Posts by RichardRodd

11) Message boards : Number crunching : A blue world for aborting...? (Message 36000) Posted 24 Jan 2009 by RichardRodd Post: Hi. I seem to have picked up a clutch of blue worlds on my small family of machines. The first one is hadsm3fub_k78a_005974157_9 which I started late December. I only just spotted it\'s blue! It\'s never trickled, is already running at 90 s/TS and the graphics display tells me it\'s only got to model date 25/03/1811 (a cold day for March...!). Can you advise me is this one for aborting? I\'m very happy to keep the faith - I have had blue worlds run for months and eventually complete, but I don\'t think I\'ve ever seen one this early in the cycle. Thanks for your help - Richard
12) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 34691) Posted 19 Aug 2008 by RichardRodd Post: It was a service installation.
13) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 34689) Posted 19 Aug 2008 by RichardRodd Post: It was a service installation.
14) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 34681) Posted 18 Aug 2008 by RichardRodd Post: Curioser and curioser... So, no worries really: I uninstalled 6.2.18 and reinstalled 5.10.45 which picked up work-in-progress (neat, that) AND \'reinstated\' all comms working as before. So summat changed between those two versions of BOINC as far as I can infer - but not on my system!! Any ideas? \'Cos the day will come when I\'ll have to move off 5.10.45. Richard
15) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 34680) Posted 18 Aug 2008 by RichardRodd Post: Greetings again. At the risk of yet more embarrassment... It seemed I\'d avoided the problem I discussed here by simply killing CyberPatrol every time I chose manually to release trickles: inelegant, but it worked. But now I\'ve upgraded to BOINC V6.2.18, ands I\'m back to square one - except this time with CyberPatrol definitely killed (as far as I can see...!) I\'m still getting all the comms error messages I was before. Am I missing something? Is there a \'reset\' I can do? I have followed all the seteps suggested above as before. Nothing else has changed in my setup, and with CyberPatrol killed, it seems I still can\'t establish comms with the BOINC server. Your help once again much appreciated... Richard
16) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 33306) Posted 12 Apr 2008 by RichardRodd Post: Mike (et al), hi. I tried the \'update\' just in case it did anything significantly - but I was already pretty sure it\'s just another way of kicking-off comms between client and server. So I\'m still looking for a little more help here. This realtime CP monitor which I switched on only reports on webpages - so it \'sees\' me writing on this page, for instance. But trickles and the like, it doesn\'t see, but still intercepts. Yesterday I mentioned the file sched_request_climateprediction.net.xml, and when I looked through it (with my almost layman\'s eyes) I spotted the three letters \'rpc\', which made me wonder about how the whole BOINC thing is architected - grid for sure, but maybe client-server using remote procedure calling? So then I start thinking that maybe we have a client-server app, with my client sending out requests (in which the CP app has no interest) and receiving stuff back from the server - an xml file which CP intercepts and corrupts, even though CP doesn\'t see it as web activity, it\'s still interfering (with the best of intentions!!). This begins to scale up from my own personal situation to a wider question of whether there\'s a design issue relating to the co-hosting (if you will) of internet activity monitors (to genericise the CP app) with BOINC (or other grid-type client-server apps). If I\'m right with this (i.e. BOINC being an RPC-based client-server app), I sort of think this needs to become a pretty technical discussion - which I might struggle to hold up at my end! Again, all comments and feedback gratefully received. If anyone watching this thread can confirm the basic BOINC architecture, then at least I could go back to the CP folk and ask them why CP intercepts/corrupts inbound RPC calls. We might just have two incompatible technologies here - and just to have teased that out might be useful for the BOINC community as a whole. Richard
17) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 33292) Posted 11 Apr 2008 by RichardRodd Post: Hello again. Sorry to be away for a few days - work and stuff called. But now it\'s the weekend...! Despite all the adverse comment about these internet guards :-) the CyberPatrol people seem genuinely keen to help me resolve this problemette. They asked me to run a utility that apparently monitors in realtime incoming and outgoing requests to, and traffic from, webpages. So I started the thing up, and asked BOINC to \'retry communications\', only to see this utility trap nothing! Okay, I hear all you real techies out there express no surprise at that!! So please help me here to have at least one more dialogue with CP. Can you help me understand to what BOINC sends the file sched_request_climateprediction.net.xml? I get the impression that this file clearly seems to get through, because (it seems to me) that it\'s the reply (from wherever) that gets intercepted by CP. I\'m trying to understand why this CP monitor doesn\'t see the outgoing or incoming traffic. Okay - all suggestion most gratefully received - as ever. I\'ll get to all the coments about reinstalling BOINC in different forms when I run this thread to exhaustion. Mike - I completely agree with your sentiments regarding workarounds, and public placing of PCs. Maybe those of us who grew up without all this PC stuff just have to let today\'s kids play with their generation\'s fire. I\'m sure in our turn we all had our own temptations... But, take it from me, it\'s distressing when one\'s kids seem drawn to the Dark Side. Maybe we\'ll have a rethink about placement within the family home -- and I do appreciate your comments. Richard
18) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 33247) Posted 7 Apr 2008 by RichardRodd Post: Argh...!!! All I did was copy-and-paste - I never read the file!! So thanks for pointing out the obvious and mea culpa. With a household of teenagers determined to plunder the darkest parts of the internet, I\'m afraid this protective parent has resorted to CyberPatrol. And running BOINC as a service means that the CP intercept (which normally displays on the screen) apparently gets caught in the sched_reply_climateprediction.net.xml file. Heavy-handed CP may be, as per the Register incident, but frankly better that than adult material on tap for my kids! But hey that\'s a personal thing and not for this forum... Anyhow, if I can find a fix, we may want to capture the resolution for other similarly afflicted folk...? Finally, if all else fails and I need to reinstall BOINC not as a service, how do I avoid losing the few hours work clocked-up on this m/c? Thanks for all your help. Richard
19) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 33242) Posted 7 Apr 2008 by RichardRodd Post: Thanks for your help Iain - my technical stature plummets by the minute! So here\'s the contents of the file sched_reply_climateprediction.net.xml -- <!-- saved from url=(0014)about:internet --> <HTML xmlns:v=\"urn:schemas-microsoft-com:vml\"> <HEAD> <TITLE>CyberPatrol Intercept</TITLE> <META HTTP-EQUIV=\"Pragma\" CONTENT=\"no-cache\"> <META HTTP-EQUIV=\"Expires\" CONTENT=\"-1\"> <style> v\\:* { behavior: url(#default#VML); } </style> <script language=\"javascript\" type=\"text/javascript\"> var g_bImageDisplayed = false; var g_iTimer = 0; function ImageLoadOK(objItem) { objItem.style.display = \"inline\"; g_bImageDisplayed = true; } function OnTimer() { if (g_iTimer != 0) { clearInterval(g_iTimer); g_iTimer = 0; } if (g_bImageDisplayed == false) { var objVML = document.all.item(\"VmlImage\"); if (objVML != null) { objVML.style.display = \"inline\"; } else { document.all.item(\"BackupText\").style.display = \"inline\"; } } } function InsertImageCode(szFilename, iWidth, iHeight, szStrokeColour) { document.write(\'<div ID=\"StdImage\">\'); document.write(\'<IMG SRC=\"\' + szFilename + \'\" style=\"DISPLAY: none\" width=\"\' + iWidth + \'\" height=\"\' + iHeight + \'\" border=\"0\" onload=\"javascript:ImageLoadOK(this);\">\'); document.write(\'</div>\'); document.write(\'<!--[if gte IE 7]><DIV id=\"VmlImage\" style=\"DISPLAY: none\"><v:rect style=\"WIDTH: \' + iWidth + \'px; HEIGHT: \' + iHeight + \'px\" coordsize = \"21600,21600\" strokecolor = \"\' + szStrokeColour + \'\">\'); document.write(\'<v:imagedata src = \"\' + szFilename + \'\"></v:imagedata></v:rect>\'); document.write(\'</DIV><![endif]-->\'); document.write(\'<div ID=\"BackupText\" style=\"DISPLAY: none\">CyberPatrol</div>\'); } g_iTimer = window.setInterval(\"OnTimer()\", 1000); </script> </HEAD> <body bgcolor=\"#F2B12A\" text=\"#FFFFFF\" link=\"#770000\" vlink=\"#770000\" alink=\"#770000\"> <CENTER> <table border=\"1\" bgcolor=\"#FFFFFF\" width=\"400\" height=\"140\" bordercolor=\"#808080\"> <tr> <td width=\"100%\" align=\"center\"><font face=\"Verdana\" color=\"black\" size=\"6\"> <script language=\"javascript\" type=\"text/javascript\"> InsertImageCode(\"file:///C:/Applications/Other Applications/Internet Filtering/STYLES/shield/cp.gif\", 266, 129, \"white\"); </script> </font></td> </tr> </table> <h1><font face=\"Verdana\" color=\"#FFFFFF\">Access Restricted</font></h1> <p><font face=\"Verdana\" size=\"4\"><b>User Profile: <Default></b></font></p> <p><font face=\"Verdana\" size=\"4\"><b>Reason</b>: An internal error has occurred - Failed to load one or more CyberLISTs. This could be due to a corrupted CyberLIST file, please re-run setup to fix this.</font></p> <p><font face=\"Verdana\" size=\"4\"><b>Category</b>: None</font></p> <p><font face=\"Verdana\" size=\"4\"><!--OVERRIDE--></font></p> <p><font face=\"Verdana\" size=\"2\"><b>To change any of the filter settings please speak to your CyberPatrol HeadquartersÃ¢â‚¬â„¢ Administrator.</b></font></p> </CENTER> </body> </HTML> I hope that provides some insight! Richard
20) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 33239) Posted 7 Apr 2008 by RichardRodd Post: Hi. I just realised I forgot to reply to Thyme\'s query about the contents of the file sched_reply_climateprediction.net.xml, which might be illuminating in its brevity...! The XML page cannot be displayed Cannot view XML input using XSL style sheet. Please correct the error and then click the Refresh button, or try again later. -------------------------------------------------------------------------------- End tag \'div\' does not match the start tag \'IMG\'. Error processing resource \'file:///C:/Program Files/BOINC/sched_reply_cli... document.write(\'</div>\'); --------------------^ So am I right in thinking that the scheduler request (whatever that is) goes off okay, but the reply gets scrambled coming back in? And if so, why? Am I best off just trying a clean re-install? And if I do that, how do I keep the work done on the two units currently running (but not trickling!)? Okay - thanks in anticipation for your help. Richard
21) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 33235) Posted 6 Apr 2008 by RichardRodd Post: Thanks for those two comments / suggestions. So for Thyme here is the set of messages that came back at me after I\'d managed to create the xml doc - not the sharpest pencil in the technical pencil-case, me...! [http_debug] [ID#0] info: About to connect() to climateprediction.net port 80 (#0) [http_debug] [ID#0] info: Trying 137.108.64.198... [http_debug] [ID#0] info: Connected to climateprediction.net (137.108.64.198) port 80 (#0) [http_debug] [ID#0] Sent header to server: GET / HTTP/1.1 User-Agent: BOINC client (windows_intelx86 5.10.45) Host: climateprediction.net Accept: / Accept-Encoding: deflate, gzip Content-Type: application/x-www-form-urlencoded [http_debug] [ID#0] Received header from server: HTTP/1.1 200 OK [http_debug] [ID#0] Received header from server: Content-Type: text/html [http_debug] [ID#0] info: no chunk, no close, no size. Assume close to signal end [http_debug] [ID#0] Received header from server: [http_debug] [ID#0] info: Closing connection #0 climateprediction.net\|[sched_op_debug] Got master file; parsing climateprediction.net\|[sched_op_debug] Found 0 scheduler URLs in master file climateprediction.net\|[sched_op_debug] Deferring communication for 1 days 0 hr 0 min 0 sec climateprediction.net\|[sched_op_debug] Reason: 8 consecutive failures fetching scheduler list (I stripped out time-stamps just to shorten messages, and left in the apparent carriage returns.) Mike - this m/c sits on a home network behind a router connected to BT Broadband. You\'ll probably ask me next about router settings... so I\'ll have to find the family friend who has the password...!! Sorry, I don\'t know if the router counts as a proxy server (see first comment above!) When you say BAM, you mean the BOINC Manager? In which case, yes. Don\'t know if that helps to be going on with... But it clearly communicated okay when I set the thing up - downloaded from BOINC, set up as a service, and off it went quite smoothly - or so it seemed. Looking forward to your next comments. Richard
22) Questions and Answers : Windows : Error Message: No start tag in scheduler reply (Message 33222) Posted 6 Apr 2008 by RichardRodd Post: Greetings to all msg board watchers! I can\'t find this error msg in the msg boards anywhere... would someone care to enlighten me please...? Is there a problem at my end, or at the server end? Do I have to do anything to fix it or just keep the faith? I ask because this is the first time this m/c (my fourth) has tried to trickle-up: this is a different type of install for me - running on a multi-user m/c as a service (and getting that to work was a bundle of fun...!! So there might be a buried problem in the setup, although it does seem to be running okay, i.e. using CPU time anyway). Okay, thanks for your help - whoever will...!! Richard
23) Questions and Answers : Windows : Pre-emptive downloads... (Message 32848) Posted 4 Mar 2008 by RichardRodd Post: Many thanks Iain.
24) Questions and Answers : Windows : Pre-emptive downloads... (Message 32843) Posted 4 Mar 2008 by RichardRodd Post: I guess I know the answer to this but I thought I\'d ask anyhow... Is there any way of downloading a new work unit sooner than it would normally be provided (at approx 97-98% complete for running task)? I have two machines - one with me - the other one stays at home and works away - connected for trickles, except of course when the line drops... If that happens while I\'m away the m/c stops. But if I could download the next units(units - it\'s a dual core) of work NOW, which I know will start in a couple of days time, then if the line drops when I\'m away, the m/c just carries on... Or is that all just too keen?! Richard
25) Questions and Answers : Windows : Trickles leaking away? (Message 32759) Posted 27 Feb 2008 by RichardRodd Post: Sorry Chris...!! Keep up the the good work! Richard
26) Questions and Answers : Windows : Trickles leaking away? (Message 32757) Posted 27 Feb 2008 by RichardRodd Post: I\'ve noticed that neither of my machines seem to have had any trickles credited for the last couple of days, despite the fact that they\'re both chuntering away steadily. THIS m/c definitely trickled yesterday and today, and I\'m sure the other one (tucked away at home) is doing likewise. Have I missed something? (I did check that s/TS is running as expected, so I\'m not heading into blue models.) Many thanks - Richard
27) Message boards : Number crunching : Iceworlds & Slowdowns hadsm3/mh - Closed - Discussion (Message 31771) Posted 18 Dec 2007 by RichardRodd Post: Yes you\'re quite right - and I tried to transfer the data sooo carefully...! So the \'big\' one (which incidentally always said it was going to take 1900 hours) was the one on 137717 with a s/TS of 22.71... And the \'little\' one (400 hours) was on 235777 with a s/TS of 1.88. But both definitely have the same total number of TS = 259248... which I guess is what you\'d expect if they\'re actually the same size pkg, despite differing estimated runtimes. I might just hold fire a short while and see if anyone else has a different insight / recommendation...! Failing that, I\'ll kill 8091_1. And many thanks for your help.
28) Message boards : Number crunching : Iceworlds & Slowdowns hadsm3/mh - Closed - Discussion (Message 31767) Posted 18 Dec 2007 by RichardRodd Post: I have TWO Blue Planets as follows...!! BLUE PLANET 1 Model/ResultID webpage is at http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=6913635 Current timestep from globe graphic = 235777 of 259248. The s/TS value is 1.88. Whether the temperature display of the globe graphic is blue. YES. CPU is Intel Core 2 Duo T5600 1.83Ghz. Whether you are overclocking. No BLUE PLANET 2 Model/ResultID webpage is at http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=6983460 Current timestep from globe graphic = 137717 of 259248. The s/TS value is 22.71. Whether the temperature display of the globe graphic is blue. YES. ===== Interesting to note the same number of timesteps but Planet 1 is a \'big\' model suggesting approx 1900 total hours to run, while the other \'small\' one comes in at 400 hours. Looking forward to hearing from you...!

Previous 20