BT 1.25

Started by Pepo, October 26, 2011, 07:39:27 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Pepo

The changeset for BT 1.25 seems to ne promising :) going to be a wonderful release!
Hope this thread will be as short as possible ;)
Peter

Pepo

#1

  • I'm not able to collapse two BOINC's notices related to hosts from remote_hosts.cfg - the Notices pane will occasionally just flash as being refreshed, but the notices stay opened. According to [-] or [+] signs' URLs (in "file:///C:/Users/UserName/AppData/Roaming/eFMer/BoincTasks/html/__collapse__162" form), the mentioned two [-] signs get always incremented numbers, while all other notices' collapse/expand signs keep their number constant.
    I'll check it later on another machine with BT 1.24, but I've never notictd such problem before. (I always do close this particular notice ;D - just missing an [X] button next to it.)
  • "-Changed: Use <Shift + click> <Ctrl + Shift + click> <Shift + Home> or <Shift + End> to select a range in the views and graphics." In the various graph dialogs (Tasks, Stats, etc.), the <Shift> button is being ignored, while the "Multiple selection" checkbox is checked.
    However as it is possibly a simplified mode, the behavior can be considered correct.
  • Although TTh was updated to 5.40 at least one hour ago and BT is connected to it, it still announces its availability in the Notices (to be updated to from previous 5.30).
    This will possibly disappear later.
Peter

fred

I've seen the collapse problem once, but when you restart BT it should go away, I'm not sure why this happens, have to check.

The multiple selection option is the default Windows setting, this is how these things should work.
Removing the check and you get the behavioral that is more in line with the regular list.

Pepo

Quote from: fred on October 26, 2011, 11:31:17 AM
I've seen the collapse problem once, but when you restart BT it should go away, I'm not sure why this happens, have to check.
In my case it persists. On the other machine, three notices related to both remote_hosts.cfg and cc_config.xml. In BT 1.24 a few minutes ago they were closing just fine. I've also noticed that the trailing numbers in URLs are being updated all three at once (like 144+145+143 -> 147+148+146), regardless of which [-] is being clicked on.

As on TTh's Temp graph dialog, BT's temp graph has also these raised color boxes next to the Core/Max/Gpu checkboxes.

Interesting or weird, until BOINC client on a particular machine (e.g. localhost, not tried with remotes) does connect to BT, BT does not display the machine's temperatures, although it is connected to its TTh. If already all runs, restarting the client does not stop the temperature from being continuously  updated.
I think that the temperature might be displayed immediately after BT+TTh being connected. This could aid with troubleshooting while trying to connect to a remote BOINC client - when TTh is connected and temperatures get updated, then the network is just fine, problem is with authentication, etc.
Peter

Pepo

Quote from: Pepo on October 26, 2011, 01:31:54 PM
Quote from: fred on October 26, 2011, 11:31:17 AM
I've seen the collapse problem once, but when you restart BT it should go away, I'm not sure why this happens, have to check.
In my case it persists. On the other machine, three notices related to both remote_hosts.cfg and cc_config.xml. In BT 1.24 a few minutes ago they were closing just fine. I've also noticed that the trailing numbers in URLs are being updated all three at once (like 144+145+143 -> 147+148+146), regardless of which [-] is being clicked on.
With the same BT instance, I've updated BOINC 6.12.34 -> 6.12.41 - I can suddenly close these three notices. Perhaps their format changed slightly in between and BT 1.25 does not support the older format correctly? (A wild guess.)
Peter

Pepo


  • "-Add: Graph toolbar: Data transfer (BOINC V7)" - there is no BOINC V7 yet to try out, just the 6.13.x series, but BT 1.25 refuses to consider them being "V7". So, how to test it in advance?
Peter

fred

Quote from: Pepo on October 27, 2011, 12:05:45 AM

  • "-Add: Graph toolbar: Data transfer (BOINC V7)" - there is no BOINC V7 yet to try out, just the 6.13.x series, but BT 1.25 refuses to consider them being "V7". So, how to test it in advance?
6.13 is the V7 beta. As soon as the feature is implemented in a 6.13 version, I will implement it.

Pepo

Quote from: fred on October 27, 2011, 06:09:58 AM
Quote from: Pepo on October 27, 2011, 12:05:45 AM
"-Add: Graph toolbar: Data transfer (BOINC V7)" - there is no BOINC V7 yet to try out, just the 6.13.x series, but BT 1.25 refuses to consider them being "V7". So, how to test it in advance?
6.13 is the V7 beta. As soon as the feature is implemented in a 6.13 version, I will implement it.
:) I've thought it already is, just could not trigger it ;D
Peter

fred

Quote from: Pepo on October 27, 2011, 09:30:34 AM
:) I've thought it already is, just could not trigger it ;D
Can only do so much .. :o... V 1.26

Pepo

(Fred, I suspect you'll hate me and will try some killing woodoo on my doll :P)

Connected to local Win x64 6.13.9 client. After testing PrimeGrid + TTh, I've noticed that under the BOINC client process, there are two orphaned PrimeGrid tasks I've aborted and updated some 20 minutes ago. They were heavily using CPU, thus TThrottle was throttling all known valid payload to nearly zero. So far all correct.

I've aborted these 2 processes and then observed the throttle% graph, slowly coming from 99% upwards to maybe 40%. The remaining tasks' CPU usage slowly rose (observed in process monitoring tools), but BT somehow got out of order. These running tasks were suddenly not green-highlighted, their Status column described them as "New" or "New - Suspended by user", instead of "Running" or anything else. Throttle column did not show any number or bar. In Computers tab, the connection to TThrottle 5.40 seemed to be OK. In the Messages tab, there was all rosa-highlighted and in the message ID column, weird numbers flashing between zero and something huge were jumping around; date-time column contained just "--":

0 -- Starting BOINC client version 6.13.9 for windows_x86_64
0 -- This a development version of BOINC and may not function properly
825309449 -- Config: report completed tasks immediately
1633951847 -- Config: GUI RPC allowed from:
1633951847 -- Config:   pavilon6
778658670 -- Config:   vetroplach
778658670 -- Config:   localhost
[.....]
537542255 WUProp@Home -- [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
1920299838 PrimeGrid -- [checkpoint] result pps_llr_104435382_0 checkpointed
1647262730 WUProp@Home -- [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
537542255 PrimeGrid -- [checkpoint] result pps_llr_104435382_0 checkpointed
1869770812 surveill@home -- Computation for task wu_1319568902_127067_0 finished
1847617390 surveill@home -- Started upload of wu_1319568902_127067_0_data
537542249 surveill@home -- Started upload of wu_1319568902_127067_0_uris
0 surveill@home -- Finished upload of wu_1319568902_127067_0_data
0 surveill@home -- Finished upload of wu_1319568902_127067_0_uris
0 surveill@home -- Sending scheduler request: To report completed tasks.
0 surveill@home -- Reporting 1 completed tasks, requesting new tasks for CPU
0 surveill@home -- Scheduler request completed: got 0 new tasks
0 surveill@home -- Not sending work - max number of probes in your network exceeded
0 WUProp@Home -- [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
0 PrimeGrid -- [checkpoint] result pps_llr_104435382_0 checkpointed
0 WUProp@Home -- [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
0 PrimeGrid -- [checkpoint] result pps_llr_104435382_0 checkpointed
0 -- Suspending computation - user request
1701995878 WUProp@Home -- [cpu_sched] Preempting wu_v3_1319038646_293619_0 (left in memory)
0 PrimeGrid -- [cpu_sched] Preempting pps_llr_104435382_0 (left in memory)
0 -- Resuming computation
0 WUProp@Home -- [cpu_sched] Resuming wu_v3_1319038646_293619_0
896081961 PrimeGrid -- [cpu_sched] Resuming pps_llr_104435382_0
1701474162 PrimeGrid -- task pps_llr_104435382_0 resumed by user
0 WUProp@Home -- [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
0 PrimeGrid -- [checkpoint] result pps_llr_104435382_0 checkpointed


Some failed attempt to reconnect to the client?

5 minutes later all was back to normal. Here comes the correct log - matching lines:

1 27.10.11 10:45 Starting BOINC client version 6.13.9 for windows_x86_64
2 27.10.11 10:45 This a development version of BOINC and may not function properly
3 27.10.11 10:45 Config: report completed tasks immediately
4 27.10.11 10:45 Config: GUI RPC allowed from:
5 27.10.11 10:45 Config:   pavilon6
6 27.10.11 10:45 Config:   vetroplach
7 27.10.11 10:45 Config:   localhost
[.....]
1181 WUProp@Home 29.10.11 00:12 [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
1182 PrimeGrid 29.10.11 00:12 [checkpoint] result pps_llr_104435382_0 checkpointed
1183 WUProp@Home 29.10.11 00:13 [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
1184 PrimeGrid 29.10.11 00:13 [checkpoint] result pps_llr_104435382_0 checkpointed
1185 surveill@home 29.10.11 00:13 Computation for task wu_1319568902_127067_0 finished
1186 surveill@home 29.10.11 00:13 Started upload of wu_1319568902_127067_0_data
1187 surveill@home 29.10.11 00:13 Started upload of wu_1319568902_127067_0_uris
1188 surveill@home 29.10.11 00:13 Finished upload of wu_1319568902_127067_0_data
1189 surveill@home 29.10.11 00:13 Finished upload of wu_1319568902_127067_0_uris
1190 surveill@home 29.10.11 00:13 Sending scheduler request: To report completed tasks.
1191 surveill@home 29.10.11 00:13 Reporting 1 completed tasks, requesting new tasks for CPU
1192 surveill@home 29.10.11 00:13 Scheduler request completed: got 0 new tasks
1193 surveill@home 29.10.11 00:13 Not sending work - max number of probes in your network exceeded
1194 WUProp@Home 29.10.11 00:14 [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
1195 PrimeGrid 29.10.11 00:14 [checkpoint] result pps_llr_104435382_0 checkpointed
1196 WUProp@Home 29.10.11 00:15 [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
1197 PrimeGrid 29.10.11 00:15 [checkpoint] result pps_llr_104435382_0 checkpointed
1198 29.10.11 00:15 Suspending computation - user request
1199 WUProp@Home 29.10.11 00:15 [cpu_sched] Preempting wu_v3_1319038646_293619_0 (left in memory)
1200 PrimeGrid 29.10.11 00:15 [cpu_sched] Preempting pps_llr_104435382_0 (left in memory)
1201 29.10.11 00:15 Resuming computation
1202 WUProp@Home 29.10.11 00:15 [cpu_sched] Resuming wu_v3_1319038646_293619_0
1203 PrimeGrid 29.10.11 00:15 [cpu_sched] Resuming pps_llr_104435382_0
1204 PrimeGrid 29.10.11 00:15 task pps_llr_104435382_0 resumed by user
1205 WUProp@Home 29.10.11 00:16 [checkpoint] result wu_v3_1319038646_293619_0 checkpointed
1206 PrimeGrid 29.10.11 00:16 [checkpoint] result pps_llr_104435382_0 checkpointed


While in such weird state, in BOINC Manager everything seemed to be fine.
Peter

fred

0         --   Starting BOINC client version 6.13.9 for windows_x86_64   
0         --   This a development version of BOINC and may not function properly  :-X

You may want to try 6.13.10.

Almost certainly a client problem.

New is 0 also.

Pepo

#11
Quote from: fred on October 29, 2011, 09:04:14 AM
You may want to try 6.13.10. Almost certainly a client problem.
If indeed, then it is almost certainly not fixed enough in 6.13.10.

Or, did the client send such havoc to the BoincTasks? Happened once more. BT was pretty CPU-busy at the moment (well, a couple of minutes, until I've stopped BOINC client). I'm trying now 6.13.10 - with the PrimeGrid tests and disabling GPU, I've fullfilled 6.13.9's prerequisites for resetting my few local projects  ;D
Peter

idahofisherman

I have been having trouble with BT using to Much CPU (98%) since about BT 1.23.  I finally changed the history to  120 seconds from 15.  This seems to have fixed the cpu problem, but now I have increased "Missed" instead of Report OK. 

Is there a time limit for missed?


Pepo

Quote from: idahofisherman on November 03, 2011, 08:35:00 PM
I finally changed the history to  120 seconds from 15.  [...] but now I have increased "Missed" instead of Report OK. 
Is there a time limit for missed?
From my experience with nCi tasks, they are often completely reported (with already a new task downloaded and started) 4-7 seconds after finishing previous task. With 4+10 sec. setting, I have 70% OK and 30% OK* + Missed report states, with 2-5 sec. I get at least 90% OK + OK* and just an occasional complete miss.
Peter

3216842

Hello to all and congratulation for this great software.
I have a little annoying problem with BT: When stopping BONIC client via menu, BT comes to a little freeze and after ~15-20 sek. BT shows up an error message "The BOINC client couldn't be shut down". Further ~10 sek. later the client  and after this all running tasks/WUs stops (in this order) :-( .
This behavior i watched in BT 1.23/.24/.25. Running XP Pro SP3, BOINC 6.12.34, BOINC Manager is not running.

Log:
04 November 2011 - 12:18:00 Shut down BoincTasks ---- The BOINC client is shutting down
04 November 2011 - 12:18:30 Shut down BoincTasks ---- The BOINC client has shut down
04 November 2011 - 12:18:30 Shut down BoincTasks ---- Der BOINC-Client konnte nicht beendet werden
04 November 2011 - 12:18:30 Connect ---- The connection was lost, because the client stopped

Hope for a fix.

__W__