I am also getting the CL_MEM_OBJECT_ALLOCATION_FAILURE errors, running the VelaJr tasks. NVidia GeForce GTX 1050 with 2GB RAM. This is the host: https://einsteinathome.org/host/12599623
The errors started at the same time as the last Windows Update (April 15), so I thought that might have something to do with it. I guess not, though?
I have just updated the driver for my GPU, in case that caused it, but that didn't fix it. I am now going to deselect this project for a while so that this computer can get back to work.
your gpu does not have enough ram to run these kinds of GW tasks.
you should be able to run the gamma ray pulsar tasks though if you want to keep contributing.
I am trying to add a new rig to E@H but I have forgotten my password. When selecting "forgot password" at the registration screen I get "Page not found". Is there some other process to recover a forgotten password?
You are logged in and posting here so I would think your browser must know your password.
Can you get your browser into a "show passwords" mode and find it that way?
If that doesn't immediately help you, a better place to post about it is in a thread with a similar issue, several of which can be found in the "Getting Started" forum.
I can't help as I don't know anything about how current procedures work. I joined in 2005 and I still use those details. Things are bound to be quite different these days. I believe the "authenticator" mechanism mentioned in the link should be a working method. You have existing hosts so you have an authenticator.
The scheduler is still wildly sending too large tasks to 2GB Nvidia cards that are not able to process them. I wonder if there's anything else on the server side that could be done with this situation.
I'm not trying to process GW tasks on a GPU with only 2GB ram. I was just wondering if this current situation is the final solution how things will be handled. If I browse through WUs I can see dozens and dozens of users receiving tasks that their computers then fail to crunch because they have 2GB (Nvidia) cards.
Unless the Einstein admins rewrite their server configuration files for a NOT simple detection algorithm to exclude 2GB cards, that is unlikely to happen. Hosts will continue to bomb out tasks until the owner looks at their host and decides to remove the low RAM card from this project or upgrades to a more capable card.
I had this effect already some weeks ago. I always could see that the handling was stopped with "error" after some pre-acticivity. The handling started and the progress counter was counting up to 10 or 11 %: The handling broke immediately after the counter had started again with 0.2 %.
My work around was to reboot windows after every calculation. It seems that only consecutive calculations can fail.
I also seems that not all signal records are affected. Meanwhile I had received many data sets which could be handled properly one after the other.
Here still my today's starting log.
08.05.2020 08:00:59 | | cc_config.xml not found - using defaults
08.05.2020 08:00:59 | | Starting BOINC client version 7.16.5 for windows_x86_64
08.05.2020 08:00:59 | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2s zlib/1.2.8
08.05.2020 08:00:59 | | Data directory: C:\ProgramData\BOINC
08.05.2020 08:00:59 | | Running under account winfried
08.05.2020 08:01:00 | | CUDA: NVIDIA GPU 0: GeForce GTX 1050 (driver version 441.93, CUDA version 10.2, compute capability 6.1, 2048MB, 1648MB available, 1911 GFLOPS peak)
08.05.2020 08:01:00 | | OpenCL: NVIDIA GPU 0: GeForce GTX 1050 (driver version 441.93, device version OpenCL 1.2 CUDA, 2048MB, 1648MB available, 1911 GFLOPS peak)
08.05.2020 08:01:00 | | OpenCL: Intel GPU 0: Intel(R) HD Graphics 630 (driver version 26.20.100.7325, device version OpenCL 2.1 NEO, 3231MB, 3231MB available, 184 GFLOPS peak)
08.05.2020 08:01:00 | | OpenCL CPU: Intel(R) Core(TM) i5-7300HQ CPU @ 2.50GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 7.6.0.0814, device version OpenCL 2.1 (Build 0))
08.05.2020 08:01:00 | | Windows processor group 0: 4 processors
08.05.2020 08:01:00 | | Host name: DESKTOP-R9HEUHO
08.05.2020 08:01:00 | | Processor: 4 GenuineIntel Intel(R) Core(TM) i5-7300HQ CPU @ 2.50GHz [Family 6 Model 158 Stepping 9]
08.05.2020 08:01:00 | | 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 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx tm2 pbe fsgsbase bmi1 smep bmi2
08.05.2020 08:01:00 | | OS: Microsoft Windows 10: Core x64 Edition, (10.00.18363.00)
08.05.2020 08:01:00 | | Memory: 7.89 GB physical, 12.89 GB virtual
08.05.2020 08:01:00 | | Disk: 224.63 GB total, 162.56 GB free
08.05.2020 08:01:00 | | Local time is UTC +2 hours
08.05.2020 08:01:00 | | No WSL found.
08.05.2020 08:01:00 | Einstein@Home | General prefs: from Einstein@Home (last modified 18-Mar-2017 21:32:56)
08.05.2020 08:01:00 | Einstein@Home | Host location: none
08.05.2020 08:01:00 | Einstein@Home | General prefs: using your defaults
08.05.2020 08:01:00 | | Reading preferences override file
08.05.2020 08:01:00 | | Preferences:
08.05.2020 08:01:00 | | max memory usage when active: 7269.32 MB
08.05.2020 08:01:00 | | max memory usage when idle: 7269.32 MB
08.05.2020 08:01:01 | | max disk usage: 162.91 GB
08.05.2020 08:01:01 | | max CPUs used: 1
08.05.2020 08:01:01 | | (to change preferences, visit a project web site or select Preferences in the Manager)
08.05.2020 08:01:01 | | Setting up project and slot directories
08.05.2020 08:01:01 | | Checking active tasks
08.05.2020 08:01:01 | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12811790; resource share 100
08.05.2020 08:01:01 | | Setting up GUI RPC socket
08.05.2020 08:01:01 | | Checking presence of 393 project files
08.05.2020 08:28:18 | Einstein@Home | Project requested delay of 60 seconds
08.05.2020 08:42:26 | Einstein@Home | update requested by user
08.05.2020 08:45:10 | | Project communication failed: attempting access to reference site
08.05.2020 08:45:11 | | Internet access OK - project servers may be temporarily down.
A GTX 1050 only has 2GB of GPU memory. This is not enough for the current batch of tasks in most cases. You should stop GW processing and do gamma ray tasks only on that GPU.
or replace the GPU with one with at least 4GB of memory.
D_S_Spence wrote:I am also
)
your gpu does not have enough ram to run these kinds of GW tasks.
you should be able to run the gamma ray pulsar tasks though if you want to keep contributing.
_________________________________________________________________________
hi guys, I am trying to add
)
hi guys,
I am trying to add a new rig to E@H but I have forgotten my password. When selecting "forgot password" at the registration screen I get "Page not found". Is there some other process to recover a forgotten password?
Thanks
Stephen
You are logged in and posting
)
You are logged in and posting here so I would think your browser must know your password.
Can you get your browser into a "show passwords" mode and find it that way?
If that doesn't immediately help you, a better place to post about it is in a thread with a similar issue, several of which can be found in the "Getting Started" forum.
I can't help as I don't know anything about how current procedures work. I joined in 2005 and I still use those details. Things are bound to be quite different these days. I believe the "authenticator" mechanism mentioned in the link should be a working method. You have existing hosts so you have an authenticator.
Cheers,
Gary.
@Stephen YMMV but in
)
@Stephen
YMMV but in Firefox... either click on [Edit] or the "hamburger" icon, ie. 3 horizontal lines, and select [Preferences].
Then -> Privacy and Security... then scroll down to "Logins and Passwords"... then -> [Saved Logins].
Other browsers no doubt differ but something like this.
The scheduler is still wildly
)
The scheduler is still wildly sending too large tasks to 2GB Nvidia cards that are not able to process them. I wonder if there's anything else on the server side that could be done with this situation.
you should stop trying to
)
you should stop trying to process GW tasks on a GPU with only 2GB ram, or upgrade the GPU for something with at least 4GB
_________________________________________________________________________
I'm not trying to process GW
)
I'm not trying to process GW tasks on a GPU with only 2GB ram. I was just wondering if this current situation is the final solution how things will be handled. If I browse through WUs I can see dozens and dozens of users receiving tasks that their computers then fail to crunch because they have 2GB (Nvidia) cards.
Unless the Einstein admins
)
Unless the Einstein admins rewrite their server configuration files for a NOT simple detection algorithm to exclude 2GB cards, that is unlikely to happen. Hosts will continue to bomb out tasks until the owner looks at their host and decides to remove the low RAM card from this project or upgrades to a more capable card.
Hi, today, I got again 21
)
Hi,
today, I got again 21 failing calculations of grav waves searchesr unning on a
NVIDIA GPU GeForce GTX 1050
For eample this one:
https://einsteinathome.org/de/task/945725001
I had this effect already some weeks ago. I always could see that the handling was stopped with "error" after some pre-acticivity. The handling started and the progress counter was counting up to 10 or 11 %: The handling broke immediately after the counter had started again with 0.2 %.
My work around was to reboot windows after every calculation. It seems that only consecutive calculations can fail.
I also seems that not all signal records are affected. Meanwhile I had received many data sets which could be handled properly one after the other.
Here still my today's starting log.
08.05.2020 08:00:59 | | cc_config.xml not found - using defaults
08.05.2020 08:00:59 | | Starting BOINC client version 7.16.5 for windows_x86_64
08.05.2020 08:00:59 | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2s zlib/1.2.8
08.05.2020 08:00:59 | | Data directory: C:\ProgramData\BOINC
08.05.2020 08:00:59 | | Running under account winfried
08.05.2020 08:01:00 | | CUDA: NVIDIA GPU 0: GeForce GTX 1050 (driver version 441.93, CUDA version 10.2, compute capability 6.1, 2048MB, 1648MB available, 1911 GFLOPS peak)
08.05.2020 08:01:00 | | OpenCL: NVIDIA GPU 0: GeForce GTX 1050 (driver version 441.93, device version OpenCL 1.2 CUDA, 2048MB, 1648MB available, 1911 GFLOPS peak)
08.05.2020 08:01:00 | | OpenCL: Intel GPU 0: Intel(R) HD Graphics 630 (driver version 26.20.100.7325, device version OpenCL 2.1 NEO, 3231MB, 3231MB available, 184 GFLOPS peak)
08.05.2020 08:01:00 | | OpenCL CPU: Intel(R) Core(TM) i5-7300HQ CPU @ 2.50GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 7.6.0.0814, device version OpenCL 2.1 (Build 0))
08.05.2020 08:01:00 | | Windows processor group 0: 4 processors
08.05.2020 08:01:00 | | Host name: DESKTOP-R9HEUHO
08.05.2020 08:01:00 | | Processor: 4 GenuineIntel Intel(R) Core(TM) i5-7300HQ CPU @ 2.50GHz [Family 6 Model 158 Stepping 9]
08.05.2020 08:01:00 | | 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 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx tm2 pbe fsgsbase bmi1 smep bmi2
08.05.2020 08:01:00 | | OS: Microsoft Windows 10: Core x64 Edition, (10.00.18363.00)
08.05.2020 08:01:00 | | Memory: 7.89 GB physical, 12.89 GB virtual
08.05.2020 08:01:00 | | Disk: 224.63 GB total, 162.56 GB free
08.05.2020 08:01:00 | | Local time is UTC +2 hours
08.05.2020 08:01:00 | | No WSL found.
08.05.2020 08:01:00 | Einstein@Home | General prefs: from Einstein@Home (last modified 18-Mar-2017 21:32:56)
08.05.2020 08:01:00 | Einstein@Home | Host location: none
08.05.2020 08:01:00 | Einstein@Home | General prefs: using your defaults
08.05.2020 08:01:00 | | Reading preferences override file
08.05.2020 08:01:00 | | Preferences:
08.05.2020 08:01:00 | | max memory usage when active: 7269.32 MB
08.05.2020 08:01:00 | | max memory usage when idle: 7269.32 MB
08.05.2020 08:01:01 | | max disk usage: 162.91 GB
08.05.2020 08:01:01 | | max CPUs used: 1
08.05.2020 08:01:01 | | (to change preferences, visit a project web site or select Preferences in the Manager)
08.05.2020 08:01:01 | | Setting up project and slot directories
08.05.2020 08:01:01 | | Checking active tasks
08.05.2020 08:01:01 | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12811790; resource share 100
08.05.2020 08:01:01 | | Setting up GUI RPC socket
08.05.2020 08:01:01 | | Checking presence of 393 project files
08.05.2020 08:28:18 | Einstein@Home | Project requested delay of 60 seconds
08.05.2020 08:42:26 | Einstein@Home | update requested by user
08.05.2020 08:45:10 | | Project communication failed: attempting access to reference site
08.05.2020 08:45:11 | | Internet access OK - project servers may be temporarily down.
A GTX 1050 only has 2GB of
)
A GTX 1050 only has 2GB of GPU memory. This is not enough for the current batch of tasks in most cases. You should stop GW processing and do gamma ray tasks only on that GPU.
or replace the GPU with one with at least 4GB of memory.
_________________________________________________________________________