| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 12
|
|
| Author |
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges:
|
After running and being switched off from time to time with no issues I have the following displayed;
----------------------------------------Waiting for shared memory??? Seen on the tasks page of a Linux cruncher. (Mint 12) This happened after stopping and re-starting Boinc(manager) and although other tasks have finished these still do not get into the queue. re-booting makes the problem worse by adding those that were crunching to those that were already waiting. Machine uses around 0.2GB at idle and there are 16GB installed I could use a little help with this as I am not proficient with Linux ![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
How much memory [RAM] do you allow BOINC to use when in Idle and Work state? I've got it at 75% (of 7935MB) and 95% (of 7935MB)... way more than enough to run 8 threads concurrent of the Memory hungry CFSW.
Anyway, please post the event/message log, from start to where the Waiting for Memory show up. Whilst not proficient at Linux, the BOINC Manager is identical in functionality... where you can read all preferences in effect via the Tools > Computing Preferences interface. --//-- (P.S. Never seen the 'shared' bit before, but BOINC core client (boinc.exe) and the science apps run in an [isolated] shared memory block. |
||
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges:
|
Start looks like this:
----------------------------------------Wed 23 May 2012 18:44:02 BST | World Community Grid | [error] ACTIVE_TASK::start(): can't create memory-mapped file: shmget() failed Set to use up to 80% of mems both in use and idle. which I just changed to 60%. I always start wcg first on this machine which also runs MJ12. perhaps this is an interaction between the two and I should run it last? ![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sorry, let me qualify the 'from start' as in 'from start - of BOINC'. Sample:
2 23-5-2012 19:02:05 Starting BOINC client version 7.0.28 for windows_x86_64 3 23-5-2012 19:02:05 log flags: file_xfer, sched_ops, task, checkpoint_debug, cpu_sched, dcf_debug 4 23-5-2012 19:02:05 log flags: sched_op_debug 5 23-5-2012 19:02:05 Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6 6 23-5-2012 19:02:05 Running as a daemon 7 23-5-2012 19:02:05 Data directory: G:\BOINC 8 23-5-2012 19:02:05 Running under account boinc_master 9 23-5-2012 19:02:05 Processor: 8 GenuineIntel Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz [Family 6 Model 42 Stepping 7] 10 23-5-2012 19:02:05 Processor: 256.00 KB cache 11 23-5-2012 19:02:05 Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx tm2 popcnt aes pbe 12 23-5-2012 19:02:05 OS: Microsoft Windows 7: Home Premium x64 Edition, Service Pack 1, (06.01.7601.00) 13 23-5-2012 19:02:05 Memory: 7.95 GB physical, 15.90 GB virtual 14 23-5-2012 19:02:05 Disk: 19.53 GB total, 18.71 GB free 15 23-5-2012 19:02:05 Local time is UTC +2 hours 16 23-5-2012 19:02:05 No usable GPUs found 17 23-5-2012 19:02:05 Config: use all coprocessors 18 23-5-2012 19:02:05 Config: GUI RPC allowed from any host 19 23-5-2012 19:02:05 Config: GUI RPC allowed from: 28 23-5-2012 19:02:05 Config: 127.0.0.1 localhost 43 World Community Grid 23-5-2012 19:02:05 URL http://www.worldcommunitygrid.org/; Computer ID 1234567; resource share 500 44 23-5-2012 19:02:05 General prefs: from http://bam.boincstats.com/ (last modified 22-May-2012 09:24:16) 45 23-5-2012 19:02:05 Host location: none 46 23-5-2012 19:02:05 General prefs: using your defaults 47 23-5-2012 19:02:05 Reading preferences override file 48 23-5-2012 19:02:05 Preferences: 49 23-5-2012 19:02:05 max memory usage when active: 6918.88MB 50 23-5-2012 19:02:05 max memory usage when idle: 7732.87MB 51 23-5-2012 19:02:05 max disk usage: 9.76GB 52 23-5-2012 19:02:05 don't use GPU while active 53 23-5-2012 19:02:05 suspend work if non-BOINC CPU load exceeds 75 % 54 23-5-2012 19:02:05 (to change preferences, visit the web site of an attached project, or select Preferences in the Manager) 56 23-5-2012 19:02:05 Not using a proxy 57 World Community Grid 23-5-2012 19:02:10 Restarting task GFAM_x3fow_PfPNP_V66I_V73I_Y160F_0021519_0083_1 using gfam version 611 in slot 4 58 World Community Grid 23-5-2012 19:02:10 Restarting task GFAM_x3fow_PfPNP_V66I_V73I_Y160F_0021584_0151_1 using gfam version 611 in slot 1 59 World Community Grid 23-5-2012 19:02:10 Restarting task GFAM_x3phb_hPNP_0021600_0001_0 using gfam version 611 in slot 7 60 World Community Grid 23-5-2012 19:02:10 Restarting task cfsw_2108_02108819_0 using cfsw version 605 in slot 3 61 World Community Grid 23-5-2012 19:02:10 Restarting task cfsw_2109_02109558_0 using cfsw version 605 in slot 5 62 World Community Grid 23-5-2012 19:02:10 Restarting task GFAM_x3phb_hPNP_0021601_0128_1 using gfam version 611 in slot 2 63 World Community Grid 23-5-2012 19:02:10 Restarting task GFAM_x3phb_hPNP_0021638_0103_1 using gfam version 611 in slot 0 64 World Community Grid 23-5-2012 19:02:10 Restarting task cfsw_2126_02126321_0 using cfsw version 605 in slot 6 This is what tells us about the base line from which we can correlate many things going wrong afterwards. Your error lines are to me [as an IIRC] a 'first reports' at the WCG forums. Don't know what MJ12 is. For sure, I have BOINC start delayed by a minute [setting in cc_config.xml], so everything else, that I value as Thé computer user comes up first, but please post your startup lines as sampled above. This is a straight crunching environment, or are there any virtual memory setups also running, and what is your security software? --//-- |
||
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges:
|
OK Sorry I misunderstood.
----------------------------------------Thu 24 May 2012 13:18:56 BST | | No config file found - using defaults Apart from anything else the first line is interesting EDIT: No security on this rig, it only crunches and crawls. The only other memory related system on this rig is a 4GB Ramdisk which initiates when the rig boots. It has been in situ throughout. Crawling with MJ12 is an internet search bot but it does not use all of the rig resources so I put WCG on here too to rectify that. Further edit: I start Boinc(manager) manually ![]() [Edit 2 times, last edit by OldChap at May 24, 2012 12:38:18 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Not happy bout your 7.0.22 [test] client version use, so fetch a stable and nearly finished bug fixed version at ppa:pkg-boinc/testing. Got this 7.0.27 running on Natty with kernel 3.0.0-19, where you're on 3.0.0.12
----------------------------------------Whilst, the rest of your log looks fine. 9.6GB for 4 threads can do 4 CEP2 concurrently (provided the rest of the sub-systems is fast too]. CFSW is a 'testing' science too. Propose you select another light one such as HCC to see if these incur same problem or if it's CFSW isolated, before and after upgrading to latest available BOINC build if problem persists. (There was a concurrency issue with CFSW on Linux, but that bug was fixed with last science version update) Whilst, a quick Google hits upon code at developers which has that message as a standard BOINC error trap: http://boinc.berkeley.edu/svn/trunk/boinc/client/app_start.cpp. Let us know what you find. --//-- P.S. Whatever MJ12 'crawls' I've got ClamAV running to make sure nothing unwanted get's 'hosted'. edit: Guess this is what it is: http://www.majestic12.co.uk/ [Edit 1 times, last edit by Former Member at May 24, 2012 1:05:40 PM] |
||
|
|
Oj101
Cruncher South Africa Joined: Oct 28, 2009 Post Count: 15 Status: Offline Project Badges:
|
When I built the box the "stable" version would not run hence getting the later beta with that issue fixed. I guess an update would help :)
----------------------------------------Seems to run well unless I stop it for some reason. Yep, That's MJ12 It collects website info and compresses it prior to sending to database. I may have to take a look at that ClamAV Thank you for your time and patience. I will let you know if this is resolved. EDIT: OOPS! just realised what I did there....posted from the rig I am helping a friend on [Edit 1 times, last edit by Oj101 at May 24, 2012 10:28:46 PM] |
||
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges:
|
Replaced Boinc version with 7.0.25, restarted and... voilà
----------------------------------------Not sure why it worked ok before and then, in the interim, had these problems but certainly this has seemed to fix things. I will wait a week then re-boot the rig to see what happens then. Thanks again SekeRob So far so good. ![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello OldChap,
Shared memory was not an uncommon error in 2005 when I first began studying the BOINC forum to learn problem solutions, but WCG might not have had it. I cannot remember problems with it after 2006, so my impression is that it has been a solved problem in BOINC. Maybe it showed up in 7.0.22 and disappeared in 7.0.23? Lawrence |
||
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges:
|
Hmmmmm. Another step along and I had to re-boot the box (for unrelated reasons) Result: the problem came back :(
----------------------------------------Then in a fit of "I wonder if..." I stopped manager and restarted in admin mode. Problem gone again :) Yet to prove this is it but at first glance it could be. I will test both ways to prove this at the weekend This probably only proves that I too am far from being a Linux adept ![]() |
||
|
|
|