21) Message boards : Number crunching : What exactly does this mean? (Message 1161153) Posted 11 Oct 2011 by gcpeters Post: Holy Jebus. Is this much manual intervention really needed? Whatever happened to coding "set and forget?" Once again my RAC is taking a nosedive due to various and sundry "under the covers" code that in my opinion could be better written. Instead, we have code that allows some folks to connect semi-regularly, but only on Thursdays, if the moon is full, and if they bounced there BOINC client 3 times while tapping their heels. Others just have to baby their systems along and then suddenly rejoice if they get some WUs to run... There has got to be a better way to do this...

22) Message boards : Number crunching : What exactly does this mean? (Message 1159233) Posted 5 Oct 2011 by gcpeters Post: But why does the one machine on this network segment upload and download just fine and suddenly this one keeps throwing out these log messages and doesn't get squat??? Is there something server side that tracks these machine's names and puts them in a queue for when they can upload and download? The code running here is just wacky...two identical machines running S@H...one works...the other no longer does. And again, what exactly does this code (Scheduler request failed: HTTP service unavailable) refer to? It would be nice if someone published a Batman decoder ring explaining all the stuff we see in the log files.

23) Message boards : Number crunching : What exactly does this mean? (Message 1158658) Posted 4 Oct 2011 by gcpeters Post: 10/3/2011 4:56:07 PM | SETI@home | Scheduler request failed: HTTP service unavailable I have multiple (identical) systems on this same network segment that are chugging along just fine...relatively speaking. I can only assume this must be server side noise...

24) Message boards : Number crunching : This is new to me... (Message 1158417) Posted 3 Oct 2011 by gcpeters Post: Weird. First time I've seen this in any of my machine's logs in the past year. When did this, to use your verbiage, 'screwy' code come into effect? Edit: And where are the change logs (if any)...for client and server side code? That would be nice to have available.

25) Message boards : Number crunching : This is new to me... (Message 1158393) Posted 2 Oct 2011 by gcpeters Post: I've never seen this before: 10/2/2011 3:35:20 PM | SETI@home | This computer has reached a limit on tasks in progress And prior to this latest hardware SNAFU at S@H, this system usually had many more tasks than this running. Where does this limitation reside? In what code is this determined? Is there a config file where this 'limit' is tracked? Furthermore, this system is set to cache 10 days worth of WUs. Why does it only pull down x amount sometimes instead of filling the cache as programmed? Is there overriding code on the S@H servers? Where is all this logic of who gets what and how much maintained? I'm just trying to understand the code here because since I recently started paying closer attention, I see a number of anomalies. I have around fifteen systems running SETI in my lab at work and at my home. To be honest, all of them rarely run S@H the same way even though all my preferences are identical. And the log files on most show differing entries despite being only a few feet apart on the same network segement. One can have a cache full of WUs and still pull down more. Another gets logs saying 0 tasks available at the exact same time.

26) Message boards : Number crunching : Can not get ANY Cache built up ! (Message 1156454) Posted 26 Sep 2011 by gcpeters Post: /me Whistles and twiddles thumbs while waiting for WUs... Btw, I was on vacation all last week and the systems in my lab have not been touched. They're just crunching along with a week's full of cached WUs...which are now empty. Ergo, something must've happened during the last "Tuesday down time." /me gestures at the guys who update the stuff in the SETI "closet"... 8^)

27) Message boards : Number crunching : Can not get ANY Cache built up ! (Message 1156295) Posted 26 Sep 2011 by gcpeters Post: Sigh... 9/26/2011 8:32:37 AM | SETI@home | Scheduler request completed: got 0 new tasks 9/26/2011 8:32:37 AM | SETI@home | Project has no tasks available SNAFU

28) Message boards : Number crunching : Another GPU question... (Message 1147719) Posted 31 Aug 2011 by gcpeters Post: Before the latest week+ RAID crash, and the subsequent "Project has no tasks available" afterward, I am now getting WUs again. Hooray! Unfortunately, one of my better number cruncher systems, that has a couple of Nvidia 9400GTs, is now only running two WUs at a time (the 2 GPU ones) instead of running 16 (system has 16 CPU cores) plus the two GPU WUs. What happened that would cause this? I haven't touched or changed any settings or prefs. Edit: Is the number of WUs that are running predicated upon the number of "Ready to start" WUs you have in the queue? My Task Mangler CPU usage is now only ~ 10% where with 16 WUs running and the two GPUs going the system normally ran @ 100%.

29) Message boards : Technical News : Drive (Aug 11 2011) (Message 1146795) Posted 29 Aug 2011 by gcpeters Post: Same here. All my new work has pretty much come to a halt. I've burned through almost all my extra WUs. The last few results I have trickling back to SETI are from older, slower machines. But most of those are also having the issue where uploads aren't working...

30) Message boards : Number crunching : HE connection problems thread (Message 1146792) Posted 29 Aug 2011 by gcpeters Post: Forgive me and my generic-speak. Everything stalled = no longer able to crunch data due to the inability to download (or even upload results on many machines) anything from this establishment. As I've mentioned before, all my systems are in a lab which basically is set and forget. Before the RAID recently croaked, everything was running fine. I haven't touched anything in weeks. RAC steadily climbing to almost 70K RAC. Since then...nada...

31) Message boards : Number crunching : HE connection problems thread (Message 1146747) Posted 29 Aug 2011 by gcpeters Post: 8/29/2011 4:25:36 AM | SETI@home | Computation for task 08mr11ai.22737.8411.5.10.191_3 finished 8/29/2011 4:25:37 AM | SETI@home | Computation for task 08mr11ai.22737.8411.5.10.233_2 finished 8/29/2011 4:25:38 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.191_3_0 8/29/2011 4:25:53 AM | SETI@home | Computation for task 08mr11ai.22737.8411.5.10.172_3 finished 8/29/2011 4:25:56 AM | SETI@home | Computation for task 08mr11ai.22737.8411.5.10.178_3 finished 8/29/2011 4:26:12 AM | SETI@home | Temporarily failed upload of 08mr11ai.22737.8411.5.10.185_3_0: HTTP error 8/29/2011 4:26:12 AM | SETI@home | Backing off 18 min 37 sec on upload of 08mr11ai.22737.8411.5.10.185_3_0 8/29/2011 4:26:55 AM | SETI@home | Temporarily failed upload of 08mr11ai.22737.8411.5.10.191_3_0: HTTP error 8/29/2011 4:26:55 AM | SETI@home | Backing off 13 min 0 sec on upload of 08mr11ai.22737.8411.5.10.191_3_0 8/29/2011 4:27:32 AM | SETI@home | Computation for task 08mr11ai.22737.8411.5.10.235_2 finished 8/29/2011 4:27:49 AM | SETI@home | Computation for task 08mr11ai.22737.8411.5.10.240_2 finished 8/29/2011 4:36:27 AM | SETI@home | Sending scheduler request: To fetch work. 8/29/2011 4:36:27 AM | SETI@home | Requesting new tasks for CPU 8/29/2011 4:37:43 AM | SETI@home | Scheduler request failed: HTTP service unavailable 8/29/2011 5:12:11 AM | SETI@home | Sending scheduler request: To fetch work. 8/29/2011 5:12:11 AM | SETI@home | Requesting new tasks for CPU 8/29/2011 5:13:28 AM | SETI@home | Scheduler request failed: HTTP service unavailable 8/29/2011 6:51:15 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.239_2_0 8/29/2011 6:51:15 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.184_3_0 8/29/2011 6:52:31 AM | SETI@home | Temporarily failed upload of 08mr11ai.22737.8411.5.10.239_2_0: HTTP error 8/29/2011 6:52:31 AM | SETI@home | Backing off 35 min 30 sec on upload of 08mr11ai.22737.8411.5.10.239_2_0 8/29/2011 6:52:31 AM | SETI@home | Temporarily failed upload of 08mr11ai.22737.8411.5.10.184_3_0: HTTP error 8/29/2011 6:52:31 AM | SETI@home | Backing off 39 min 1 sec on upload of 08mr11ai.22737.8411.5.10.184_3_0 8/29/2011 7:07:12 AM | SETI@home | Sending scheduler request: To fetch work. 8/29/2011 7:07:12 AM | SETI@home | Requesting new tasks for CPU 8/29/2011 7:08:28 AM | SETI@home | Scheduler request failed: HTTP service unavailable 8/29/2011 8:19:31 AM | SETI@home | update requested by user 8/29/2011 8:19:32 AM | SETI@home | Sending scheduler request: Requested by user. 8/29/2011 8:19:32 AM | SETI@home | Requesting new tasks for CPU 8/29/2011 8:19:33 AM | SETI@home | Scheduler request completed: got 0 new tasks 8/29/2011 8:19:33 AM | SETI@home | Project has no tasks available 8/29/2011 8:26:39 AM | SETI@home | Sending scheduler request: To fetch work. 8/29/2011 8:26:39 AM | SETI@home | Requesting new tasks for CPU 8/29/2011 8:26:40 AM | SETI@home | Scheduler request completed: got 0 new tasks 8/29/2011 8:26:40 AM | SETI@home | Project has no tasks available 8/29/2011 8:47:58 AM | SETI@home | Sending scheduler request: To fetch work. 8/29/2011 8:47:58 AM | SETI@home | Requesting new tasks for CPU 8/29/2011 8:48:02 AM | SETI@home | Scheduler request completed: got 0 new tasks 8/29/2011 8:48:02 AM | SETI@home | Project has no tasks available 8/29/2011 9:26:34 AM | SETI@home | Sending scheduler request: To fetch work. 8/29/2011 9:26:34 AM | SETI@home | Requesting new tasks for CPU 8/29/2011 9:26:35 AM | SETI@home | Scheduler request completed: got 0 new tasks 8/29/2011 9:26:35 AM | SETI@home | Project has no tasks available 8/29/2011 9:32:51 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.239_2_0 8/29/2011 9:32:51 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.184_3_0 8/29/2011 9:32:58 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.239_2_0 8/29/2011 9:32:58 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.185_3_0 8/29/2011 9:33:00 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.184_3_0 8/29/2011 9:33:00 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.191_3_0 8/29/2011 9:33:01 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.185_3_0 8/29/2011 9:33:01 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.233_2_0 8/29/2011 9:33:02 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.191_3_0 8/29/2011 9:33:02 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.172_3_0 8/29/2011 9:33:03 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.233_2_0 8/29/2011 9:33:03 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.178_3_0 8/29/2011 9:33:04 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.172_3_0 8/29/2011 9:33:04 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.235_2_0 8/29/2011 9:33:05 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.178_3_0 8/29/2011 9:33:05 AM | SETI@home | Started upload of 08mr11ai.22737.8411.5.10.240_2_0 8/29/2011 9:33:06 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.235_2_0 8/29/2011 9:33:07 AM | SETI@home | Finished upload of 08mr11ai.22737.8411.5.10.240_2_0 8/29/2011 9:33:11 AM | SETI@home | Sending scheduler request: To fetch work. 8/29/2011 9:33:11 AM | SETI@home | Reporting 9 completed tasks, requesting new tasks for CPU 8/29/2011 9:33:13 AM | SETI@home | Scheduler request completed: got 0 new tasks 8/29/2011 9:33:13 AM | SETI@home | Project has no tasks available

32) Message boards : Number crunching : HE connection problems thread (Message 1146745) Posted 29 Aug 2011 by gcpeters Post: @ Brad corson Same here :( Everything is stalled... Dead in the water... SNAFU...

33) Message boards : Number crunching : "...not requesting new tasks." ??? (Message 1132039) Posted 26 Jul 2011 by gcpeters Post: Fair enough. These systems I have running S@H have nothing but spare cycles so for me it's gravy. Otherwise I would throw most of the machines on my list in the scrap heap. It's obscene *cough* criminal the amount of hardware we throw out each month where I work...

34) Message boards : Number crunching : "...not requesting new tasks." ??? (Message 1131984) Posted 25 Jul 2011 by gcpeters Post: @ Claggy Maybe your Cache is full? Negative. Maybe you have more than twice the uploads to CPU cores. Negative. Maybe you have No New Tasks set. Negative. Maybe you have a task suspended (Boinc won't ask for work from a project with a suspended task). Negative. Maybe you have Network Activity Suspended. Negative. Maybe you have work fetch for that device disabled in your Project Preferences. Negative. Maybe Boinc 6.12.x is in extended Backoff. No idea what this is and have no idea why this would spontanteously happen. I have been in Mexico for the past week. I left my lab systems crunching. I get back and several of them have piles of completed tasks they aren't uploading and others say "not requesting new tasks" in the logs but have no new tasks uploading or waiting to start. Still others have a pile of tasks in the uploading and downloading queues. Isn't this supposed to be a "screensaver?" How much do I really have to fiddle with this after I set all my preferences? Sometimes it works and chugs along for weeks. I make absolutely zero changes to any settings to any machines in my lab. Suddenly some of my systems start to exhibit these weird symptoms with BOINC/SETI functionality. Did anything happen last week while I was away? Did servers go down? Did the network pipe yet again have issues? Yes, I have been a member of this search since 1999, but I didn't start seriously using my lab machines for all the spare cycles they had to number crunch SETI until the last 6 months. Is it just a given that when you get to this level of number crunching with BOINC/SETI that you have to baby-sit it along for it to work somewhat smoothly? I hate to always come across as a pain in the backside, but it seems this really could be designed/coded to work better with less fiddling. Maybe it's by design to keep out the riff-raff...? By now many who have read my comments/rants are probably pointing the finger at me on that last point ;) I literally just want to set and forget this app. and have my lab machines chug along on autopilot. Next year I want to look at these machines and see 16 million units completed and a decent RAC.

35) Message boards : Number crunching : "...not requesting new tasks." ??? (Message 1131840) Posted 25 Jul 2011 by gcpeters Post: Why not?

36) Message boards : Number crunching : How are people acheiving the super high RACs? (Message 1127223) Posted 13 Jul 2011 by gcpeters Post: So did I read correctly that even if I am not running any GPUs I should still install the Lunatics package because it is optimised for number crunching?

37) Message boards : Number crunching : How are people acheiving the super high RACs? (Message 1127126) Posted 13 Jul 2011 by gcpeters Post: Ok, I took one of the 16 core machines and put in two nVidia older GPU cards I had collecting dust (2x GeForce 9400GT). I also installed the latest Lunatics app (v0.38). Now, is there any easy way to see a difference in performance? Or do I just need to sit back and wait for the RAC to climb?

38) Message boards : Number crunching : How are people acheiving the super high RACs? (Message 1125971) Posted 8 Jul 2011 by gcpeters Post: @ Grant: I'm running these servers exclusively for SETI number crunching. They are mostly unused servers and some older admin workstations in my lab.

39) Message boards : Number crunching : How are people acheiving the super high RACs? (Message 1125920) Posted 8 Jul 2011 by gcpeters Post: I'm running Dell big-iron servers (R900, R910) with massive cooling and baseboard management...also power pigs btw ;) I'll throw some GPUs in them and see what happens...

40) Message boards : Number crunching : How are people acheiving the super high RACs? (Message 1125752) Posted 8 Jul 2011 by gcpeters Post: I have a pile of nVidia and ATI cards in the lab...will any GPU work, ala ATI/AMD or are we talking nVidia only? My servers have multiple x8 and x16 PCIe slots each. I could load them up...

Previous 20 · Next 20

 
©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.