I got Boinc View running on Windows XP and it seems to update fine from the client on Solaris 10. I can see progress with cpu time and percent done.
i don't understand.... it worked also better for me yesterday at 11:00 pm...
i saw the percentages goign forward...slowly..but forward :-)
and today..... the strange behaviour is back....
something is strange on my computer.... :-) but i don't know what...
the 2 e@h are still working, but since hours one is at 2.48% and one at 3.12%.... but they are running !!!!!
Currently, how many boinc clients do you have running? Are you still running two different instances of Boinc or are you running one? To get changes to the preferences on the webpage to update, you have to stop the Boinc client and restart it.
What are your preferences for 'general' and each project set to and how many projects do you currently have attached?
Currently, how many boinc clients do you have running? Are you still running two different instances of Boinc or are you running one? To get changes to the preferences on the webpage to update, you have to stop the Boinc client and restart it.
What are your preferences for 'general' and each project set to and how many projects do you currently have attached?
currently two boinc.... both with the same preferences...
this evening i changed the times preferences , because after a restart the wu was showing 22 hours and 0% !!!! only because of this break from 1 hour ...
i didn't kill the second boinc client yet, because one wu was working, but tomorrow, if this wu isn't at the end, i'll kill one ....
here my general preferences for this machine
Do work while computer is running on batteries? yes
Do work while computer is in use? yes
Do work only between the hours of (no restriction)
Leave applications in memory while preempted? no
Switch between applications every 10 minutes
On multiprocessors, use at most 1 processors
Disk and memory usage
Use no more than 100 GB disk space
Leave at least 0.001 GB disk space free
Use no more than 50% of total disk space
Write to disk at most every 10 seconds
Use no more than 25% of total virtual memory
Network usage
Connect to network about every 0.1 days
Confirm before connecting to Internet? no
Disconnect when done? no
Maximum download rate: no limit
Maximum upload rate: no limit
Use network only between the hours of (no restriction)
Skip image file verification? no
I think what is happening is your S@H and E@H applications are both running on each of your Boinc clients. That is if I understand what you are running currently. For the preferences, it takes the last update from a website when you restart Boinc. So, if you restarted Boinc after your general preferences were in, they should be just like you have down below. I have used 'Leave applications in memory while preempted? YES' before and it seems to be quicker at changing between E@H and S@H for instance.
Are the preferences for both S@H and E@H set to the same number under Resource?
Are you seeing any slowdown in your S@H processing? In your Boincview, when you have only your Sun computer selected, do you have a lot of jobs waiting for either project? Do you have your two versions of Boinc installed in different directories? The Oracle that is installed, is that idle most of the time along with any other applications on the machine?
I came across something in the forums that Boinc will attempt to catch up a project if the workunits are falling behind and won't make the deadline to report back. Search for EDF mode.
I think you should try to get your setup as simple as possible. Leave your preferences on the websites set just as they are. On one of your Boincs, detach from one project so you only have one active project like E@H. Run only one Boinc. I am hoping when you run it that way, we will be able to tell more
I think what is happening is your S@H and E@H applications are both running on each of your Boinc clients.
I have used 'Leave applications in memory while preempted? YES' before and it seems to be quicker at changing between E@H and S@H for instance.
yes, both projects were defined in both instances, but only E@H were active...
Quote:
Are the preferences for both S@H and E@H set to the same number under Resource?
yes it was....
Quote:
Do you have your two versions of Boinc installed in different directories?
yes it was...
Quote:
The Oracle that is installed, is that idle most of the time along with any other applications on the machine?
most of the time are the other applications idle, and oracle is only there for some queries.....
Quote:
I think you should try to get your setup as simple as possible. Leave your preferences on the websites set just as they are. On one of your Boincs, detach from one project so you only have one active project like E@H. Run only one Boinc. I am hoping when you run it that way, we will be able to tell more
now, i only have 1 boinc running, with only E@H defined, with the WU 555 working and the wu 554 waiting...
the preferences are....
Do work while computer is running on batteries? yes
Do work while computer is in use? yes
Do work only between the hours of (no restriction)
Leave applications in memory while preempted? yes
Switch between applications every 60 minutes
On multiprocessors, use at most 1 processors
Use no more than 100 GB disk space
Leave at least 0.001 GB disk space free
Use no more than 80% of total disk space
Write to disk at most every 10 seconds
Use no more than 25% of total virtual memory
Connect to network about every 0.1 days
Confirm before connecting to Internet? no
Disconnect when done? no
Maximum download rate: no limit
Maximum upload rate: no limit
Use network only between the hours of (no restriction)
Skip image file verification? no
and the machine is only at 50% of his resources...
You will have to let it run for at least two days that way. You may have two E@H workunits that it will switch between. This may have happened because of all the preference changes. Don't change anything for a week and just let it run. I don't expect you to see much more than 50% on processor usage since there are two processors. From your stats on the website, you take about 22 hours to process a workunit. I would just leave it alone for the week and just watch.
I created a new post about the processing times. Lets see if we get any answers.
ok... this wu 555 ended after 24 hours.... and i saw no move between yesterday and this "51:33 minutes CPU, and 3.61%" and the end....
this morning one hour befor the end....
now it is going the same way with the wu 554... (i'll wait a week before changing anything, but it seems to work as it worked with 2 proc. or 2 projects... that is, without moves on the "fraction_done"....)
wow.... now after 3 hours without move.... now it's growing.......
very very slowly...... but regulary growing....
the only thing i've just made, but i can't think that it had an effect on the E@H, was to apply the pfile Unix-Command to the einstein process.....
and now i see the percentages !!!!!! hourrahhhhh
i hope it will go on.....
i'll see with the next WU, tomorrow :-) ... if the same thing can be produced twice....
ok... today it didn't grow anymore.... and i tried to apply the unix-command pfiles on the process, and it wooooooorks !!!!!
i don't know why, but the client_xml file and the are only updated if i apply the pfile command......
now, i'll try to start the second boinc, with other preferences, and with only the einstein project on it, to see what will happen....
RE: I got Boinc View
)
i don't understand.... it worked also better for me yesterday at 11:00 pm...
i saw the percentages goign forward...slowly..but forward :-)
and today..... the strange behaviour is back....
something is strange on my computer.... :-) but i don't know what...
the 2 e@h are still working, but since hours one is at 2.48% and one at 3.12%.... but they are running !!!!!
strange...really strange....
PID USERNAME THR PR NCE SIZE RES STATE TIME FLTS CPU COMMAND
18900 pp 3 0 19 7184K 6496K run 118:29 0 44.48% albert_4.36_spa
24194 pp 3 0 19 7376K 5680K run 919:34 0 44.02% albert_4.36_spa
75 root 6 45 0 3952K 2736K sleep 301:35 0 0.09% picld
3088 pms 1 52 2 852M 808M sleep 0:00 0 0.09% oracle
3019 pms 1 52 2 852M 808M sleep 0:00 0 0.05% oracle
2904 oracle 1 58 0 853M 808M sleep 0:00 0 0.05% oracle
Currently, how many boinc
)
Currently, how many boinc clients do you have running? Are you still running two different instances of Boinc or are you running one? To get changes to the preferences on the webpage to update, you have to stop the Boinc client and restart it.
What are your preferences for 'general' and each project set to and how many projects do you currently have attached?
RE: Currently, how many
)
currently two boinc.... both with the same preferences...
this evening i changed the times preferences , because after a restart the wu was showing 22 hours and 0% !!!! only because of this break from 1 hour ...
i didn't kill the second boinc client yet, because one wu was working, but tomorrow, if this wu isn't at the end, i'll kill one ....
here my general preferences for this machine
Do work while computer is running on batteries? yes
Do work while computer is in use? yes
Do work only between the hours of (no restriction)
Leave applications in memory while preempted? no
Switch between applications every 10 minutes
On multiprocessors, use at most 1 processors
Disk and memory usage
Use no more than 100 GB disk space
Leave at least 0.001 GB disk space free
Use no more than 50% of total disk space
Write to disk at most every 10 seconds
Use no more than 25% of total virtual memory
Network usage
Connect to network about every 0.1 days
Confirm before connecting to Internet? no
Disconnect when done? no
Maximum download rate: no limit
Maximum upload rate: no limit
Use network only between the hours of (no restriction)
Skip image file verification? no
I think what is happening is
)
I think what is happening is your S@H and E@H applications are both running on each of your Boinc clients. That is if I understand what you are running currently. For the preferences, it takes the last update from a website when you restart Boinc. So, if you restarted Boinc after your general preferences were in, they should be just like you have down below. I have used 'Leave applications in memory while preempted? YES' before and it seems to be quicker at changing between E@H and S@H for instance.
Are the preferences for both S@H and E@H set to the same number under Resource?
Are you seeing any slowdown in your S@H processing? In your Boincview, when you have only your Sun computer selected, do you have a lot of jobs waiting for either project? Do you have your two versions of Boinc installed in different directories? The Oracle that is installed, is that idle most of the time along with any other applications on the machine?
I came across something in the forums that Boinc will attempt to catch up a project if the workunits are falling behind and won't make the deadline to report back. Search for EDF mode.
I think you should try to get your setup as simple as possible. Leave your preferences on the websites set just as they are. On one of your Boincs, detach from one project so you only have one active project like E@H. Run only one Boinc. I am hoping when you run it that way, we will be able to tell more
RE: I think what is
)
yes, both projects were defined in both instances, but only E@H were active...
yes it was....
yes it was...
most of the time are the other applications idle, and oracle is only there for some queries.....
now, i only have 1 boinc running, with only E@H defined, with the WU 555 working and the wu 554 waiting...
the preferences are....
Do work while computer is running on batteries? yes
Do work while computer is in use? yes
Do work only between the hours of (no restriction)
Leave applications in memory while preempted? yes
Switch between applications every 60 minutes
On multiprocessors, use at most 1 processors
Use no more than 100 GB disk space
Leave at least 0.001 GB disk space free
Use no more than 80% of total disk space
Write to disk at most every 10 seconds
Use no more than 25% of total virtual memory
Connect to network about every 0.1 days
Confirm before connecting to Internet? no
Disconnect when done? no
Maximum download rate: no limit
Maximum upload rate: no limit
Use network only between the hours of (no restriction)
Skip image file verification? no
and the machine is only at 50% of his resources...
load averages: 1.18, 1.12, 1.12 SUN-Fire 10:55:29
188 processes: 184 sleeping, 1 zombie, 1 stopped, 2 on cpu
CPU states: 36.9% idle, 55.7% user, 5.7% kernel, 1.8% iowait, 0.0% swap
Memory: 4.0G real, 1.5G free, 2.3G swap in use, 3.2G swap free
PID USERNAME THR PR NCE SIZE RES STATE TIME FLTS CPU COMMAND
14708 pp 3 0 19 8776K 8240K cpu00 24:22 0 48.73% albert_4.36_spa
75 root 6 45 0 3952K 2736K sleep 303:54 0 0.10% picld
17207 pp 1 58 0 2600K 1624K cpu01 0:00 0 0.06% top
now let's see what E@H will do.... it started at 10:30 am.... i will do some checks every hour....
ok, i could not see every
)
ok, i could not see every hour what happend, lol...
but after one hour the results were the same than after 5 hours...
51:33 minutes CPU, and 3.61% ...
the boinc is updated every 20 minutes... the differences are
pp@SUN-Fire ~/BOINC_4.43>diff client_state_prev.xml client_state.xml
26c26
0.976026
---
> 0.976037
28,29c28,29
0.971908
1139758246.100126
---
> 0.971921
> 1139759448.020443
and since this morning nothing moved in the directory
pp@SUN-Fire ~/BOINC_4.43/projects/einstein.phys.uwm.edu>lst
total 35606
drwxr-xr-x 3 pp staff 512 Feb 12 10:29 ..
-rw-rw-rw- 1 pp staff 2745667 Feb 12 10:29 earth_05_09
-rw-rw-rw- 1 pp staff 274843 Feb 12 10:29 sun_05_09
-rw-rw-rw- 1 pp staff 166 Feb 12 10:29 config_S4R2a.cfg
-rwxrwxrwx 1 pp staff 6829516 Feb 12 10:30 albert_4.36_sparc-sun-sola
ris2.7
-rw-rw-rw- 1 pp staff 266461 Feb 12 10:30 skygrid_0990_r_T07.dat
-rw-rw-rw- 1 pp staff 7368000 Feb 12 10:30 r1_0985.5
-rw-r--r-- 1 pp staff 670461 Feb 12 11:23 r1_0985.5__555_S4R2a_1_0
drwxr-xr-x 2 pp staff 512 Feb 12 11:23 .
You will have to let it run
)
You will have to let it run for at least two days that way. You may have two E@H workunits that it will switch between. This may have happened because of all the preference changes. Don't change anything for a week and just let it run. I don't expect you to see much more than 50% on processor usage since there are two processors. From your stats on the website, you take about 22 hours to process a workunit. I would just leave it alone for the week and just watch.
I created a new post about the processing times. Lets see if we get any answers.
http://einsteinathome.org/node/190763
ok... this wu 555 ended after
)
ok... this wu 555 ended after 24 hours.... and i saw no move between yesterday and this "51:33 minutes CPU, and 3.61%" and the end....
this morning one hour befor the end....
pp@SUN-Fire ~/BOINC_4.43>diff client_state_prev.xml client_state.xml
26c26
0.976574
---
> 0.976585
28,29c28,29
0.972551
1139818249.910196
---
> 0.972563
> 1139819451.360114
pp@SUN-Fire ~/BOINC_4.43/projects/einstein.phys.uwm.edu>lst
total 39702
drwxr-xr-x 3 pp staff 512 Feb 12 10:29 ..
-rw-rw-rw- 1 pp staff 2745667 Feb 12 10:29 earth_05_09
-rw-rw-rw- 1 pp staff 274843 Feb 12 10:29 sun_05_09
-rw-rw-rw- 1 pp staff 166 Feb 12 10:29 config_S4R2a.cfg
-rwxrwxrwx 1 pp staff 6829516 Feb 12 10:30 albert_4.36_sparc-sun-solaris2.7
-rw-rw-rw- 1 pp staff 266461 Feb 12 10:30 skygrid_0990_r_T07.dat
-rw-rw-rw- 1 pp staff 7368000 Feb 12 10:30 r1_0985.5
drwxr-xr-x 2 pp staff 512 Feb 12 11:23 .
-rw-r--r-- 1 pp staff 2767558 Feb 13 07:33 r1_0985.5__555_S4R2a_1_0
one hour later it was finished.... and here is the client_state file...
pp@SUN-Fire ~/BOINC_4.43>diff client_state_prev.xml client_state.xml
26c26
0.976609
---
> 0.976610
29c29
1139822134.700157
---
> 1139822144.750107
32,33c32,33
1540.82
392058
---
> 1540.77
> 392059
45c45
2
---
> 3
54c54
0.000000
---
> 1139822207.755213
58d57
193,217d191
r1_0985.5__555_S4R2a_1_0
151177.000000
6000000.000000
8e502cbdc5a936b62e4d73cebbfcebcb
0
http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/file_upload_handler
r1_0985.5__555_S4R2a_1_0
6000000
http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/file_upload_handler
272,303d245
r1_0985.5__555_S4R2a
albert
436
33227203661312.000000
166136018306560.000000
60000000.000000
100000000.000000
earth_05_09
earth
sun_05_09
sun
config_S4R2a.cfg
conf
r1_0985.5
data.sft
skygrid_0990_r_T07.dat
skygrid_0990_r_T07.dat
368,392d309
r1_0985.5__555_S4R2a_1
84556.920000
0
5
compactifying ... done.
r1_0985.5__555_S4R2a
1140946392.000000
r1_0985.5__555_S4R2a_1_0
Fstat.out
now it is going the same way with the wu 554... (i'll wait a week before changing anything, but it seems to work as it worked with 2 proc. or 2 projects... that is, without moves on the "fraction_done"....)
wow.... now after 3 hours
)
wow.... now after 3 hours without move.... now it's growing.......
very very slowly...... but regulary growing....
the only thing i've just made, but i can't think that it had an effect on the E@H, was to apply the pfile Unix-Command to the einstein process.....
and now i see the percentages !!!!!! hourrahhhhh
i hope it will go on.....
i'll see with the next WU, tomorrow :-) ... if the same thing can be produced twice....
ok... today it didn't grow
)
ok... today it didn't grow anymore.... and i tried to apply the unix-command pfiles on the process, and it wooooooorks !!!!!
i don't know why, but the client_xml file and the are only updated if i apply the pfile command......
now, i'll try to start the second boinc, with other preferences, and with only the einstein project on it, to see what will happen....
thank you wumpus for your help :-)