I've got 20 2.6Ghz AMD 64 FX machines running, plus my home 3.2Ghz Quad Phenom II running 4 BOINC projects. At the moment Seti and ABC have no work for some reason, and both Einstien and Rosetta are telling me I've reached the quota at 23??? Doesn't make sense to have the machines running with nothing working??? How does one change this setting, or do I have to join another project...
Most machines still have work, but this one has run out, and is getting nothing from any of the projects.
Rosetta has plenty of work available, but don't know if this limit is based on machine, user, or what.

---

I've got 20 2.6Ghz AMD 64 FX machines running, plus my home 3.2Ghz Quad Phenom II running 4 BOINC projects. At the moment Seti and ABC have no work for some reason, and both Einstien and Rosetta are telling me I've reached the quota at 23??? Doesn't make sense to have the machines running with nothing working??? How does one change this setting, or do I have to join another project...
Most machines still have work, but this one has run out, and is getting nothing from any of the projects.
Rosetta has plenty of work available, but don't know if this limit is based on machine, user, or what.

I looked at the results of some of your machines, here is one of them. And it seems that allmost all of the tasks error with no computation on your machines at all.
All the tasks have this error message:

<core_client_version>6.10.17</core_client_version>
<![CDATA[
<message>
Can't write init file: -108
</message>
]]>

This is why the quota is lowered.
As this happens on all of your machines this suggests that there might be a probem with the way BOINC is installed on those machines, strange though that there are some succesfull results.

AdeB

---

I've got 20 2.6Ghz AMD 64 FX machines running, plus my home 3.2Ghz Quad Phenom II running 4 BOINC projects. At the moment Seti and ABC have no work for some reason, and both Einstien and Rosetta are telling me I've reached the quota at 23??? Doesn't make sense to have the machines running with nothing working??? How does one change this setting, or do I have to join another project...
Most machines still have work, but this one has run out, and is getting nothing from any of the projects.
Rosetta has plenty of work available, but don't know if this limit is based on machine, user, or what.

Are you perhaps overclocking, or trying to run multiple projects on one machine? If the later make sure you have the setting to keep units in memory set to YES for all projects! The setting can be found under Your Account, Computing Preferences and then in the top section processor usage you will see this line:
'Leave applications in memory while suspended?' Make sure it is set to YES for each project.

If you are overclocking you might want to back off a little bit, it can be fine for gaming or whatever but when trying to do math problems it can be very sensitive.

The quota is per machine. For each invalid result returned the quota is reduced by one until it reaches one workunit per day. This is to prevent machines from uselessly and endlessly downloading WUs they can't successfully complete. For each successfully validated result the daily quota is doubled until it reaches the project specific maximum. On Rosetta the maximum daily WU quota per CPU is 100.

Followed AdeB's link to find the exit status:

Exit status -185 (0xffffff47)

From the BOINC FAQ service:

ERR_RESULT_START -185

This is an error that will occur:
- if BOINC couldn't start the application
- if files are missing
- upon catch of other error returns
- on nonzero exit or signal
- if exceeded resource limit
- as catch-all for resume/start errors

Rebooting your machine will most of the times fix these.

Snags

I've got 20 2.6Ghz AMD 64 FX machines running, plus my home 3.2Ghz Quad Phenom II running 4 BOINC projects. At the moment Seti and ABC have no work for some reason, and both Einstien and Rosetta are telling me I've reached the quota at 23??? Doesn't make sense to have the machines running with nothing working??? How does one change this setting, or do I have to join another project...
Most machines still have work, but this one has run out, and is getting nothing from any of the projects.
Rosetta has plenty of work available, but don't know if this limit is based on machine, user, or what.

Are you perhaps overclocking, or trying to run multiple projects on one machine? If the later make sure you have the setting to keep units in memory set to YES for all projects! The setting can be found under Your Account, Computing Preferences and then in the top section processor usage you will see this line:
'Leave applications in memory while suspended?' Make sure it is set to YES for each project.

If you are overclocking you might want to back off a little bit, it can be fine for gaming or whatever but when trying to do math problems it can be very sensitive.

Thanks for the info. They recently swapped the computers in my lab to different machines that are now AMD 64 FX chips with ASUS A8V motherboards. I had to reinstall the BOINC on each machine. The setting was not to leave in memory, so I changed it. These motherboards do have a feature called AI Overclocking, and the default is Auto? Perhaps I need to change it to standard, which I assume would be no overclocking.

So, will see if making those changes to the setup and machines resolves the problems. Have noted that only 4 of the 20 machines were having this issue. Have noticed that some of these machines case fans are not working at normal speed.

Thanks for all the info.

---

I've got 20 2.6Ghz AMD 64 FX machines running, plus my home 3.2Ghz Quad Phenom II running 4 BOINC projects. At the moment Seti and ABC have no work for some reason, and both Einstien and Rosetta are telling me I've reached the quota at 23??? Doesn't make sense to have the machines running with nothing working??? How does one change this setting, or do I have to join another project...
Most machines still have work, but this one has run out, and is getting nothing from any of the projects.
Rosetta has plenty of work available, but don't know if this limit is based on machine, user, or what.

Are you perhaps overclocking, or trying to run multiple projects on one machine? If the later make sure you have the setting to keep units in memory set to YES for all projects! The setting can be found under Your Account, Computing Preferences and then in the top section processor usage you will see this line:
'Leave applications in memory while suspended?' Make sure it is set to YES for each project.

If you are overclocking you might want to back off a little bit, it can be fine for gaming or whatever but when trying to do math problems it can be very sensitive.

Thanks for the info. They recently swapped the computers in my lab to different machines that are now AMD 64 FX chips with ASUS A8V motherboards. I had to reinstall the BOINC on each machine. The setting was not to leave in memory, so I changed it. These motherboards do have a feature called AI Overclocking, and the default is Auto? Perhaps I need to change it to standard, which I assume would be no overclocking.

So, will see if making those changes to the setup and machines resolves the problems. Have noted that only 4 of the 20 machines were having this issue. Have noticed that some of these machines case fans are not working at normal speed.

Thanks for all the info.

I would not think the AI Overclocking is the problem, but it is probably okay to turn it off anyway. I think the problem is probably the fans and the pc overheating. Boinc makes the pc use 100% of the cpu, 100% of the time, that can make for a very warm pc and cause overheating issues.

I've got 20 2.6Ghz AMD 64 FX machines running, plus my home 3.2Ghz Quad Phenom II running 4 BOINC projects. At the moment Seti and ABC have no work for some reason, and both Einstien and Rosetta are telling me I've reached the quota at 23??? Doesn't make sense to have the machines running with nothing working??? How does one change this setting, or do I have to join another project...
Most machines still have work, but this one has run out, and is getting nothing from any of the projects.
Rosetta has plenty of work available, but don't know if this limit is based on machine, user, or what.

Are you perhaps overclocking, or trying to run multiple projects on one machine? If the later make sure you have the setting to keep units in memory set to YES for all projects! The setting can be found under Your Account, Computing Preferences and then in the top section processor usage you will see this line:
'Leave applications in memory while suspended?' Make sure it is set to YES for each project.

If you are overclocking you might want to back off a little bit, it can be fine for gaming or whatever but when trying to do math problems it can be very sensitive.

Thanks for the info. They recently swapped the computers in my lab to different machines that are now AMD 64 FX chips with ASUS A8V motherboards. I had to reinstall the BOINC on each machine. The setting was not to leave in memory, so I changed it. These motherboards do have a feature called AI Overclocking, and the default is Auto? Perhaps I need to change it to standard, which I assume would be no overclocking.

So, will see if making those changes to the setup and machines resolves the problems. Have noted that only 4 of the 20 machines were having this issue. Have noticed that some of these machines case fans are not working at normal speed.

Thanks for all the info.

I would not think the AI Overclocking is the problem, but it is probably okay to turn it off anyway. I think the problem is probably the fans and the pc overheating. Boinc makes the pc use 100% of the cpu, 100% of the time, that can make for a very warm pc and cause overheating issues.

I've set it to standard, and seem to see less of the errors, but still seeing some with various projects on different machines. I've just added the gnome-applet-sensors, and machines CPU temps vary from 40C to 58C, where the room AC is set to 24C. On one of the machine that was running at 58C, I found that stopping BOINC would drop the machine to 48C. Doing this remotely from Home, so will have to check machines later to see it this is linked to fan or other things. The CPU fans all seem to be fine, but case fans are the issue on some systems. Have changed the BOINC setting to 90% for the CPU, but don't see it actually adjusting CPU use?? My Home Phenom II Quad machine is running at 57C, but it has no errors on units. Only time I've had any issues is when I've seen a big power dip.

I had run a full memtest scan on machines, and no errors. These machines are Running the Fedora 12 64bit OS.

Thanks.

---

Have changed the BOINC setting to 90% for the CPU, but don't see it actually adjusting CPU use??

Do not do this!! Boinc has the setting but it really doesn't do what you want it to do, pc's have clock ticks that are used by the OS and programs to determine the priority a particular program will run at. For instance Boinc always uses 100% of the cpu when it runs, what you did was say run at 100% but only 90% of the time. So it runs full bore for 9 clock ticks then takes one clock tick off, then runs at 100% for 9 clock ticks, etc, etc. Boinc by default runs at the lowest priority of any program on the pc, that means that any other program running will bounce it from the next clock tick and Boinc will pick back up when it is done. So what you have done is make a change that can cause some serious problems, I highly recommend you change that back to 100%.

I had run a full memtest scan on machines, and no errors. These machines are Running the Fedora 12 64bit OS.
Thanks.

I don't know anything about Linux, not enough to help anyone anyway, but running the memtest was a good thing.

Have changed the BOINC setting to 90% for the CPU, but don't see it actually adjusting CPU use??

Do not do this!! Boinc has the setting but it really doesn't do what you want it to do, pc's have clock ticks that are used by the OS and programs to determine the priority a particular program will run at. For instance Boinc always uses 100% of the cpu when it runs, what you did was say run at 100% but only 90% of the time. So it runs full bore for 9 clock ticks then takes one clock tick off, then runs at 100% for 9 clock ticks, etc, etc. Boinc by default runs at the lowest priority of any program on the pc, that means that any other program running will bounce it from the next clock tick and Boinc will pick back up when it is done. So what you have done is make a change that can cause some serious problems, I highly recommend you change that back to 100%.

I had run a full memtest scan on machines, and no errors. These machines are Running the Fedora 12 64bit OS.
Thanks.

I don't know anything about Linux, not enough to help anyone anyway, but running the memtest was a good thing.

Thanks for the info. I just checked all 20 machines, and had no errors on any of them listed. I had just upgraded to the latest boinc, so perhaps something in that fixed the problem. The temps seem to be the same for the machines.

Will change the setting back to 100, and see if the problem stays away.

---

Have changed the BOINC setting to 90% for the CPU, but don't see it actually adjusting CPU use??

Do not do this!! Boinc has the setting but it really doesn't do what you want it to do, pc's have clock ticks that are used by the OS and programs to determine the priority a particular program will run at. For instance Boinc always uses 100% of the cpu when it runs, what you did was say run at 100% but only 90% of the time. So it runs full bore for 9 clock ticks then takes one clock tick off, then runs at 100% for 9 clock ticks, etc, etc. Boinc by default runs at the lowest priority of any program on the pc, that means that any other program running will bounce it from the next clock tick and Boinc will pick back up when it is done. So what you have done is make a change that can cause some serious problems, I highly recommend you change that back to 100%.

I had run a full memtest scan on machines, and no errors. These machines are Running the Fedora 12 64bit OS.
Thanks.

I don't know anything about Linux, not enough to help anyone anyway, but running the memtest was a good thing.

Thanks for the info. I just checked all 20 machines, and had no errors on any of them listed. I had just upgraded to the latest boinc, so perhaps something in that fixed the problem. The temps seem to be the same for the machines.

Will change the setting back to 100, and see if the problem stays away.

I am glad all is well again, good luck and keep on crunching!