Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 
World Community Grid Forums
  Category: Support
  Forum: BOINC Agent Support
  Thread: Help needed: BOINC Agent crunches out its entire stash!

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 18
Posts: 18   Pages: 2   [ 1 2 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 1463 times and has 17 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Help needed: BOINC Agent crunches out its entire stash!
I am running BOINC agent on 2 machines under the same profile. One is XPSP2, the seond is Vista, both x86, and both have Intel Core Duo II CPUs. The profile specifies 2 days of work queueing.

One of these maintains the queue. The other also used to; but after it had been turned off for ~40 hours, its behavior changed significantly: the queue is empty, except for the 2 active work bits. The window shows 3 lines only if there is one waiting to be reported; 2 always active, but none are waiting (I could not spot when it downloads a slice of work, i.e. whether that happens slightly before or slightly after a calculation completes, but I have not seen even once that only only one task be running, or that any one task be waiting in the queue). Web stats page confirms that I have only 2 jobs allocated for that device, while there are many more for the device that queues its work properly.

Questions:
1. Is it possible to fix that by some trick?
2. If such a problem (and its solution) is unknown, I'll just completely reinstall the software. In this case, I want to completely remove its settings (registry, setting files, what else?). Where can I find instructions as to do that? I think some stat counter became corrupt, and the program end up with too small an estimate of how much it can do in 2 days. (I happen to know the OS and its utilities very well, so instruction in the form “check the value X under the regstry key Y,” while omitting the part how to start the regedit and navigate around in it, would be the best readable one for me!)

Will be awfully grateful for the answers to these Qs, as well as for any other relevant suggestions!
[Apr 26, 2008 11:54:18 PM]   Link   Report threatening or abusive post: please login first  Go to top 
retsof
Former Community Advisor
USA
Joined: Jul 31, 2005
Post Count: 6824
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
There's no need for a reinstall. It's just a scheduler foible.

Each machine maintains its own queue independently. After the download, turning one off for awhile (nearly for the queue length of two days) prodded it into action. It wasn't downloading more work because it was trying to finish what it had already once it resumed, to meet the deadline.

If you ask for 2 days of work and then turn the machine off for 2 days, what should it do?

If you know that you are going to be gone for awhile, you can hit the "no new tasks" button on the advanced view project tab 2 days ahead of time to drain the queue yourself.

An excessive queue length used to cause scheduler problems in the past with earlier BOINC versions, but two days is no problem at all.
----------------------------------------
SUPPORT ADVISOR
Work+GPU i7 8700 12threads
School i7 4770 8threads
Default+GPU Ryzen 7 3700X 16threads
Ryzen 7 3800X 16 threads
Ryzen 9 3900X 24threads
Home i7 3540M 4threads50%
----------------------------------------
[Edit 5 times, last edit by retsof at Apr 27, 2008 1:00:08 AM]
[Apr 27, 2008 12:50:30 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
Thank you very much for your reply, but I am afraid that I did not explain my problem properly, so the answer is missing the point. That's entirely my fault.

I mentioned that turning off the machine seems to have caused the behavior, but did not mention when it started. It was about 3 weeks ago.

My problem is not that the queue overflows — no, the reverse is true: there is no tasks waiting in the queue! By the way, I finally observed when a new job is downloaded: it is about 10 minutes before one of the 2 active jobs is about to complete.

So, in other words, the device is told to maintain a queue of ~2 days of work. Instead, it maintains the queue length of only 10 minutes. It does not sit without work, but any network hiccup will cause it to.
[Apr 27, 2008 6:55:05 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
I had this problem once and as far as I can recollect, I exited Boinc and restarted it which seemed to cure the problem.....
[Apr 27, 2008 7:06:41 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Sekerob
Ace Cruncher
Joined: Jul 24, 2005
Post Count: 20043
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
fregimus,

would you open up your file explorer and navigate to the BOINC data dir (the place is mentioned in your start up log of BOINC Message window) and open with notepad the client_state.xml. When done copy the following info and post it here:

<time_stats>
    <on_frac>0.963488</on_frac>
    <connected_frac>0.222719</connected_frac>
    <active_frac>0.999739</active_frac>
    <cpu_efficiency>0.953925</cpu_efficiency>
    <last_update>1209279971.875000</last_update>
</time_stats>

<project>
    <master_url>http://www.worldcommunitygrid.org/</master_url>
    <project_name>World Community Grid</project_name>
    <symstore></symstore>
    <user_name>Sekerob</user_name>
    <host_venue>work</host_venue>
    <short_term_debt>4203.775734</short_term_debt>
    <long_term_debt>474704.317209</long_term_debt>
    <resource_share>100.000000</resource_share>
    <duration_correction_factor>1.815018</duration_correction_factor>
    <sched_rpc_pending>0</sched_rpc_pending>
    <send_time_stats_log>0</send_time_stats_log>
    <send_job_log>0</send_job_log>
    <verify_files_on_app_start/>
    <ams_resource_share>0.000000</ams_resource_share>
    <scheduler_url>https://secure.worldcommunitygrid.org/boinc/wcg_cgi/fcgi</scheduler_url>
    <code_sign_key>
The values of our interest are <active_frac> and <duraction_correction_factors>

Mine is still upset over the miss estimated FightAIDS jobs that were twice as long so the duration_correction_factor, one of the work fetch controls is 1.8. It will adjust automatically, long as you keep crunching.

After, exit notepad WITHOUT saving the file!

ttyl
----------------------------------------
WCG Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All!
----------------------------------------
[Edit 1 times, last edit by Sekerob at Apr 27, 2008 7:25:46 AM]
[Apr 27, 2008 7:24:58 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
Thanks for your help Sekerob! Here are the 2 config elements from the failing host. I am posting them whole, except the last long paragraph of gibberish encoded binary value, that has been snipped, 'cause it is ugly. : )
<time_stats>
    <on_frac>0.936234</on_frac>
    <connected_frac>0.999980</connected_frac>
    <active_frac>0.999832</active_frac>
    <cpu_efficiency>0.986408</cpu_efficiency>
    <last_update>1209298353.968050</last_update>
</time_stats>
<project>
    <master_url>http://www.worldcommunitygrid.org/</master_url>
    <project_name>World Community Grid</project_name>
    <symstore></symstore>
    <user_name>fregimus</user_name>
    <team_name></team_name>
    <host_venue></host_venue>
    <email_hash>aaf5827d09d4eb31df2e3e25d1de9bd8</email_hash>
    <cross_project_id>82fe8eb8040cdd90c611a840669ca0ef</cross_project_id>
    <user_total_credit>296380.940280</user_total_credit>
    <user_expavg_credit>1375.116263</user_expavg_credit>
    <user_create_time>1180425546.000000</user_create_time>
    <rpc_seqno>4069</rpc_seqno>
    <hostid>200125</hostid>
    <host_total_credit>226571.460248</host_total_credit>
    <host_expavg_credit>683.274741</host_expavg_credit>
    <host_create_time>1180425897.000000</host_create_time>
    <nrpc_failures>0</nrpc_failures>
    <master_fetch_failures>2</master_fetch_failures>
    <min_rpc_time>1209289630.137749</min_rpc_time>
    <next_rpc_time>1209635169.137749</next_rpc_time>
    <short_term_debt>0.000000</short_term_debt>
    <long_term_debt>0.000000</long_term_debt>
    <resource_share>100.000000</resource_share>
    <duration_correction_factor>0.906428</duration_correction_factor>
    <sched_rpc_pending>0</sched_rpc_pending>
    <verify_files_on_app_start/>
    <ams_resource_share>0.000000</ams_resource_share>
    <scheduler_url>https://secure.worldcommunitygrid.org/boinc/wcg_cgi/fcgi</scheduler_url>
    <code_sign_key>
{{...snip...}}
</code_sign_key>
</project>
[Apr 27, 2008 1:40:49 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
All that looks completely normal to me.

It is possible that the work buffer isn't configured how you expect. Please will you post the contents of the sched_request_www.worldcommunitygrid.org.xml file?

Thank you.
[Apr 27, 2008 2:15:57 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
Thanks for looking. The file contents is below. Note the CPU is overclocked slightly (2.4 GHz really vs. 1.86 GHz as reported).

The question that I still have is, does it make sense to try to diagnose the problem? It is cheap enough, worktimewise, to block new tasks and, when current ones are finished, then reinstall the program. If the diagnostics we are engaging in is important for bug fixing etc., then I am all for it; but if our goal now is only to make my device working properly, then should we waste any more time on that, if a fixed time, inexpensive solution is available? : )

What if I run all pending tasks to completion, uninstall, kill all remaining xml files and then reinstall? It will take just 10 minutes of my work. We are already spending much more time collectively. Would my plan B be harmful? Unlikely to fix the problem? I just hate the idea of wasting your time.

<scheduler_request>
    <authenticator>fa97daaf79719cc1f0d925bab24b24c7</authenticator>
    <hostid>200125</hostid>
    <rpc_seqno>4070</rpc_seqno>
    <platform_name>windows_intelx86</platform_name>
    <core_client_major_version>5</core_client_major_version>
    <core_client_minor_version>8</core_client_minor_version>
    <core_client_release>15</core_client_release>
    <work_req_seconds>15.708310</work_req_seconds>
    <resource_share_fraction>1.000000</resource_share_fraction>
    <rrs_fraction>1.000000</rrs_fraction>
    <prrs_fraction>1.000000</prrs_fraction>
    <estimated_delay>4675.133781</estimated_delay>
    <duration_correction_factor>0.906428</duration_correction_factor>
    <code_sign_key>
1024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.
</code_sign_key>
<working_global_preferences>
<global_preferences>
   <mod_time>1208599987</mod_time>
   <run_if_user_active/>
   <start_hour>0</start_hour>
   <end_hour>0</end_hour>
   <net_start_hour>0</net_start_hour>
   <net_end_hour>0</net_end_hour>
   <work_buf_min_days>0.060000</work_buf_min_days>
   <max_cpus>16</max_cpus>
   <cpu_scheduling_period_minutes>120.000000</cpu_scheduling_period_minutes>
   <disk_interval>15.000000</disk_interval>
   <disk_max_used_gb>4.000000</disk_max_used_gb>
   <disk_max_used_pct>80.000000</disk_max_used_pct>
   <disk_min_free_gb>0.500000</disk_min_free_gb>
   <vm_max_used_pct>75.000000</vm_max_used_pct>
   <ram_max_used_busy_pct>75.000000</ram_max_used_busy_pct>
   <ram_max_used_idle_pct>75.000000</ram_max_used_idle_pct>
   <idle_time_to_run>3.000000</idle_time_to_run>
   <max_bytes_sec_up>0.000000</max_bytes_sec_up>
   <max_bytes_sec_down>0.000000</max_bytes_sec_down>
   <cpu_usage_limit>100.000000</cpu_usage_limit>
</global_preferences>
</working_global_preferences>
<global_preferences>
    <source_project>http://www.worldcommunitygrid.org/</source_project>
    <source_scheduler>https://secure.worldcommunitygrid.org/boinc/w...</source_scheduler>
    <mod_time>1208599987</mod_time>
    <cpu_scheduling_period_minutes>120</cpu_scheduling_period_minutes>
    <disk_interval>15.0</disk_interval>
    <disk_max_used_gb>4.0</disk_max_used_gb>
    <disk_max_used_pct>80.0</disk_max_used_pct>
    <disk_min_free_gb>0.5</disk_min_free_gb>
    <idle_time_to_run>3.0</idle_time_to_run>
    <max_bytes_sec_down>0.0</max_bytes_sec_down>
    <max_bytes_sec_up>0.0</max_bytes_sec_up>
    <max_cpus>16</max_cpus>
    <run_if_user_active/>
    <cpu_usage_limit>100.0</cpu_usage_limit>
    <ram_max_used_busy_pct>75.0</ram_max_used_busy_pct>
    <ram_max_used_idle_pct>75.0</ram_max_used_idle_pct>
    <work_buf_min_days>0.06</work_buf_min_days>
    <end_hour>0</end_hour>
    <net_end_hour>0</net_end_hour>
    <net_start_hour>0</net_start_hour>
    <start_hour>0</start_hour>
    <vm_max_used_pct>75.0</vm_max_used_pct>
    <work_buf_additional_days>3.0</work_buf_additional_days>
    <suspend_if_no_recent_input>0.0</suspend_if_no_recent_input>
</global_preferences>
<global_prefs_source_email_hash>aaf5827d09d4eb31df2e3e25d1de9bd8</global_prefs_source_email_hash>
<cross_project_id>82fe8eb8040cdd90c611a840669ca0ef</cross_project_id>
<time_stats>
    <on_frac>0.936869</on_frac>
    <connected_frac>0.999980</connected_frac>
    <active_frac>0.999833</active_frac>
    <cpu_efficiency>0.986426</cpu_efficiency>
</time_stats>
<net_stats>
    <bwup>7304.39</bwup>
    <bwdown>86646.3</bwdown>
</net_stats>
<host_info>
    <timezone>-28800</timezone>
    <domain_name>cantos.foo.c</domain_name>
    <ip_addr>10.2.0.150</ip_addr>
    <host_cpid>6ec4f49987d040fab8cd6d88e24df422</host_cpid>
    <p_ncpus>2</p_ncpus>
    <p_vendor>GenuineIntel</p_vendor>
    <p_model>Intel(R) Core(TM)2 CPU          6300  @ 1.86GHz 
[x86 Family 6 Model 15 Stepping 2] [fpu tsc pae nx sse sse2 mmx]</p_model>
    <p_fpops>2271606290.727339</p_fpops>
    <p_iops>4733359725.197291</p_iops>
    <p_membw>1000000000.000000</p_membw>
    <p_calculated>1209028784.763641</p_calculated>
    <m_nbytes>2145820672.000000</m_nbytes>
    <m_cache>1000000.000000</m_cache>
    <m_swap>10570137600.000000</m_swap>
    <d_total>120582045696.000000</d_total>
    <d_free>79926657024.000000</d_free>
    <os_name>Microsoft Windows XP</os_name>
    <os_version>Professional Edition, Service Pack 2, (05.01.2600.00)</os_version>
    <accelerators>Radeon X1800 GTO/Radeon X1800 GTO Secondary</accelerators>
</host_info>
    <disk_usage>
        <d_boinc_used_total>65490545.000000</d_boinc_used_total>
        <d_boinc_used_project>49943785.000000</d_boinc_used_project>
    </disk_usage>
<other_results>
    <other_result>
        <name>ls042_00032_11</name>
    </other_result>
    <other_result>
        <name>X0000045061278200501272008_3</name>
    </other_result>
</other_results>
<in_progress_results>
    <ip_result>
        <report_deadline>1211014908.000000</report_deadline>
        <cpu_time_remaining>8639.633385</cpu_time_remaining>
    </ip_result>
    <ip_result>
        <report_deadline>1209465955.000000</report_deadline>
        <cpu_time_remaining>4776.634976</cpu_time_remaining>
    </ip_result>
</in_progress_results>
</scheduler_request>
[Apr 27, 2008 3:56:03 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
Hi fregimus,
Another possibility. Your Messages tab should say whether you are using Global preferences or Local preferences. Make sure that you are using the Global preferences that specify 2 days cache.

Lawrence
[Apr 27, 2008 3:58:04 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Help needed: BOINC Agent crunches out its entire stash!
Thank you. I do not know how to interpret the log; I am copying the beginning of it, all lines up to the very first resuming task message:

080416 224458||Starting BOINC client version 5.8.15 for windows_intelx86
080416 224458||log flags: task, file_xfer, sched_ops
080416 224458||Libraries: libcurl/7.16.0 OpenSSL/0.9.8a zlib/1.2.3
080416 224458||Data directory: C:\Program Files\BOINC
080416 224458||Processor: 2 GenuineIntel Intel(R) Core(TM)2 CPU 6300 @ 1.86GHz [x86 Family 6 Model 15 Stepping 2] [fpu tsc pae nx sse sse2 mmx]
080416 224458||Memory: 2.00 GB physical, 9.84 GB virtual
080416 224458||Disk: 112.30 GB total, 74.39 GB free
080416 224458|World Community Grid|URL: http://www.worldcommunitygrid.org/; Computer ID: 200125; location: (none); project prefs: default
080416 224458||General prefs: from World Community Grid (last modified 2008-01-18 01:05:16)
080416 224458||Host location: none
080416 224458||General prefs: using your defaults
080416 224458||Reading preferences override file
[Apr 27, 2008 4:35:57 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 18   Pages: 2   [ 1 2 | Next Page ]
[ Jump to Last Post ]
Post new Thread