I crunch astropulse only due to the length of them, one task and that's 14-16 hrs of crunching, excluding the B3_P1 & B6_P0 "shorties". Currently my cache has dwindled to a solitary task :( Member of the People Encouraging Niceness In Society club. |
Speedy Send message Joined: 26 Jun 04 Posts: 1643 Credit: 12,921,799 RAC: 89 |
I'm not having a go at anyone. I thought the idea of this thread was to talk about server related problems? For example if download/upload speeds are slow. How come at some point in the thread does the topic of APs come up? This topic was also reasonably prominent in panic mode thread 84. Or am I missing something somewhere is this related to a server problem? Thanks Atkayn for clarifying what this thread can be used for. |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
Can't you crunching MB's instead of AP's during these lean times ? It's always better to crunching all WU's.... There's WU's for everyone and we would never fall down... Maybe ;) Same here. It's easier to keep up with burning through 6 tasks per 24 hours, and keeps the client_state file relatively small. Of course with the bigger pipe to contact the servers, that's not much of an issue anymore as it was in the past. The mega crunchers (before the limits) would have client_state's that were upwards of 10MiB! That has to get sent to the server on every scheduler contact (was my understanding of it back then..maybe it's different now). So even though the data files for APs are much larger and take up more disk space and use a lot more bandwidth to transfer, it keeps the client_state smaller, and also helps to keep your tasks page on the website (and more importantly, for the database) shorter, and the tasks list in BOINC Manager shorter, too. The database performance issues are the reason for the limits anyway, and the way I see it, having fewer tasks is me doing my part to help the database just that little tiny bit... and also because of CPU-only and 6/day, I can hold a 10-day cache and not be affected by the limits, so that's a plus. Even though there is no downtime, I'm still able to keep my CPUs busy during this dark period of no new APs being split, all thanks to a 10-day cache that has turned into 15-18 thanks to the re-sends I've been able to scrounge up. That's how I see it. There.. see? Did my B&M-ing as well as threw legitimate server-related facts into it. :p Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
The mega crunchers (before the limits) would have client_state's that were upwards of 10MiB! That has to get sent to the server on every scheduler contact (was my understanding of it back then..maybe it's different now). The client_state.xml was never sent to the servers, it has always been the sched_request_[projectURL].xml, which contains much less information than the client_state and therefore is also much smaller. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.