Posts by [B^S] sTrey

21) Message boards : RALPH@home bug list : Bug reports for Ralph 5.21 (Message 1796)
Posted 7 Jun 2006 by [B^S] sTrey
Post:
I think the most common refrain on ralph message boards is that there's not enough work. So we're trying a new strategy for our workunit queue -- from now on, there will always be work in the queue to help us debug continuously! We didn't do this before because we needed quick turnaround for certain new workunits every couple days. Now we've changed the workunit buffer size and priority system to let us send out new jobs quickly while maintaining a trickle of regular jobs at other times. Does that sound OK to everyone?


Sounds great, thanks!

4 results no delay, no graphics problems so far. As soon as it works off a little debt will be back crunching :)
22) Message boards : RALPH@home bug list : Forum bug? (Message 1795)
Posted 7 Jun 2006 by [B^S] sTrey
Post:
Different issue on same subject: I now get this trying to get to ralph's home page:
can't read ../cache/65/index.php; lastmod 1149661629


(p.s. yes this also shows up as "in 5 minutes" as I edit it)

[edit - Jun 07 07:30:26 UTC homepage loads again]
23) Message boards : RALPH@home bug list : Bug reports for Ralph 5.21 (Message 1774)
Posted 5 Jun 2006 by [B^S] sTrey
Post:
My one 5.21 wu ran before I got up, but from the log I'd say it finished without the delay. Could use some more wus :) to check graphics etc.
One thing, the watchdog timer bug is said to have been introduced with 5.20 but I definitely saw the delayed-finishing behavior in 5.19. Makes me wonder.
24) Message boards : RALPH@home bug list : Bug reports for Ralph 5.20 (Message 1759)
Posted 3 Jun 2006 by [B^S] sTrey
Post:
My Ralph calculates the WUs to 100% but doesnt send them, and they are still "active" but there is no further calculation, the programm continues with my rosetta WUs...


Oh it DID send the WU after some time, sry!!!


I've noticed this too with 5.19 and 5.20. My pref is set to 2 hours and my crunching interval is 2:01. The wus I've been getting happen to finish early, say 1:45, go to 100% but then pause instead of completing. Nothing else such as downloads has triggered early rescheduling. The next time the wu gets crunch-time it completes immediately and uploads.

Not causing any problems but it's definitely different behavior, and after about 5 in a row not counting one that errored out, it doesn't seem coincidental.

sample result
25) Message boards : RALPH@home bug list : Bug reports for Ralph 5.17-5.19 (Message 1740)
Posted 1 Jun 2006 by [B^S] sTrey
Post:
Ditto, same error shown in this result.

Errored out immediately (cpu = 14 seconds) with the exit code 1 @ barcode_classes.cc line:500

app is 5.19, client 5.4.9, windows XPsp2
26) Message boards : RALPH@home bug list : Bug reports for Ralph 5.16 (Message 1651)
Posted 16 May 2006 by [B^S] sTrey
Post:
The MAPRELAX wu I reported earlier is now just shy of halfway through its 4-hour task. 149MB in memory, peak of 405MB, VM size 416MB. Not as bad as some, but more than I'm comfortable with. I'm going to suspend some other projects and restart boinc to lessen the competition until this wu finishes.


Just missed the edit-post deadline... I waited for a checkpoint then restarted boinc. At the time of the restart it was just beginning model 8. ralph was consuming 217MB memory peak 405, VM size 468MB.

On the restart, it went up to about 93 MB peak 95, VM 117 MB. It's growing though; as I type this it's at 130MB memory peak 13, VM 145MB.

27) Message boards : Number crunching : Can you tell when the next checkpoint is for a given work unit? (Message 1650)
Posted 16 May 2006 by [B^S] sTrey
Post:
CPDN is quite nice this way, allowing me to minimize work loss when I must restart boinc or the OS. Recent memory issues with ralph have increased my interest in being able to do restarts with minimal loss.

I realize not all applications are predictable in the way CPDN units are, but I'm just wondering, is there any way to tell from Ralph graphics how close the next checkpoint is? E.g. my understanding is that rosetta checkpoints once per model; if that's correct then is there a known # of steps per model?
Or do you just wait for the model # or progress % to change and therefore know after the fact that a checkpoint just occurred?
28) Message boards : RALPH@home bug list : Bug reports for Ralph 5.16 (Message 1649)
Posted 16 May 2006 by [B^S] sTrey
Post:
The MAPRELAX wu I reported earlier is now just shy of halfway through its 4-hour task. 149MB in memory, peak of 405MB, VM size 416MB. Not as bad as some, but more than I'm comfortable with. I'm going to suspend some other projects and restart boinc to lessen the competition until this wu finishes.

If this large memory usage is in fact due to the type of wu and not leakage or some other bug, it would be nice if we could set willingness to crunch memory-gobbling tasks in host preferences.

Of course there are no host preferences, but even a project-wide preference setting would be helpful for many of us.
29) Message boards : RALPH@home bug list : Bug reports for Ralph 5.16 (Message 1642)
Posted 15 May 2006 by [B^S] sTrey
Post:
Update before I hit the road: 1 hr 31 minutes into the wu, memory now is 325MB peak 329, Vm size is 341MB. PF ranging from 0-1K/second. Avail memory down to 74K, will see if the box is still functioning when I get home!
30) Message boards : RALPH@home bug list : Bug reports for Ralph 5.16 (Message 1641)
Posted 15 May 2006 by [B^S] sTrey
Post:
Running a 5.16 wu now for all of 30 minutes (XP Pro SP2, P4, 1 GB mem), and memory usage is already high: 174K memory, 191K VM. These numbers aren't intolerable but they're higher than I'm used to seeing from rosetta, and make me think it will climb into the range I reported on earlier. I'll be away most of the day,;will report back when the wu has more mileage.

See also this post on the number-crunching thread; it reports 690MB which is higher than I've seen.

p.s. I leave boinc apps in memory when switched. Are these high numbers only happening to machines which leave rosetta in memory when preempted?
31) Message boards : RALPH@home bug list : Bug reports for Ralph 5.15 (Message 1619)
Posted 14 May 2006 by [B^S] sTrey
Post:
Running a MAPRELAX_TEST unit now and either this is a bad unit or the work towards reducing memory usage is going backwards. Currently on my XP box Task Manager gives the following values for this work unit: memory usage 233MB, Peak memory usage 351MB, VMsize 391 MB. Ouch, this is significantly worse than previous units/ralph versions I've run. It's taken the memory-hog prize away from FightAIDS@Home (30,113,277MB) which was the undisputed front runner for months.
It's pretty painful for a 1GB machine; pagefile usage is about 1.5GB now and 40% of that is Ralph!

p.s.
PF is about 400 every 2-4 seconds, which is much lower than 5.12 units in insanity mode (2-4K/second). It's about halfway through my 4-hour preference setting and just got pre-empted, will see what happens on the last half.

Graphics: Boinc screensaver set to go to blank screen after 5 minutes. I also inspected graphics on this wu manually and I do not know what the memory usage was before that.
32) Message boards : RALPH@home bug list : Bug reports for Ralph 5.13 (Message 1583)
Posted 11 May 2006 by [B^S] sTrey
Post:
I'm seeing the same thing in the graphics, the display alternates between normal appearance and partial chains as shown by wizzszz's pics.

However I much prefer this to the problems 5.12 was causing...
33) Message boards : RALPH@home bug list : Bug reports for Ralph 5.13 (Message 1578)
Posted 11 May 2006 by [B^S] sTrey
Post:
Thank you Rhiju, for responding so fast when only a few people were complaining about this, and also to Rom for responding quickly and accurately on a fix/workaround. All is still going well on my pc.

34) Message boards : RALPH@home bug list : Bug reports for Ralph 5.13 (Message 1574)
Posted 11 May 2006 by [B^S] sTrey
Post:
5.13 has only been running 30 minutes here, but so far so good!
I need to run things overnight with my normal project mix to be sure.

Any short technical description on the debugging code removed? Or a pointer to it if it's public? I may beg for details to troubleshoot another app which handled this particularly badly; I don't think telling the vendor to run ralph/rosetta 5.12 is going to fly...
35) Message boards : RALPH@home bug list : Bug reports for Ralph 5.11 and 5.12 (Message 1563)
Posted 10 May 2006 by [B^S] sTrey
Post:
This is anecdotal only, but having seen Son Goku's post I will chip in that I too have seen my system get noticeably sluggish when Ralph is loaded,
only since about 5.11 or 5.12. I've been running BOINC over a year with the same hardware (P4, HT restricted to one pseudo-cpu, 1GB memory)
and this hasn't happened before, and there are other projects I run like WCGrid's FightAids@Home which use more memory (virtual and not)
which don't cause this. I don't understand the reason, unless Task Mgr lies about memory and VM sizes (quite possible)...
but I've had things happen like my specific applications consistently failing or small applications taking huge amounts of time to start up,
only once I have let Ralph run. Even if Ralph is not currently running, as long as it has run and is still in memory.

I know this sounds a bit nuts but I've watched it several times by going through the sequence from boot up with all boinc projects suspended,
enabling different combinations of projects -- including combinations which add up to more memory usage -- and watching things flake out
as soon as Ralph has loaded and only when Ralph is loaded, with no other evidence of problems with the box.

Out of time to pursue it more, and I'll probably NNW Ralph for a while as I need this machine to function properly, and this problem seems like
a Twilight Zone ep. If more than one or two of us are affected I'm happy to pursue it further when I get some time.

p.s. I do use the screen saver, going to blank screen after 5 minutes. Occasionally I'll run the graphics manually but not often. Single-user install.
36) Message boards : RALPH@home bug list : Bug reports for Ralph 5.05 and higher (Message 1411)
Posted 27 Apr 2006 by [B^S] sTrey
Post:
4 days? Ouch, I'd hoped for 6 or 7, and definitely with smaller quotas.

seti beta has a painfully small return rate due to huge quotas. Shorter deadlines aren't as direct, well I guess they are here but if you're running a quorum of more than 1, short deadlines drag things out having to resend after earlier results time out...

Meanwhile I've preferred to test with 16-hr runtimes, and I do run other projects. With my current mix I can probably just make 4 days. Of course when you want really fast returns those hit-and-quit wu's you've been sending, do the job.

So I'm wondering is there little value for testing longer time settings here? Easy enough to drop back to 2 or 4 hour runtimes.

p.s.
If this discussion continues, maybe it's better moved out of the bug-report thread?
37) Message boards : RALPH@home bug list : Bug reports for Ralph 5.05 and higher (Message 1397)
Posted 27 Apr 2006 by [B^S] sTrey
Post:
wu 79004 killed by watchdog just after 2 hrs' runtime (Stuck at score -115.914 for 3600 seconds)
(Sorry this was 5.05, can't get any 5.06 wus)
38) Message boards : RALPH@home bug list : Bug reports for Ralph 5.04 (Message 1365)
Posted 26 Apr 2006 by [B^S] sTrey
Post:
wu 85055 was killed by the watchdog; credits were granted.

cpu_run_time_pref: 57600
Watchdog is killing the run! Stuck at score 2.48287 for 3600 seconds
39) Message boards : RALPH@home bug list : Bug reports for Ralph 5.04 (Message 1349)
Posted 25 Apr 2006 by [B^S] sTrey
Post:
My one 5.04 wu, and one 5.03 wu for that matter, both finished successfully. However I'm wondering why they both finished in under 2 hours when my preference is set to 16 hours and earlier wu's did run for that time.

wu 95168, 5.04
wu 94408, 5.03

Did you "Update" with BOINC Manager after changing your preferences on the web site? If not, the preferences will not be changed until the client contacts the server. Pardon me if this response is too obvious, but you never know.... :)


No problem, appreciate the thought (however I did update). I should have said in my original post that the result files show the preference set to 16 hours.
40) Message boards : RALPH@home bug list : Bug reports for Ralph 5.04 (Message 1347)
Posted 25 Apr 2006 by [B^S] sTrey
Post:
My one 5.04 wu, and one 5.03 wu for that matter, both finished successfully. However I'm wondering why they both finished in under 2 hours when my preference is set to 16 hours and earlier wu's did run for that time.

wu 95168, 5.04
wu 94408, 5.03


Previous 20 · Next 20



©2024 University of Washington
http://www.bakerlab.org