After I installed the latest Seti/Lunatics for V7 & V8 I edited the app_info count settings for both V7 & V8 to allow 3 GPU work units simultaneously. Count went from stock <count>1</count> to <count>.33</count> for 3 at a time. I've started and re-started Boinc and also the computer. It still only runs 1 GPU unit at a time. There is no app_config.xml to conflict settings.
Did I miss something?
I installed to a Windows 7 Pro x64 system with a Bloomfield i960 processor, GTX770 GPU and 24GB RAM. Two other systems pending the Lunatics installer both Win 7 pro x64 with GTX670 and GTX470 GPUs. The 670 will also be set to 3 counts if allowed. I have run this way for the last few years without problems. Also where can I find info on setup and using app.config?
Thanks

---

Delta-V

Did you tell BOINC to Reread the Config files???

I would first Suspend the work; then Reread the Config files, and Resume work. This should implement your changes to run 3 WUs at a time. :-)


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

For app_info.xml, that has to be a full client (BOINC) restart.

Only app_ err, _config is read by read config files. (plus cc_config, of course, but that doesn't concern us here).

For app_info.xml, that has to be a full client (BOINC) restart.

Only app_ err, _config is read by read config files. (plus cc_config, of course, but that doesn't concern us here).

Thanks Richard. :-)

I learned something new... ;-)


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

There is nothing wrong with the app_info counts.
I did this on all 9 of my rigs yesterday.
You have to bring Boinc to a full stop and restart it.

The only other possible issue is that early on, there were simply not enough GPU tasks in my cache to run 3 at a time on all GPUs.
So, it could be possible that Lee was in that situation as well, but did not realize it.

But, of course, it's Boinc that reads the app_info, not the apps.
As long as the installer was pointed at the same location as in the past, there should be no issues with Boinc not finding and reading it. The updated apps have nothing to do with that. Boinc decides how many instances of 'app.exe' to run, not the app itself.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

I did the Option " Read config files" and followed it with "Read local perfs file". Here is the event log of the read functions:

1/23/2016 1:32:50 PM | | Re-reading cc_config.xml
1/23/2016 1:32:50 PM | | Not using a proxy
1/23/2016 1:32:50 PM | | log flags: sched_ops, task
1/23/2016 1:33:27 PM | SETI@home | General prefs: from SETI@home (last modified 08-Jun-2013 23:01:29)
1/23/2016 1:33:27 PM | SETI@home | Host location: none
1/23/2016 1:33:27 PM | SETI@home | General prefs: using your defaults
1/23/2016 1:33:27 PM | | Reading preferences override file
1/23/2016 1:33:27 PM | | Preferences:
1/23/2016 1:33:27 PM | | max memory usage when active: 14740.27MB
1/23/2016 1:33:27 PM | | max memory usage when idle: 24567.11MB
1/23/2016 1:33:27 PM | | max disk usage: 30.00GB
1/23/2016 1:33:27 PM | | max CPUs used: 7
1/23/2016 1:33:27 PM | | suspend work if non-BOINC CPU load exceeds 85%
1/23/2016 1:33:27 PM | | (to change preferences, visit a project web site or select Preferences in the Manager)

Here is the event log For a full shut down/restart of Boinc:

1/23/2016 1:41:43 PM | | Starting BOINC client version 7.6.9 for windows_x86_64
1/23/2016 1:41:43 PM | | log flags: sched_ops, task
1/23/2016 1:41:43 PM | | Libraries: libcurl/7.39.0 OpenSSL/1.0.2a zlib/1.2.8
1/23/2016 1:41:43 PM | | Data directory: C:\Program Files\BOINC\Data
1/23/2016 1:41:43 PM | | Running under account Admin
1/23/2016 1:41:44 PM | | CUDA: NVIDIA GPU 0: GeForce GTX 770 (driver version 359.06, CUDA version 7.5, compute capability 3.0, 2048MB, 1901MB available, 3332 GFLOPS peak)
1/23/2016 1:41:44 PM | | OpenCL: NVIDIA GPU 0: GeForce GTX 770 (driver version 359.06, device version OpenCL 1.2 CUDA, 2048MB, 1901MB available, 3332 GFLOPS peak)
1/23/2016 1:41:44 PM | SETI@home | Found app_info.xml; using anonymous platform
1/23/2016 1:41:44 PM | | Host name: Delta-V
1/23/2016 1:41:44 PM | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7 CPU 960 @ 3.20GHz [Family 6 Model 26 Stepping 5]
1/23/2016 1:41:44 PM | | 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 popcnt syscall nx lm vmx tm2 pbe
1/23/2016 1:41:44 PM | | OS: Microsoft Windows 7: Professional x64 Edition, Service Pack 1, (06.01.7601.00)
1/23/2016 1:41:44 PM | | Memory: 23.99 GB physical, 47.98 GB virtual
1/23/2016 1:41:44 PM | | Disk: 465.76 GB total, 350.46 GB free
1/23/2016 1:41:44 PM | | Local time is UTC -6 hours
1/23/2016 1:41:44 PM | | A new version of BOINC is available. (7.6.22) <a href=http://boinc.berkeley.edu/download.php>Download</a>
1/23/2016 1:41:44 PM | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 7552120; resource share 100
1/23/2016 1:41:44 PM | SETI@home | General prefs: from SETI@home (last modified 08-Jun-2013 23:01:29)
1/23/2016 1:41:44 PM | SETI@home | Host location: none
1/23/2016 1:41:44 PM | SETI@home | General prefs: using your defaults
1/23/2016 1:41:44 PM | | Reading preferences override file
1/23/2016 1:41:44 PM | | Preferences:
1/23/2016 1:41:44 PM | | max memory usage when active: 14740.27MB
1/23/2016 1:41:44 PM | | max memory usage when idle: 24567.11MB
1/23/2016 1:41:44 PM | | max disk usage: 30.00GB
1/23/2016 1:41:44 PM | | max CPUs used: 7
1/23/2016 1:41:44 PM | | suspend work if non-BOINC CPU load exceeds 85%
1/23/2016 1:41:44 PM | | (to change preferences, visit a project web site or select Preferences in the Manager)
1/23/2016 1:41:44 PM | | Not using a proxy

---

Delta-V

In the previous post the shut down included going to event viewer and closing Boinc Tray and Boinc.exe

---

Delta-V

What did the next section of the log say, after the re-start?

The section about starting or re-starting SETI tasks?

The section about starting or re-starting SETI tasks?

Unless he's got [cpu_sched] set, I don't think BOINC 7.6.9 is going to show task restarts. (One of my gripes with recent BOINC versions. :^))

---

The section about starting or re-starting SETI tasks?

Unless he's got [cpu_sched] set, I don't think BOINC 7.6.9 is going to show task restarts. (One of my gripes with recent BOINC versions. :^))

Damn. I'm used to that for resuming (from pre-emption, while BOINC stays running), but I'd forgotten about restarting too.

Lee, how are you determining what the number of active tasks per GPU is, after the re-start?

I never paid much attention to entries in log about active tasks because never had a problem like this before. I just shut down and restarted Boinc again but only entries beyond hardware and software loading referred to requesting work from server. Not short on cuda tasks either. I have many cuda50 tasks in the cache.

---

Delta-V

Problem Solved. I need to read instructions more thoroughly. I tried setting count at 0.33 and 0.30 but the only acceptable entry is 0.3. My bad. The 0.33 I tried first came from the program "Performance64"
"<maxAutoCount> 2 = 0.5, 3 = 0.33, 4 = 0.25 So fill in 1 or 2 or 3 or 4". I wonder if 0.25 still works in the new Lunatics?
Thanks for all the help

---

Delta-V

I tried setting count at 0.33 and 0.30 but the only acceptable entry is 0.3

?? - This just can't be true - unless there is strange bug in BOINC
Are you sure you edited ALL <count>1</count> (there are many) (unless you know you have tasks from only one <version_num> / <plan_class> combination)

I wonder if 0.25 still works in the new Lunatics?

This can't depend on any "new Lunatics"
- this value (and all of app_info.xml) is read and interpreted/used by BOINC, not by any app

---

 


- ALF - "Find out what you don't do well ..... then don't do it!" :)