Posts by SuperSluether

1) (Message 4272)
Posted 27 Mar 2015 by SuperSluether
Post:
I don't mind everyone not getting tasks. On the Cobblestone lists, I was getting fed up being overtaken by people with super-fast computers who have only been crunching for a couple of weeks and can complete a task in minutes instead of four hours like my laptop. At least us slower are holding position for an extra few hours now.


I hate to break it to you, but not everybody has a laptop. Desktops are a lot faster than laptops by nature, and since a lot of people run BOINC on their gaming rig like me, there's some seriously suped-up crunchers out there.
2) (Message 4195)
Posted 21 Mar 2015 by SuperSluether
Post:
Well, I took some compressed air and blew out my computer. Everything's running at normal temps again. If it gets too hot during summer, I'll just go from my 4.30Ghz overclock back to the stock 3.5 clocks.
3) (Message 4127)
Posted 13 Mar 2015 by SuperSluether
Post:
I've only got 1 machine in my bedroom, but my 70-80 C range is starting to creep into the mid 80s and 90s. Looks like I'll have to adjust my overclock for the summer.
4) (Message 4027)
Posted 16 Feb 2015 by SuperSluether
Post:
It's back up to 200,000+ tasks now. I don't see a work generator for this project, so goodness knows where the work units come from. No matter the project, the work generator always takes a break from time to time and the amount of tasks falls behind for a few hours. It's BOINC, I think it might be normal. :P
5) (Message 4026)
Posted 15 Feb 2015 by SuperSluether
Post:

Lots of wasted processor time over the last few days.


Not necessarily. If you have a download error, BOINC doesn't run the task, meaning that you only process tasks that download correctly.


This is the 3rd project I have dedicated my computer time to.


Then why are download errors from this one so problematic for you? Unless you have limited bandwidth, it doesn't really matter. There are tons of BOINC projects out there, and each one has a good intent and needs computing power.
6) (Message 4019)
Posted 13 Feb 2015 by SuperSluether
Post:
A temporary solution is to modify your cc_config.xml file, which is located in the BOINC data directory. Look for this line of code:

<dont_check_file_sizes>0|1</dont_check_file_sizes>

Change the number to 1, and BOINC will ignore "wrong size" errors.

"Checksum or signature error," "MD5 check failed," and "permanent HTTP error" will have to be fixed on the server side.
7) (Message 3996)
Posted 9 Feb 2015 by SuperSluether
Post:
From the news thread, it sounds like Kyong is having some IRL complications. If the servers are still in his old apartment and his work is making it hard to check on them, it might be a few days before he sees the problem. Hopefully it'll be easy to fix.
8) (Message 3703)
Posted 16 Oct 2014 by SuperSluether
Post:
Back up to 99 K. Whatever happened, it looks like the database is filling up again.
9) (Message 3701)
Posted 16 Oct 2014 by SuperSluether
Post:
Well SOMETHING's going on, because I just got 40 tasks for CPU and 6 for NVIDIA GPU.
10) (Message 3696)
Posted 16 Oct 2014 by SuperSluether
Post:
Hopefully everyone has a backup project or two, or nine... :P


I actually run 13 projects on my computer. :P Anyway, what happened? This project used to have just over a million tasks ready to send at any given time, and now there's nothing?! Is the work generator broken, or are the servers still recovering from the "update every 5 minutes" issue?
11) (Message 3682)
Posted 11 Oct 2014 by SuperSluether
Post:
I have removed the next_rpc_delay setting.


Finally, the nightmare of constant requests is over! Hopefully now the server can catch up on its work generating. Looks like it dropped down quite a bit.
12) (Message 3653)
Posted 29 Sep 2014 by SuperSluether
Post:
All I know about so-called Military Class components is that MSi leans heavily on it for their hardware. Supposedly, the Military Class parts last longer, even in bad conditions.
13) (Message 3652)
Posted 29 Sep 2014 by SuperSluether
Post:
It is when a project chooses to award a fixed number of credits per workunit completed and validated. Some projects choose to do this way, some projects choose other ways, there are several.


IF tasks are taking longer due to MORE DATA BEING analyzed the we ALL deserve MORE CREDITS.


1 billion credits and 5 bucks can get you a latte at Starbucks! Without the 5 bucks though you are NOT getting your latte.


Lol
14) (Message 3651)
Posted 29 Sep 2014 by SuperSluether
Post:
Here's what my event log has been looking like:

9/29/2014 8:39:06 AM | Asteroids@home | Sending scheduler request: Requested by project.
9/29/2014 8:39:06 AM | Asteroids@home | Not requesting tasks: don't need
9/29/2014 8:39:11 AM | Asteroids@home | Scheduler request completed
9/29/2014 8:44:11 AM | Asteroids@home | Sending scheduler request: Requested by project.
9/29/2014 8:44:11 AM | Asteroids@home | Not requesting tasks: don't need
9/29/2014 8:44:20 AM | Asteroids@home | Scheduler request completed
9/29/2014 8:44:20 AM | Asteroids@home | Server can't open database
15) (Message 3650)
Posted 29 Sep 2014 by SuperSluether
Post:
For the past 48 hours I have been receiving the following errors. Any ideas?

9/28/2014 4:33:37 PM | Asteroids@home | [error] File input_313587_3 has wrong size: expected 5164, got 6100
9/28/2014 4:33:37 PM | Asteroids@home | [error] Checksum or signature error for input_313587_3
9/28/2014 4:33:37 PM | Asteroids@home | [error] File input_313587_2 has wrong size: expected 5164, got 6100
9/28/2014 4:33:37 PM | Asteroids@home | [error] Checksum or signature error for input_313587_2
9/28/2014 4:33:38 PM | Asteroids@home | Finished download of input_313587_1
9/28/2014 4:33:38 PM | Asteroids@home | Finished download of input_312491_3
9/28/2014 4:33:38 PM | Asteroids@home | [error] File input_313587_1 has wrong size: expected 5164, got 6100
9/28/2014 4:33:38 PM | Asteroids@home | [error] Checksum or signature error for input_313587_1
9/28/2014 4:33:38 PM | Asteroids@home | [error] File input_312491_3 has wrong size: expected 5325, got 5771
9/28/2014 4:33:38 PM | Asteroids@home | [error] Checksum or signature error for input_312491_3

Crunching in Clermont, FL


I had about 30 tasks that failed to download with the same error. Every file either has a checksum/signature error or the wrong file size.
16) (Message 3635)
Posted 25 Sep 2014 by SuperSluether
Post:
There aren't any GPU applications for this project. How are you running tasks on NVIDIA?
17) (Message 3630)
Posted 24 Sep 2014 by SuperSluether
Post:
I can upload, download, and report tasks without problems, but this project keeps saying the project requested an update. After it updates automatically or manually, communication is deferred for an hour, and the message returns after a few minutes. What is this?

Resetting and removing and re adding the project did not fix it.
18) (Message 3538)
Posted 16 Aug 2014 by SuperSluether
Post:
I feel bad for Kyong. It seems like he's the only admin, or at least the only one who fixes the server. It must be hard knowing that you can't even go on vacation without something going wrong. :(

Well, that's what the work buffer is for! It's not like the tasks are that big, so just download enough for a couple days and you'll be good! :)

Or like me and just do work for a different project while the server is down.
19) (Message 3517)
Posted 14 Aug 2014 by SuperSluether
Post:
Oh, I remember now. Without the feeder, nobody gets new tasks. Without the purge, old tasks never get deleted. These 2 server functions are essentially the bottleneck, and now they're not working. :/
20) (Message 3495)
Posted 13 Aug 2014 by SuperSluether
Post:
I noticed I wasn't getting any new tasks, and the communication was deferred for 1 hour instead of the usual 10 seconds. Apparently the feeder is off...


Next 20