Panic Mode On (82) Server Problems?

Author	Message
William Volunteer tester Send message Joined: 14 Feb 13 Posts: 801 Credit: 709,699 RAC: 3,016	Message 1341263 - Posted: 27 Feb 2013, 13:13:00 UTC
Eric has been considering using the AP compression on MB as well, but it's unclear if all the apps in circulation can handle it. Boinc offers the option of zipped files, but that mechanism is under review. Probably another development waiting for a round touit. ____________ A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1341263 ·

Wedge009 Volunteer tester Send message Joined: 3 Apr 99 Posts: 230 Credit: 60,807,632 RAC: 159,503	Message 1341316 - Posted: 27 Feb 2013, 18:17:24 UTC
AstroPulse work seems to be getting sent out again - local proxy is keeping things downloading. All but one host has had its download queue completed while I slept. Looking good so far... ____________ Soli Deo Gloria
ID: 1341316 ·

TBar Send message Joined: 22 May 99 Posts: 460 Credit: 9,026,579 RAC: 41,963	Message 1341325 - Posted: 27 Feb 2013, 18:48:48 UTC
Things are working really well here. Here is your BOINC on Windows Networking; 24-Feb-2013 14:48:36 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu: transient HTTP error 24-Feb-2013 14:48:36 [SETI@home] Backing off 3 hr 3 min 2 sec on download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu 24-Feb-2013 14:48:36 [SETI@home] Started download of ap_27jl12ad_B6_P0_00149_20130224_12797.wu 24-Feb-2013 14:50:25 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu: transient HTTP error 24-Feb-2013 14:50:25 [SETI@home] Backing off 4 hr 20 min 55 sec on download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu 24-Feb-2013 14:50:25 [SETI@home] Started download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu 24-Feb-2013 14:50:39 [SETI@home] Temporarily failed download of ap_04au12aa_B3_P1_00058_20130224_10430.wu: transient HTTP error 24-Feb-2013 14:50:39 [SETI@home] Backing off 4 hr 6 min 46 sec on download of ap_04au12aa_B3_P1_00058_20130224_10430.wu 24-Feb-2013 14:50:39 [SETI@home] Started download of ap_04au12ad_B0_P0_00084_20130224_17943.wu 24-Feb-2013 14:50:40 [SETI@home] Temporarily failed download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu: transient HTTP error 24-Feb-2013 14:50:40 [SETI@home] Backing off 4 hr 17 min 3 sec on download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu 24-Feb-2013 14:50:40 [SETI@home] Started download of ap_04au12ad_B1_P0_00050_20130224_09896.wu 24-Feb-2013 14:52:04 [SETI@home] Temporarily failed download of ap_27jl12ad_B5_P1_00314_20130224_07707.wu: transient HTTP error 24-Feb-2013 14:52:04 [SETI@home] Backing off 3 hr 7 min 26 sec on download of ap_27jl12ad_B5_P1_00314_20130224_07707.wu 24-Feb-2013 14:52:42 [SETI@home] Temporarily failed download of ap_04au12ad_B0_P0_00090_20130224_17943.wu: transient HTTP error 24-Feb-2013 14:52:42 [SETI@home] Backing off 35 min 38 sec on download of ap_04au12ad_B0_P0_00090_20130224_17943.wu 24-Feb-2013 14:52:48 [SETI@home] Started download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu 24-Feb-2013 14:52:48 [SETI@home] Started download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu 24-Feb-2013 14:53:04 [SETI@home] Temporarily failed download of ap_27jl12ad_B4_P1_00330_20130224_07577.wu: transient HTTP error 24-Feb-2013 14:53:04 [SETI@home] Backing off 2 hr 53 min 59 sec on download of ap_27jl12ad_B4_P1_00330_20130224_07577.wu 24-Feb-2013 14:53:04 [SETI@home] Started download of ap_04au12aa_B3_P1_00058_20130224_10430.wu 24-Feb-2013 14:53:39 [SETI@home] Temporarily failed download of ap_27jl12ad_B6_P0_00149_20130224_12797.wu: transient HTTP error 24-Feb-2013 14:53:39 [SETI@home] Backing off 27 min 8 sec on download of ap_27jl12ad_B6_P0_00149_20130224_12797.wu 24-Feb-2013 14:53:39 [SETI@home] Started download of ap_04ja13ae_B6_P1_00059_20130214_03790.wu 24-Feb-2013 14:54:11 [SETI@home] Temporarily failed download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu: transient HTTP error 24-Feb-2013 14:54:11 [SETI@home] Backing off 5 hr 17 min 8 sec on download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu 24-Feb-2013 14:54:11 [SETI@home] Started download of ap_27jl12ad_B4_P1_00330_20130224_07577.wu 24-Feb-2013 14:55:22 [SETI@home] Sending scheduler request: To fetch work. 24-Feb-2013 14:55:22 [SETI@home] Requesting new tasks for CPU and NVIDIA and ATI 24-Feb-2013 14:55:27 [SETI@home] Scheduler request completed: got 0 new tasks 24-Feb-2013 14:55:27 [SETI@home] No tasks sent 24-Feb-2013 14:55:27 [SETI@home] No tasks are available for AstroPulse v6 24-Feb-2013 14:55:27 [SETI@home] No tasks are available for the applications you have selected. 24-Feb-2013 14:55:48 [SETI@home] Temporarily failed download of ap_04au12ad_B0_P0_00084_20130224_17943.wu: transient HTTP error 24-Feb-2013 14:55:48 [SETI@home] Backing off 23 min 52 sec on download of ap_04au12ad_B0_P0_00084_20130224_17943.wu 24-Feb-2013 14:55:48 [SETI@home] Started download of ap_27jl12ad_B6_P1_00393_20130224_13352.wu 24-Feb-2013 14:56:13 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu: transient HTTP error 24-Feb-2013 14:56:13 [SETI@home] Backing off 3 hr 55 min 27 sec on download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu 24-Feb-2013 14:56:13 [SETI@home] Started download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu 24-Feb-2013 14:57:25 [SETI@home] Temporarily failed download of ap_04au12ad_B1_P0_00050_20130224_09896.wu: transient HTTP error 24-Feb-2013 14:57:25 [SETI@home] Backing off 2 min 12 sec on download of ap_04au12ad_B1_P0_00050_20130224_09896.wu 24-Feb-2013 14:57:48 [SETI@home] Started download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu 24-Feb-2013 14:57:57 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu: transient HTTP error 24-Feb-2013 14:57:57 [SETI@home] Backing off 5 hr 25 min 47 sec on download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu Here is your BOINC on Linux Networking; 2/27/2013 1:00:40 PM \| SETI@home \| Requesting new tasks for CPU and NVIDIA 2/27/2013 1:00:53 PM \| SETI@home \| Scheduler request completed: got 1 new tasks 2/27/2013 1:00:56 PM \| SETI@home \| Started download of ap_20oc12aa_B5_P0_00110_20130227_04393.wu 2/27/2013 1:05:59 PM \| SETI@home \| Sending scheduler request: To fetch work. 2/27/2013 1:05:59 PM \| SETI@home \| Requesting new tasks for CPU and NVIDIA 2/27/2013 1:06:05 PM \| SETI@home \| Scheduler request completed: got 1 new tasks 2/27/2013 1:06:07 PM \| SETI@home \| Started download of ap_27my12ah_B1_P1_00074_20130227_04438.wu 2/27/2013 1:11:10 PM \| SETI@home \| Sending scheduler request: To fetch work. 2/27/2013 1:11:10 PM \| SETI@home \| Requesting new tasks for CPU and NVIDIA 2/27/2013 1:11:24 PM \| SETI@home \| Scheduler request completed: got 1 new tasks 2/27/2013 1:11:26 PM \| SETI@home \| Started download of ap_20oc12aa_B4_P1_00190_20130227_04358.wu 2/27/2013 1:12:08 PM \| SETI@home \| Finished download of ap_20oc12aa_B5_P0_00110_20130227_04393.wu 2/27/2013 1:13:42 PM \| SETI@home \| Computation for task ap_12jl12ab_B3_P0_00232_20130225_03562.wu_0 finished 2/27/2013 1:13:42 PM \| SETI@home \| Starting task ap_16my12aa_B0_P1_00395_20130225_30963.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 0 2/27/2013 1:13:45 PM \| SETI@home \| Started upload of ap_12jl12ab_B3_P0_00232_20130225_03562.wu_0_0 2/27/2013 1:13:50 PM \| SETI@home \| Finished upload of ap_12jl12ab_B3_P0_00232_20130225_03562.wu_0_0 2/27/2013 1:16:13 PM \| SETI@home \| Finished download of ap_27my12ah_B1_P1_00074_20130227_04438.wu 2/27/2013 1:16:30 PM \| SETI@home \| Sending scheduler request: To fetch work. 2/27/2013 1:16:30 PM \| SETI@home \| Reporting 1 completed tasks 2/27/2013 1:16:30 PM \| SETI@home \| Requesting new tasks for CPU and NVIDIA and ATI 2/27/2013 1:16:41 PM \| SETI@home \| Scheduler request completed: got 3 new tasks 2/27/2013 1:16:44 PM \| SETI@home \| Started download of ap_27my12ah_B1_P1_00130_20130227_04438.wu 2/27/2013 1:16:44 PM \| SETI@home \| Started download of ap_20oc12aa_B5_P1_00140_20130227_04442.wu 2/27/2013 1:16:44 PM \| SETI@home \| Started download of ap_20oc12aa_B5_P1_00139_20130227_04442.wu 2/27/2013 1:21:47 PM \| SETI@home \| Sending scheduler request: To fetch work. 2/27/2013 1:21:47 PM \| SETI@home \| Requesting new tasks for CPU and NVIDIA and ATI 2/27/2013 1:21:52 PM \| SETI@home \| Scheduler request completed: got 2 new tasks 2/27/2013 1:21:54 PM \| SETI@home \| Started download of ap_20oc12aa_B5_P0_00243_20130227_04393.wu 2/27/2013 1:21:54 PM \| SETI@home \| Started download of ap_27my12ah_B1_P1_00158_20130227_04438.wu 2/27/2013 1:23:02 PM \| SETI@home \| Finished download of ap_20oc12aa_B4_P1_00190_20130227_04358.wu 2/27/2013 1:24:15 PM \| SETI@home \| Computation for task 12jl12ab.18380.81318.14.10.193_1 finished 2/27/2013 1:24:15 PM \| SETI@home \| Starting task 12jl12ab.18380.87453.14.10.28_0 using setiathome_enhanced version 609 (cuda23) in slot 4 2/27/2013 1:24:18 PM \| SETI@home \| Started upload of 12jl12ab.18380.81318.14.10.193_1_0 2/27/2013 1:24:20 PM \| SETI@home \| Finished upload of 12jl12ab.18380.81318.14.10.193_1_0 2/27/2013 1:26:53 PM \| SETI@home \| Finished download of ap_27my12ah_B1_P1_00130_20130227_04438.wu 2/27/2013 1:26:58 PM \| SETI@home \| Sending scheduler request: To fetch work. 2/27/2013 1:26:58 PM \| SETI@home \| Reporting 1 completed tasks 2/27/2013 1:26:58 PM \| SETI@home \| Requesting new tasks for NVIDIA and ATI 2/27/2013 1:27:15 PM \| SETI@home \| Scheduler request completed: got 1 new tasks 2/27/2013 1:27:17 PM \| SETI@home \| Started download of ap_20oc12aa_B4_P1_00280_20130227_04358.wu 2/27/2013 1:27:43 PM \| SETI@home \| Finished download of ap_20oc12aa_B5_P1_00139_20130227_04442.wu 2/27/2013 1:28:11 PM \| SETI@home \| Finished download of ap_20oc12aa_B5_P1_00140_20130227_04442.wu 2/27/2013 1:32:21 PM \| SETI@home \| Sending scheduler request: To fetch work. 2/27/2013 1:32:21 PM \| SETI@home \| Requesting new tasks for NVIDIA and ATI 2/27/2013 1:32:24 PM \| SETI@home \| Scheduler request completed: got 0 new tasks 2/27/2013 1:32:24 PM \| SETI@home \| No tasks sent 2/27/2013 1:32:24 PM \| SETI@home \| No tasks are available for AstroPulse v6 2/27/2013 1:32:24 PM \| SETI@home \| This computer has reached a limit on tasks in progress 2/27/2013 1:32:24 PM \| SETI@home \| Project has no tasks available 2/27/2013 1:33:26 PM \| SETI@home \| Finished download of ap_27my12ah_B1_P1_00158_20130227_04438.wu 2/27/2013 1:33:41 PM \| SETI@home \| Finished download of ap_20oc12aa_B5_P0_00243_20130227_04393.wu 2/27/2013 1:38:07 PM \| SETI@home \| Finished download of ap_20oc12aa_B4_P1_00280_20130227_04358.wu 2/27/2013 1:40:18 PM \| SETI@home \| Computation for task ap_03ja13ai_B1_P0_00057_20130130_05224.wu_4 finished 2/27/2013 1:40:18 PM \| SETI@home \| Starting task ap_23jn12aa_B6_P0_00068_20130221_26617.wu_0 using astropulse_v6 version 601 in slot 1 2/27/2013 1:40:20 PM \| SETI@home \| Started upload of ap_03ja13ai_B1_P0_00057_20130130_05224.wu_4_0 2/27/2013 1:40:24 PM \| SETI@home \| Finished upload of ap_03ja13ai_B1_P0_00057_20130130_05224.wu_4_0 2/27/2013 1:42:26 PM \| SETI@home \| Sending scheduler request: To fetch work. 2/27/2013 1:42:26 PM \| SETI@home \| Reporting 1 completed tasks 2/27/2013 1:42:26 PM \| SETI@home \| Requesting new tasks for ATI 2/27/2013 1:42:29 PM \| SETI@home \| Scheduler request completed: got 0 new tasks 2/27/2013 1:42:29 PM \| SETI@home \| No tasks sent 2/27/2013 1:42:29 PM \| SETI@home \| No tasks are available for AstroPulse v6 2/27/2013 1:42:29 PM \| SETI@home \| This computer has reached a limit on tasks in progress See any difference? The Squid has just put SIV out of work.
ID: 1341325 ·

Siran d'Vel'nahr Volunteer tester Send message Joined: 23 May 99 Posts: 5650 Credit: 3,589,072 RAC: 2,688	Message 1341327 - Posted: 27 Feb 2013, 18:53:12 UTC
Greetings, 15+ minutes and still only 19% done on a download? I'd like to... Keep on BOINCing...! :) much more smoothly! ____________ CMDR Siran d'Vel'nahr XO USS Vre'kasht NCC-33187 Siran's website: [ ONLINE! ]
ID: 1341327 ·

Wedge009 Volunteer tester Send message Joined: 3 Apr 99 Posts: 230 Credit: 60,807,632 RAC: 159,503	Message 1341353 - Posted: 27 Feb 2013, 20:30:28 UTC - in response to Message 1341325.
See any difference? The Squid has just put SIV out of work. Pardon my ignorance, but what's SIV? But certainly, this is the stark black-and-white difference I saw between BOINC connecting directly to Berkeley and using a local proxy server. Given that AstroPulse WUs are being generated again, I think we've proved beyond reasonable doubt that even a local proxy can help alleviate the constant download back-off issues. ____________ Soli Deo Gloria
ID: 1341353 ·

Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 2086 Credit: 6,842,706 RAC: 4,048	Message 1341361 - Posted: 27 Feb 2013, 20:37:59 UTC
Well I'm not using a local proxy, myself, though I have thought about it several times, but using one of the free proxies is working well for now. Got an AP to download from start to finish with no interruptions in 9 minutes: 2013-02-27 14:52:59 SETI@home Started download of ap_27my12ah_B2_P0_00234_20130227_15489.wu 2013-02-27 15:00:32 SETI@home Finished download of ap_27my12ah_B2_P0_00234_20130227_15489.wu ____________ Linux laptop uptime: 1484d 22h 42m Ended due to UPS failure, found 14 hours after the fact
ID: 1341361 ·

Claggy Volunteer tester Send message Joined: 5 Jul 99 Posts: 3513 Credit: 26,382,708 RAC: 17,867	Message 1341371 - Posted: 27 Feb 2013, 20:58:11 UTC - in response to Message 1341353. Last modified: 27 Feb 2013, 21:05:12 UTC
See any difference? The Squid has just put SIV out of work. Pardon my ignorance, but what's SIV? SIV (or System Information Viewer) is available here: http://rh-software.com/ As well as showing CPU and GPU load, temps, voltages, clock speeds and a thousand other bits of info, it has a Boinc Status screen, on that screen is an Auto (download) Retry selection, because this screen needs to be open for it to work i run two instances of SIV. Claggy
ID: 1341371 ·

Wedge009 Volunteer tester Send message Joined: 3 Apr 99 Posts: 230 Credit: 60,807,632 RAC: 159,503	Message 1341374 - Posted: 27 Feb 2013, 21:01:12 UTC
Thanks, I thought it might have been one of those external utilities. As for proxies, I didn't really want to try using an external one because of all the mixed results I'd heard - for some people it works really well, others not so much. Also the fact that proxies can be unreliable and then you're stuck having to find another one. But with the local proxy, everything is under your control, which I prefer even if it means making sure that the proxy itself is working. At least it means only monitoring one process instead of manually kicking stuck downloads on multiple hosts all the time. ____________ Soli Deo Gloria
ID: 1341374 ·

TBar Send message Joined: 22 May 99 Posts: 460 Credit: 9,026,579 RAC: 41,963	Message 1341379 - Posted: 27 Feb 2013, 21:10:00 UTC - in response to Message 1341353.
See any difference? The Squid has just put SIV out of work. Pardon my ignorance, but what's SIV? But certainly, this is the stark black-and-white difference I saw between BOINC connecting directly to Berkeley and using a local proxy server. Given that AstroPulse WUs are being generated again, I think we've proved beyond reasonable doubt that even a local proxy can help alleviate the constant download back-off issues. SIV is the App you see in the lower left of this image. If you look in the upper left corner of the SIV BOINC Status Window you will see an option to "Auto Retry in xxxx" It defaults to every 15 minutes until it detects a stalled process or you recheck the box. Then it Auto Retries every 2.5 minutes. Not bad when your downloads are stalling on an average of every 2 minutes. It's much better if your downloads don't stall to begin with...
ID: 1341379 ·

Wedge009 Volunteer tester Send message Joined: 3 Apr 99 Posts: 230 Credit: 60,807,632 RAC: 159,503	Message 1341381 - Posted: 27 Feb 2013, 21:12:21 UTC
That's what I would have needed... but I think this local proxy trick does a much better job, and in a central location (as opposed to multiple hosts). Thanks again for sharing your experience of it. ____________ Soli Deo Gloria
ID: 1341381 ·

msattler Volunteer tester Send message Joined: 9 Jul 00 Posts: 34718 Credit: 355,247,118 RAC: 380,401	Message 1341568 - Posted: 28 Feb 2013, 7:45:42 UTC
Hmmmmmm... No stats dump to Boincstats for a couple of days now. ____________ ****** "Ask not, what your kitty can do for you. Ask what you can do for your kitty." As it is kitten, so shall it be done.
ID: 1341568 ·

fscheel Send message Joined: 13 Apr 12 Posts: 73 Credit: 11,135,641 RAC: 3	Message 1341607 - Posted: 28 Feb 2013, 12:05:59 UTC
Could someone offer some incite as to why a linux based PC works while a windows based PC fails?
ID: 1341607 ·

msattler Volunteer tester Send message Joined: 9 Jul 00 Posts: 34718 Credit: 355,247,118 RAC: 380,401	Message 1341613 - Posted: 28 Feb 2013, 13:08:17 UTC
'Nuther interesting note..... Although AP is being split, currently there are some 20,000 ready to send. So it does not appear for whatever reason that much of it is being sent out. Wonder if they twiddled with the scheduler or feeder settings to throttle outbound AP? Or is something tangled up related to no stats going out? Dunno. ____________ ****** "Ask not, what your kitty can do for you. Ask what you can do for your kitty." As it is kitten, so shall it be done.
ID: 1341613 ·

Wedge009 Volunteer tester Send message Joined: 3 Apr 99 Posts: 230 Credit: 60,807,632 RAC: 159,503	Message 1341615 - Posted: 28 Feb 2013, 13:16:58 UTC
Luck of the draw, I think. I seem to be getting more AP WUs than usual, so maybe you're just missing out for the moment. As for Windows vs Linux, I'm not sure if the focus was meant to be on the operating system - certainly both my Windows and Linux hosts suffered terribly with trying to download WUs. It was only after installing a local proxy server (which happens to be Linux-based) that the download situation improved, at least for my personal LAN. ____________ Soli Deo Gloria
ID: 1341615 ·

msattler Volunteer tester Send message Joined: 9 Jul 00 Posts: 34718 Credit: 355,247,118 RAC: 380,401	Message 1341616 - Posted: 28 Feb 2013, 13:31:28 UTC - in response to Message 1341615. Last modified: 28 Feb 2013, 13:50:01 UTC
Luck of the draw, I think. I seem to be getting more AP WUs than usual, so maybe you're just missing out for the moment. As for Windows vs Linux, I'm not sure if the focus was meant to be on the operating system - certainly both my Windows and Linux hosts suffered terribly with trying to download WUs. It was only after installing a local proxy server (which happens to be Linux-based) that the download situation improved, at least for my personal LAN. Well, I can tell ya for a fact that something different is going on. AP just does NOT normally build any ready to send cache under usual circumstances. Not 20,500 ready to send. EDIT.... AP splitting rate is around 1/second currently. That's about 600 in a 10 minute status page update interval. In the last update, ready to send went up around 300, out in the field went up around 220, and that's with around 200 completed APs being returned in the same time frame. So, they are going out. It would appear that for whatever reason, AP splitting has gone wild. Hope nothing is amiss with the data or the splitter process. Or maybe georgem just got his dander up and decided to split AP like crazy. ____________ ****** "Ask not, what your kitty can do for you. Ask what you can do for your kitty." As it is kitten, so shall it be done.
ID: 1341616 ·

William Volunteer tester Send message Joined: 14 Feb 13 Posts: 801 Credit: 709,699 RAC: 3,016	Message 1341620 - Posted: 28 Feb 2013, 14:05:08 UTC
Looking at the graphs AP are going out at the usual rate. Being on a page with turnaround slight fluctuations in the result creation rate are lost from view. i.e. we can't see if that has increased. ____________ A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1341620 ·

Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 5290 Credit: 41,837,817 RAC: 32,870	Message 1341627 - Posted: 28 Feb 2013, 14:39:43 UTC - in response to Message 1341616. Last modified: 28 Feb 2013, 14:47:16 UTC
EDIT.... AP splitting rate is around 1/second currently. That's about 600 in a 10 minute status page update interval. In the last update, ready to send went up around 300, out in the field went up around 220, and that's with around 200 completed APs being returned in the same time frame. So, they are going out. It would appear that for whatever reason, AP splitting has gone wild. The number in progress has been climbing steadily since splitting re-started, but as you say the number ready to send is also climbing steadily. It was OK for an hour or so, then it developed a very small ready to send buffer (maybe a few 100 WUs) & then after another couple of hours it started it's present steady climb. EDIT- it might also explain the really, really slow (1-1.5kB/s) download rate for the last day or so. Although whatever was done while i was away on holidays appears to have sorted the Scheduler out- most replies within 10 seconds. ____________ Grant Darwin NT.
ID: 1341627 ·

Richard Haselgrove Volunteer tester Send message Joined: 4 Jul 99 Posts: 7670 Credit: 40,274,929 RAC: 21,546	Message 1341628 - Posted: 28 Feb 2013, 14:42:27 UTC
Creation rate is normal: but there are more out in the field: and at the same time more ready to send: Does that add up to a much reduced failure (early return) rate? Whether fewer with 100% blanking, fewer from a faulty B3_P1 tape, or just fewer app crashes I couldn't say - but it feels like good news.
ID: 1341628 ·

Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 5290 Credit: 41,837,817 RAC: 32,870	Message 1341644 - Posted: 28 Feb 2013, 15:37:55 UTC - in response to Message 1341628.
Pinging 208.68.240.18 results in 100% packet loss every time. Pinging 208.68.240.13 results in 25% packet loss on 1 in every 3-5 goes. ____________ Grant Darwin NT.
ID: 1341644 ·

Richard Haselgrove Volunteer tester Send message Joined: 4 Jul 99 Posts: 7670 Credit: 40,274,929 RAC: 21,546	Message 1341659 - Posted: 28 Feb 2013, 16:29:04 UTC - in response to Message 1341644.
Pinging 208.68.240.18 results in 100% packet loss every time. Pinging 208.68.240.13 results in 25% packet loss on 1 in every 3-5 goes. 208.68.240.18 was retired in October: the alternate download server has been 208.68.240.21 since then.
ID: 1341659 ·

Author

Message

Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 801
Credit: 709,699
RAC: 3,016

Eric has been considering using the AP compression on MB as well, but it's unclear if all the apps in circulation can handle it.
Boinc offers the option of zipped files, but that mechanism is under review.
Probably another development waiting for a round touit.
____________
A person who won't read has no advantage over one who can't read. (Mark Twain)

ID: 1341263 ·

Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 230
Credit: 60,807,632
RAC: 159,503

AstroPulse work seems to be getting sent out again - local proxy is keeping things downloading. All but one host has had its download queue completed while I slept. Looking good so far...
____________
Soli Deo Gloria

ID: 1341316 ·

TBar
Send message
Joined: 22 May 99
Posts: 460
Credit: 9,026,579
RAC: 41,963

Things are working really well here. Here is your BOINC on Windows Networking;

24-Feb-2013 14:48:36 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu: transient HTTP error
24-Feb-2013 14:48:36 [SETI@home] Backing off 3 hr 3 min 2 sec on download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu
24-Feb-2013 14:48:36 [SETI@home] Started download of ap_27jl12ad_B6_P0_00149_20130224_12797.wu
24-Feb-2013 14:50:25 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu: transient HTTP error
24-Feb-2013 14:50:25 [SETI@home] Backing off 4 hr 20 min 55 sec on download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu
24-Feb-2013 14:50:25 [SETI@home] Started download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu
24-Feb-2013 14:50:39 [SETI@home] Temporarily failed download of ap_04au12aa_B3_P1_00058_20130224_10430.wu: transient HTTP error
24-Feb-2013 14:50:39 [SETI@home] Backing off 4 hr 6 min 46 sec on download of ap_04au12aa_B3_P1_00058_20130224_10430.wu
24-Feb-2013 14:50:39 [SETI@home] Started download of ap_04au12ad_B0_P0_00084_20130224_17943.wu
24-Feb-2013 14:50:40 [SETI@home] Temporarily failed download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu: transient HTTP error
24-Feb-2013 14:50:40 [SETI@home] Backing off 4 hr 17 min 3 sec on download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu
24-Feb-2013 14:50:40 [SETI@home] Started download of ap_04au12ad_B1_P0_00050_20130224_09896.wu
24-Feb-2013 14:52:04 [SETI@home] Temporarily failed download of ap_27jl12ad_B5_P1_00314_20130224_07707.wu: transient HTTP error
24-Feb-2013 14:52:04 [SETI@home] Backing off 3 hr 7 min 26 sec on download of ap_27jl12ad_B5_P1_00314_20130224_07707.wu
24-Feb-2013 14:52:42 [SETI@home] Temporarily failed download of ap_04au12ad_B0_P0_00090_20130224_17943.wu: transient HTTP error
24-Feb-2013 14:52:42 [SETI@home] Backing off 35 min 38 sec on download of ap_04au12ad_B0_P0_00090_20130224_17943.wu
24-Feb-2013 14:52:48 [SETI@home] Started download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu
24-Feb-2013 14:52:48 [SETI@home] Started download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu
24-Feb-2013 14:53:04 [SETI@home] Temporarily failed download of ap_27jl12ad_B4_P1_00330_20130224_07577.wu: transient HTTP error
24-Feb-2013 14:53:04 [SETI@home] Backing off 2 hr 53 min 59 sec on download of ap_27jl12ad_B4_P1_00330_20130224_07577.wu
24-Feb-2013 14:53:04 [SETI@home] Started download of ap_04au12aa_B3_P1_00058_20130224_10430.wu
24-Feb-2013 14:53:39 [SETI@home] Temporarily failed download of ap_27jl12ad_B6_P0_00149_20130224_12797.wu: transient HTTP error
24-Feb-2013 14:53:39 [SETI@home] Backing off 27 min 8 sec on download of ap_27jl12ad_B6_P0_00149_20130224_12797.wu
24-Feb-2013 14:53:39 [SETI@home] Started download of ap_04ja13ae_B6_P1_00059_20130214_03790.wu
24-Feb-2013 14:54:11 [SETI@home] Temporarily failed download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu: transient HTTP error
24-Feb-2013 14:54:11 [SETI@home] Backing off 5 hr 17 min 8 sec on download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu
24-Feb-2013 14:54:11 [SETI@home] Started download of ap_27jl12ad_B4_P1_00330_20130224_07577.wu
24-Feb-2013 14:55:22 [SETI@home] Sending scheduler request: To fetch work.
24-Feb-2013 14:55:22 [SETI@home] Requesting new tasks for CPU and NVIDIA and ATI
24-Feb-2013 14:55:27 [SETI@home] Scheduler request completed: got 0 new tasks
24-Feb-2013 14:55:27 [SETI@home] No tasks sent
24-Feb-2013 14:55:27 [SETI@home] No tasks are available for AstroPulse v6
24-Feb-2013 14:55:27 [SETI@home] No tasks are available for the applications you have selected.
24-Feb-2013 14:55:48 [SETI@home] Temporarily failed download of ap_04au12ad_B0_P0_00084_20130224_17943.wu: transient HTTP error
24-Feb-2013 14:55:48 [SETI@home] Backing off 23 min 52 sec on download of ap_04au12ad_B0_P0_00084_20130224_17943.wu
24-Feb-2013 14:55:48 [SETI@home] Started download of ap_27jl12ad_B6_P1_00393_20130224_13352.wu
24-Feb-2013 14:56:13 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu: transient HTTP error
24-Feb-2013 14:56:13 [SETI@home] Backing off 3 hr 55 min 27 sec on download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu
24-Feb-2013 14:56:13 [SETI@home] Started download of ap_03ja13aa_B2_P1_00174_20130130_31206.wu
24-Feb-2013 14:57:25 [SETI@home] Temporarily failed download of ap_04au12ad_B1_P0_00050_20130224_09896.wu: transient HTTP error
24-Feb-2013 14:57:25 [SETI@home] Backing off 2 min 12 sec on download of ap_04au12ad_B1_P0_00050_20130224_09896.wu
24-Feb-2013 14:57:48 [SETI@home] Started download of ap_03ja13ai_B1_P0_00057_20130130_05224.wu
24-Feb-2013 14:57:57 [SETI@home] Temporarily failed download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu: transient HTTP error
24-Feb-2013 14:57:57 [SETI@home] Backing off 5 hr 25 min 47 sec on download of ap_03ja13ai_B1_P0_00089_20130130_05224.wu

Here is your BOINC on Linux Networking;

See any difference? The Squid has just put SIV out of work.

ID: 1341325 ·

Siran d'Vel'nahr
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 5650
Credit: 3,589,072
RAC: 2,688

Greetings,

15+ minutes and still only 19% done on a download?

I'd like to...

Keep on BOINCing...! :)

much more smoothly!

____________
CMDR Siran d'Vel'nahr XO
USS Vre'kasht NCC-33187

Siran's website: [ ONLINE! ]

ID: 1341327 ·

Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 230
Credit: 60,807,632
RAC: 159,503

See any difference? The Squid has just put SIV out of work.

Pardon my ignorance, but what's SIV?

But certainly, this is the stark black-and-white difference I saw between BOINC connecting directly to Berkeley and using a local proxy server. Given that AstroPulse WUs are being generated again, I think we've proved beyond reasonable doubt that even a local proxy can help alleviate the constant download back-off issues.
____________
Soli Deo Gloria

ID: 1341353 ·

Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 2086
Credit: 6,842,706
RAC: 4,048

Well I'm not using a local proxy, myself, though I have thought about it several times, but using one of the free proxies is working well for now. Got an AP to download from start to finish with no interruptions in 9 minutes:

2013-02-27 14:52:59 SETI@home Started download of ap_27my12ah_B2_P0_00234_20130227_15489.wu
2013-02-27 15:00:32 SETI@home Finished download of ap_27my12ah_B2_P0_00234_20130227_15489.wu

____________

Linux laptop uptime: 1484d 22h 42m
Ended due to UPS failure, found 14 hours after the fact

ID: 1341361 ·

Claggy

Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 3513
Credit: 26,382,708
RAC: 17,867

See any difference? The Squid has just put SIV out of work.

Pardon my ignorance, but what's SIV?

SIV (or System Information Viewer) is available here:

http://rh-software.com/

As well as showing CPU and GPU load, temps, voltages, clock speeds and a thousand other bits of info, it has a Boinc Status screen, on that screen is an Auto (download) Retry selection,
because this screen needs to be open for it to work i run two instances of SIV.

Claggy

ID: 1341371 ·

Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 230
Credit: 60,807,632
RAC: 159,503

Thanks, I thought it might have been one of those external utilities.

As for proxies, I didn't really want to try using an external one because of all the mixed results I'd heard - for some people it works really well, others not so much. Also the fact that proxies can be unreliable and then you're stuck having to find another one.

But with the local proxy, everything is under your control, which I prefer even if it means making sure that the proxy itself is working. At least it means only monitoring one process instead of manually kicking stuck downloads on multiple hosts all the time.
____________
Soli Deo Gloria

ID: 1341374 ·

TBar
Send message
Joined: 22 May 99
Posts: 460
Credit: 9,026,579
RAC: 41,963

See any difference? The Squid has just put SIV out of work.

Pardon my ignorance, but what's SIV?

But certainly, this is the stark black-and-white difference I saw between BOINC connecting directly to Berkeley and using a local proxy server. Given that AstroPulse WUs are being generated again, I think we've proved beyond reasonable doubt that even a local proxy can help alleviate the constant download back-off issues.

SIV is the App you see in the lower left of this image. If you look in the upper left corner of the SIV BOINC Status Window you will see an option to "Auto Retry in xxxx" It defaults to every 15 minutes until it detects a stalled process or you recheck the box. Then it Auto Retries every 2.5 minutes. Not bad when your downloads are stalling on an average of every 2 minutes.

It's much better if your downloads don't stall to begin with...

ID: 1341379 ·

Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 230
Credit: 60,807,632
RAC: 159,503

That's what I would have needed... but I think this local proxy trick does a much better job, and in a central location (as opposed to multiple hosts). Thanks again for sharing your experience of it.
____________
Soli Deo Gloria

ID: 1341381 ·

msattler

Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 34718
Credit: 355,247,118
RAC: 380,401

Hmmmmmm...
No stats dump to Boincstats for a couple of days now.
____________
******

"Ask not, what your kitty can do for you. Ask what you can do for your kitty."

As it is kitten, so shall it be done.

ID: 1341568 ·

fscheel

Send message
Joined: 13 Apr 12
Posts: 73
Credit: 11,135,641
RAC: 3

Could someone offer some incite as to why a linux based PC works while a windows based PC fails?

ID: 1341607 ·

msattler

Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 34718
Credit: 355,247,118
RAC: 380,401

'Nuther interesting note.....
Although AP is being split, currently there are some 20,000 ready to send.
So it does not appear for whatever reason that much of it is being sent out.
Wonder if they twiddled with the scheduler or feeder settings to throttle outbound AP?
Or is something tangled up related to no stats going out?
Dunno.
____________
******

"Ask not, what your kitty can do for you. Ask what you can do for your kitty."

As it is kitten, so shall it be done.

ID: 1341613 ·

Wedge009
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 230
Credit: 60,807,632
RAC: 159,503

Luck of the draw, I think. I seem to be getting more AP WUs than usual, so maybe you're just missing out for the moment.

As for Windows vs Linux, I'm not sure if the focus was meant to be on the operating system - certainly both my Windows and Linux hosts suffered terribly with trying to download WUs. It was only after installing a local proxy server (which happens to be Linux-based) that the download situation improved, at least for my personal LAN.
____________
Soli Deo Gloria

ID: 1341615 ·

msattler

Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 34718
Credit: 355,247,118
RAC: 380,401

Luck of the draw, I think. I seem to be getting more AP WUs than usual, so maybe you're just missing out for the moment.

As for Windows vs Linux, I'm not sure if the focus was meant to be on the operating system - certainly both my Windows and Linux hosts suffered terribly with trying to download WUs. It was only after installing a local proxy server (which happens to be Linux-based) that the download situation improved, at least for my personal LAN.

Well, I can tell ya for a fact that something different is going on. AP just does NOT normally build any ready to send cache under usual circumstances. Not 20,500 ready to send.

EDIT....
AP splitting rate is around 1/second currently. That's about 600 in a 10 minute status page update interval. In the last update, ready to send went up around 300, out in the field went up around 220, and that's with around 200 completed APs being returned in the same time frame. So, they are going out. It would appear that for whatever reason, AP splitting has gone wild. Hope nothing is amiss with the data or the splitter process.

Or maybe georgem just got his dander up and decided to split AP like crazy.
____________
******

"Ask not, what your kitty can do for you. Ask what you can do for your kitty."

As it is kitten, so shall it be done.

ID: 1341616 ·

William

Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 801
Credit: 709,699
RAC: 3,016

Looking at the graphs AP are going out at the usual rate.
Being on a page with turnaround slight fluctuations in the result creation rate are lost from view. i.e. we can't see if that has increased.
____________
A person who won't read has no advantage over one who can't read. (Mark Twain)

ID: 1341620 ·

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5290
Credit: 41,837,817
RAC: 32,870

EDIT....
AP splitting rate is around 1/second currently. That's about 600 in a 10 minute status page update interval. In the last update, ready to send went up around 300, out in the field went up around 220, and that's with around 200 completed APs being returned in the same time frame. So, they are going out. It would appear that for whatever reason, AP splitting has gone wild.

The number in progress has been climbing steadily since splitting re-started, but as you say the number ready to send is also climbing steadily. It was OK for an hour or so, then it developed a very small ready to send buffer (maybe a few 100 WUs) & then after another couple of hours it started it's present steady climb.

EDIT- it might also explain the really, really slow (1-1.5kB/s) download rate for the last day or so.
Although whatever was done while i was away on holidays appears to have sorted the Scheduler out- most replies within 10 seconds.
____________
Grant
Darwin NT.

ID: 1341627 ·

Richard Haselgrove

Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 7670
Credit: 40,274,929
RAC: 21,546

Creation rate is normal:

but there are more out in the field:

and at the same time more ready to send:

Does that add up to a much reduced failure (early return) rate? Whether fewer with 100% blanking, fewer from a faulty B3_P1 tape, or just fewer app crashes I couldn't say - but it feels like good news.

ID: 1341628 ·

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5290
Credit: 41,837,817
RAC: 32,870

Pinging 208.68.240.18 results in 100% packet loss every time.
Pinging 208.68.240.13 results in 25% packet loss on 1 in every 3-5 goes.

____________
Grant
Darwin NT.

ID: 1341644 ·

Richard Haselgrove

Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 7670
Credit: 40,274,929
RAC: 21,546

Pinging 208.68.240.18 results in 100% packet loss every time.
Pinging 208.68.240.13 results in 25% packet loss on 1 in every 3-5 goes.

208.68.240.18 was retired in October: the alternate download server has been 208.68.240.21 since then.

ID: 1341659 ·