For some time Boinc after a day or two has been using all the memory (16GB) and a good part of the swap and my machine therefore becomes very slow, I participate in Einstein@home and LHC@home, how to solve the problem?

ven. 13 août 2021 09:44:59 |  | Starting BOINC client version 7.16.6 for x86_64-pc-linux-gnu
ven. 13 août 2021 09:44:59 |  | log flags: file_xfer, sched_ops, task
ven. 13 août 2021 09:44:59 |  | Libraries: libcurl/7.68.0 OpenSSL/1.1.1f zlib/1.2.11 brotli/1.0.7 libidn2/2.2.0 libpsl/0.21.0 (+libidn2/2.2.0) libssh/0.9.3/openssl/zlib nghttp2/1.40.0 librtmp/2.3
ven. 13 août 2021 09:44:59 |  | Data directory: /var/lib/boinc-client
ven. 13 août 2021 09:44:59 |  | No usable GPUs found
ven. 13 août 2021 09:45:04 |  | libc: Ubuntu GLIBC 2.31-0ubuntu9.2 version 2.31
ven. 13 août 2021 09:45:04 |  | Host name: jupiter
ven. 13 août 2021 09:45:04 |  | Processor: 4 GenuineIntel Intel(R) Core(TM) i5-7500 CPU @ 3.40GHz [Family 6 Model 158 Stepping 9]
ven. 13 août 2021 09:45:04 |  | 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 ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d
ven. 13 août 2021 09:45:04 |  | OS: Linux Ubuntu: Ubuntu 20.04.2 LTS [5.4.0-81-generic|libc 2.31 (Ubuntu GLIBC 2.31-0ubuntu9.2)]
ven. 13 août 2021 09:45:04 |  | Memory: 15.53 GB physical, 11.99 GB virtual
ven. 13 août 2021 09:45:04 |  | Disk: 906.41 GB total, 204.55 GB free
ven. 13 août 2021 09:45:04 |  | Local time is UTC +2 hours
ven. 13 août 2021 09:45:04 |  | VirtualBox version: 6.1.22_Ubuntur144080
ven. 13 août 2021 09:45:04 |  | Config: GUI RPCs allowed from:
ven. 13 août 2021 09:45:04 |  | 127.0.0.1
ven. 13 août 2021 09:45:05 | LHC@home | General prefs: from LHC@home (last modified 08-Aug-2021 22:54:37)
ven. 13 août 2021 09:45:05 | LHC@home | Computer location: home
ven. 13 août 2021 09:45:05 |  | General prefs: using separate prefs for home
ven. 13 août 2021 09:45:05 |  | Reading preferences override file
ven. 13 août 2021 09:45:05 |  | Preferences:
ven. 13 août 2021 09:45:05 |  | max memory usage when active: 12721.37 MB
ven. 13 août 2021 09:45:05 |  | max memory usage when idle: 14311.54 MB
ven. 13 août 2021 09:45:06 |  | max disk usage: 100.00 GB
ven. 13 août 2021 09:45:06 |  | (to change preferences, visit a project web site or select Preferences in the Manager)
ven. 13 août 2021 09:45:06 |  | Setting up project and slot directories
ven. 13 août 2021 09:45:06 |  | Checking active tasks
ven. 13 août 2021 09:45:06 | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12887977; resource share 100
ven. 13 août 2021 09:45:06 | LHC@home | URL https://lhcathome.cern.ch/lhcathome/; Computer ID 10691960; resource share 100
ven. 13 août 2021 09:45:06 |  | Setting up GUI RPC socket
ven. 13 août 2021 09:45:06 |  | gui_rpc_auth.cfg is empty - no GUI RPC password protection
ven. 13 août 2021 09:45:06 |  | Checking presence of 343 project files

I use Boinc v 7.16.6 x86_64-pc-linux-gnu

Thanks in advance for your help.

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

Either run fewer tasks or get more RAM is the simple answer.

What are you running at the same time as BOINC if anything? LHC@home shouldn't require more than 3GB/task as far as I can make out from their help pages. Einstein is giving me a page not found error when I try and go to their FAQ page or forums so can't comment on their requirements. While not what I found on their pages, I thought LHC had had some tasks that required upwards of 5GB/task in the past so with only 16 for the OS, anything else yo are running and BOINC, that really isn't very much. I am currently regretting only getting 32GB on my new(ish) machine.

Edit got into the Einstein Forums via a different link.

Thank you for your answer, but what is curious is that I had not noticed this problem before for a year, on the other hand if I run "boinc-client restart" the memory use decreases but goes up gradually ...

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

Then check what is using that memory. It shouldn't be BOINC itself as it doesn't do any of the calculations, it's just a managing program.

Thank you for your answer, but what is curious is that I had not noticed this problem before for a year, on the other hand if I run "boinc-client restart" the memory use decreases but goes up gradually ...
Sam.

In a terminal run

top

You will be able to see what processes are running and what percentage of memory they are using in real time. This should let you see which of the two projects you are running (if any) is the culprit.

I always have a terminal with a top to monitor CPU and Memory usage.

top - 13:25:07 up  3:41,  1 user,  load average: 6,57, 6,34, 6,48
Tâches: 321 total,   6 en cours, 314 en veille,   0 arrêté,   1 zombie
%Cpu(s):  5,7 ut,  2,2 sy, 91,9 ni,  0,0 id,  0,0 wa,  0,0 hi,  0,2 si,  0,0 st
MiB Mem :  15901,7 total,   3101,0 libr,  10952,3 util,   1848,4 tamp/cache
MiB Éch:  12275,0 total,  10514,3 libr,   1760,7 util.   4184,9 dispo Mem 

    PID UTIL.     PR  NI    VIRT    RES    SHR S  %CPU  %MEM    TEMPS+ COM.                                                                           
 177577 boinc     39  19 2107020   2,0g   6352 R  86,7  12,9  51:07.42 einstein_O2MD1_                                                                
 177520 boinc     39  19 2105884   2,0g   6352 R  82,7  12,9  51:05.50 einstein_O2MD1_                                                                
 177539 boinc     39  19 2096484   2,0g   6352 R  69,8  12,9  51:25.24 einstein_O2MD1_                                                                
 177482 boinc     39  19 2200952   2,1g   6352 R  67,8  13,5  51:10.66 einstein_O2MD1_                                                                
   4816 samp      39  19 1617448  32296   8536 S  53,2   0,2 132:05.25 tracker-extract                                                                
  11865 samp      20   0   21,3g 138076  90764 S  12,3   0,8   5:16.17 chrome                                                                         
 232668 root      20   0       0      0      0 Z   6,6   0,0   0:00.20 /usr/local/shar                                                                
   4896 samp      20   0  946508  76096  58068 R   3,7   0,5   4:49.14 Xorg                    

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

The last column shows the actual process taking up the memory, which in this case is an Einstein application (times 4). If these are the ones taking up all the memory, then you'll have to report this to the Einstein project https://einsteinathome.org/community/forum

The last column shows the actual process taking up the memory, which in this case is an Einstein application (times 4). If these are the ones taking up all the memory, then you'll have to report this to the Einstein project https://einsteinathome.org/community/forum

They use on average 13% of memory each for the moment it is stable after restarting boinc-client, I will report it to the Einstein project.

Thanks again !

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

LHC: Atlas tasks can use over 10 GB memory/task depending on how you have configured them to run. The minimum is 3900 MB per Atlas tasks if you run them with 1 CPU core/task. The memory consumption formula is 3000 + (900 x n_cores) for Atlas task. CMS tasks use about 2900 MB per task and Theory about 700 MB per task.

I have the same problem with LHC@home ATLAS Simulation and VBoxHeadLess on startup the memory usage goes up to 95% I have to run "boinc-client restart" to get the memory usage back to normal.

I don't understand where the problem comes from and how to solve it.

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

In fact I have the problem with each change of project...

For the moment I will have to stop using Boinc on my machine

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

What is your setup for Atlas tasks? How many CPU cores are you using per task? This can be setup on LHC web site or by an app_config.xml. Are you using VirtualBox or are you running Atlas tasks as native linux tasks? On LHC website project preferences you can also limit how many tasks you have on your computer at one time. On Boinc preferences (locally with Boinc manager or on project website) you can limit how many CPU cores you allow Boinc to use at any time.

What is your setup for Atlas tasks? How many CPU cores are you using per task? This can be setup on LHC web site or by an app_config.xml. Are you using VirtualBox or are you running Atlas tasks as native linux tasks? On LHC website project preferences you can also limit how many tasks you have on your computer at one time. On Boinc preferences (locally with Boinc manager or on project website) you can limit how many CPU cores you allow Boinc to use at any time.

The problem occurs whatever the project, I have the same thing with Einstein@home the only way to correct the problem is to run "boinc-client restart" when it happens ...

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

Because that exits the tasks out of memory and restarts them, possibly from the beginning.

Because that exits the tasks out of memory and restarts them, possibly from the beginning.

Apparently they start again where they were stopped...

The problem is that it must be done automatically at each change of project, and that I do not know how to do...

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

You shouldn't need to do anything like that, as with each change of project applications are swapped out of memory and new ones start. At least at the default setting of "Leave applications in memory" set to No.

Even applications within the same project, when tasks end and new ones start, they do so with their own science application, leaving memory and starting anew.

You shouldn't need to do anything like that, as with each change of project applications are swapped out of memory and new ones start. At least at the default setting of "Leave applications in memory" set to No.

Even applications within the same project, when tasks end and new ones start, they do so with their own science application, leaving memory and starting anew.

Ok, I modified this parameter in the preferences of each project I restart Boinc to see. I am at 80% memory usage for the Boinc projects, I ordered 16GB of additional memory.

Thanks again for your help.

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

You shouldn't need to do anything like that, as with each change of project applications are swapped out of memory and new ones start. At least at the default setting of "Leave applications in memory" set to No.

Even applications within the same project, when tasks end and new ones start, they do so with their own science application, leaving memory and starting anew.

Apparently this seems to solve the memory usage problem, after several task and project changes.

Thanks a lot !

Sam.

---

Powered by Debian & Ubuntu 20.04 LTS
Boinc version 7.16.6 x86_64-pc-linux-gnu

Boinc doesn't use that much memory.
It's more than likely a bunch of WUs that are paused, or pending (switched with others after crunching data for long amounts of time) still remnant in memory.
It's really project issues, not a boinc issue.
Perhaps selecting to not keep GPU projects in memory, might alleviate some of the issue?

You shouldn't need to do anything like that, as with each change of project applications are swapped out of memory and new ones start. At least at the default setting of "Leave applications in memory" set to No.

Some projects (CPDN) being the one I know from experience, give a much higher failure rate if this option is not selected. Because of this, I rarely have tasks suspended after they have started running. My one exception is if work comes along for CPDN testing branch especially if it is a batch were quick answers are wanted. I realise that my way of running BOINC doesn't fit well for those who want to run several different projects at once.