Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 49
|
![]() |
Author |
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Challenging you to venture into the scary cc_config manual yourself and find the answer ;?
----------------------------------------(there's 2, yes two options, an old to influence the log file size and a new [I've never tried] to influence how much is shown in the log pane) edit: Maybe there's Beta fatigue developing... me clients are on a second helping of the 'only 20,000' (which I interpret a 2x20K). None of the first found a wingman yet. [Edit 1 times, last edit by Former Member at Dec 13, 2013 2:38:12 AM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi ccandido,
The size is limited to about 2 megabytes. When the Event Log reaches 2 megabytes, the old Event Log, is deleted. The current Event Log is renamed stdoutdae.old from stdoutdae.txt and a new Event Log is started in a new stdoutdae.txt file. When you shut down or reboot, the current Event Log is appended to the end of stdoutdae.old. If appending would exceed the limit, the old Event Log is deleted and stdoutdae.txt becomes the new stdoutdae.old, starting a (chuckle) new old Event Log. Lawrence |
||
|
ccandido
Senior Cruncher Joined: Jun 22, 2011 Post Count: 182 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Seke, I accepted the challenge:
----------------------------------------<max_stdout_file_size>104857600</max_stdout_file_size> I'm going to try 100 MB max size On another topic, some fatigue might have an advantage, more jobs left... Thanks to Lawrence too ![]() ![]() [Edit 1 times, last edit by ccandido at Dec 13, 2013 3:31:36 AM] |
||
|
ccandido
Senior Cruncher Joined: Jun 22, 2011 Post Count: 182 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Rebooted the machine with 8 running betas and all restarted from checkpoint.
----------------------------------------![]() ![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
7.28 is looking good: 3 of my Betas have turned Valid, all 3 of mine were suspended/resumed at least once with LAIM off, all 3 wingmen showed only one Initializing line, my Result.out was identical to each wingman's.
|
||
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
(Sek):"Maybe there's Beta fatigue developing..."
Yep, been running around the farm, mustering the goats. Am getting quite footsore, and will soon be quite LAIM. Have given most bleaters a stop and resume, but some did escape, so I'm feeling quite betan. ---- Seriously, with 9 devices it's hard work keeping up with micromanaging the new arrivals and making sure that I give a restart to as many WUs as I can. The job is not helped by the weird choices that BOINC client 7.2.28 makes when it decides which task to start when I suspend a running task, or which task it decides to suspend in situations such as when it grabs a thread to run a task high pri, or when I reduce the number of cores allowed. I was not impressed when it unexpectedly suspended a CEP2 WU on 1 occasion at least. Sure hope the problems are all fixed, and that all/most of the techs can have their weekend away from The Monster. |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Rob, I accepted the challenge: <max_stdout_file_size>104857600</max_stdout_file_size> I'm going to try 100 MB max size On another topic, some fatigue might have an advantage, more jobs left... Thanks to Lawrence too [OT]Did write 2 settings... you missed the one that changes the number of lines held visible in the Event window (warning: Going overboard on log file size and active event window lines may impact BOINC Manager responsiveness... same as buffering too many jobs)[/OT] On Topic, 3rd helping for 18 of 22 cores [4 of batch 216 are into 7th hour at ~45%], and too having a "LAIM off suspend" treatment (Rickjb, first suspend all 'ready to start' tasks that are not Beta before the LAIM game... had a more complex dance on one machine that had 7.26 prod. running too, so had to alternate LAIM in between to keep those in memory to not risk them going invalid). [Edit 1 times, last edit by Former Member at Dec 13, 2013 9:16:07 AM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Now running 21 of 22 cores with 7.28... no more new coming [but those 216's are long ones... racing towards emerald, if you can call it racing :]. Tested the restart issue ad nauseam ... me fatigues are drenched and regrettable missed to switch LAIM at one test so 7.26 got also unloaded... high probs they go invalid, worse meet with 4 more to get the Too Late route treatment [which oh oh, gives more crdt than an invalid... for consistency, but I do understand the reasoning behind this :O].
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
BTW, noticed on me 'crunch only' Linux-64 quad & W7-64 octo, that the efficiency is showing as > 99.9% [the CPU : Elapsed ratio]. Did not pay close attention how this is prior and after a restart so screenshotted a before and after and find there's no particular % efficiency shift... no lost time recorded. Greater than 99.9% is incredible... very lean code [but then there's enough RAM to work with... no swapping needed], also a testament to the OS's going largely into snooze mode, purely servicing our crunshing. The W8.1-64 is slightly less efficient, but noticed if you explicitly sign-out not just lock screen, to get 1% better performance i.e the Metro/Modern active tile-interface may keep sucking CPU juice to keep the background fresh, for when screen up again [for those who wish to squeeze the last percent]. Eagerly waiting on 8.2 [seems to be in the works with a quasi-permanent return to the W7 like GUI... eclectic I suppose]
|
||
|
rbotterb
Senior Cruncher United States Joined: Jul 21, 2005 Post Count: 401 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Restarted my laptop with 4 MCM1 7.28 WUs and they appear to have restarted fine, though their remaining times were about 1 hour shorted than where they were when I shut them down last night.
Maybe the new trick to finish MCM1 WUs now is to restart my laptop a couple times, and have the remaining time drop one hour each time until doing a quick finish...LOL ![]() Anyway they seem to be behaving, so since I'm out of the house now I'll have to check them later today and see if they finished out OK..... |
||
|
|
![]() |