Message boards :
Number crunching :
Invalid rate.
Message board moderation
Previous · 1 · 2
Author | Message |
---|---|
Send message Joined: 21 Apr 19 Posts: 25 Credit: 12,699 RAC: 0 |
100% invalid rate on my two Windows 8.1 machines w/ BOINC 7.14.1 and VB 5.1.26 and 5.1.28 but 100% success rate (6/6) on my Windows 7 laptop with BOINC 7.8.3 and VBox 5.1.28. The other machine with Windows 7 has BOINC 7.14.1 with Virtual Box 5.1.30 and is completing 100% valid so far. Seems to be an issue with Windows 8.1 as the host OS. See: https://boinc.nanohub.org/nanoHUB_at_home/forum_thread.php?id=57&postid=261#261 |
Send message Joined: 8 Feb 18 Posts: 9 Credit: 64,768 RAC: 0 |
I installed VBox v6.0.6 on my Win 10 box that was taking up to 60 min's to run the Wu's, now they all seem to take 10 min's or less ... |
Send message Joined: 21 Apr 19 Posts: 25 Credit: 12,699 RAC: 0 |
My Windows 7 Professional 64 with VBox 5.1.30 finished 397 valid, 0 invalid and 1 error (196 (0x000000C4) EXIT_DISK_LIMIT_EXCEEDED). Runs great on that machine, 100% invalid rate on the nearly identical hardware machines but with Windows 8.1. |
Send message Joined: 8 Jan 19 Posts: 24 Credit: 2,501 RAC: 0 |
>>> Seems to be an issue with Windows 8.1 as the host OS. I think there must be more too it than that. I have 8.1 on this system, and VirtualBox 5.1.28, today I have had 60 completed and validated so far. I had a single error yesterday, and 200+ Normal completion and validated. |
Send message Joined: 21 Apr 19 Posts: 25 Credit: 12,699 RAC: 0 |
>>> Seems to be an issue with Windows 8.1 as the host OS. Those machines are running VM's all day, every day, and completed 20,000 hours of Theory VM's each, plus ran ATLAS and Cosmology camb_boinc2docker (beta test) which, I thought, was another version of the boinc2docker run here. The errors in the logs are the same whether valid or not (need to change the boot order to eliminate the earliest ones). Got any idea what to look for? Those 2 separate machines run everything else they've come across, so this is about finding out what is broken with these WU's. |
Send message Joined: 8 Jan 19 Posts: 24 Credit: 2,501 RAC: 0 |
I've had another failure today: 1672983 1204102 818 24 Apr 2019, 9:30:36 UTC 24 Apr 2019, 9:56:04 UTC Completed and validated 146.55 99.14 1.15 boinc2docker v1.12 (vbox64_mt) windows_x86_64 1673020 1204118 818 24 Apr 2019, 9:30:36 UTC 24 Apr 2019, 12:23:53 UTC Error while computing 6,825.86 5,924.72 --- boinc2docker v1.12 (vbox64_mt) windows_x86_64 1673021 1204119 818 24 Apr 2019, 9:30:36 UTC 24 Apr 2019, 9:56:04 UTC Completed and validated 292.68 252.41 2.31 boinc2docker v1.12 (vbox64_mt) windows_x86_64 .. note the REALLY long run time, the abort code is the: 197 (0x000000C5) EXIT_TIME_LIMIT_EXCEEDED ... the task is getting into a loop or similar from which it has no way out, and so it runs until it is killed by the task itself. I've had 80+ Normal Completion task as well on the same machine today. So whatever the loop is, it is an "unusual", on my machine, BUT, you said you are losing a lot? I can't see your machine. I only have the project on this machine, the other I have here is pretty much identical, (same motherboard, processor, RAM, SSD, graphics, OS, VBox), so I doubt we'd learn anything by putting it on there as well. Both my machines here are connected to about ten projects, some being Virtual Box set ups, I have seen more errors and problems from the work units here than probably the rest of my projects this centuary combined. |
Send message Joined: 8 Feb 18 Posts: 9 Credit: 64,768 RAC: 0 |
I don't think the Version of VBox much matters, I've ran 3 different Version on my 3 Box's with the same result. They will run 100 or so okay then go into 50% error mode for the next 20-30 Wu's & then run another 100 or so okay, so to me it's just that certain Wu's are FUBAR & there's not much you can do about it. I'm just in it for the Hours for the WuProp Project & the Credit doesn't mean squat to me so if the Project wants to send crap then I'll run crap ... :D |
Send message Joined: 8 Jan 19 Posts: 24 Credit: 2,501 RAC: 0 |
What is important to me, is my machines running and completing valuable scientific work for the benefit of responsible scientists to aid their research. On regular occasions, the project here is throwing faulty work units to crunchers who are wasting their time. Recently, I've had few errors, others have many, but am still seeing validation errors - something is not right here. Another project, I can't remember which, had problems with Virtual Box 6 and asked that crunchers who wanted to continue their work should retain, or return to 5. I have set no new tasks again. |
Send message Joined: 8 Feb 18 Posts: 9 Credit: 64,768 RAC: 0 |
I have set no new tasks again. Yeah right and that's your way of aiding the Project. Better off to run the Wu's good or bad so they can find out whats wrong with them, but then you would be wasting your time & we wouldn't want that ... |
Send message Joined: 8 Jan 19 Posts: 24 Credit: 2,501 RAC: 0 |
We are volunteers, we are free to choose what we volunteer our resources towards. It is the nature of the platform. |
Send message Joined: 21 Apr 19 Posts: 25 Credit: 12,699 RAC: 0 |
Actually open up the VM and interact with the long running WU and take a screen shot of it. The long runner I opened was sitting at a screen where, once I hit 'enter', it shut down and sent it's results. Wish I'd taken a screen shot; it was waiting on user input and I can't remember the phrasing of the prompt. |
Send message Joined: 8 Jan 19 Posts: 24 Credit: 2,501 RAC: 0 |
That would be of potential use in fixing this. If the task is getting to a point where it needs something, and is not getting it, it may just wait until it times out. I hope the project team are watching. |
Send message Joined: 8 Jan 19 Posts: 24 Credit: 2,501 RAC: 0 |
Task number 07213152_016_1 >>> 194 (0x000000C2) EXIT_ABORTED_BY_CLIENT While I was out. Too many unexplained and unexplainable errors. I'm backing off for a longer period now, if the project is still there in six months, I may give it another try. |
Send message Joined: 20 Feb 18 Posts: 4 Credit: 7,091 RAC: 0 |
Tasks are still ending in an error and this was my first one 15 May 2019, 6:51:25 UTC https://boinc.nanohub.org/nanoHUB_at_home/result.php?resultid=1972199 Every one since then has errored out. |
Send message Joined: 11 Jan 17 Posts: 99 Credit: 224,673 RAC: 0 |
They have bad batches, due to the large number of apps they are running from different users. The admin posted a while ago that they fix them as they find them, but with over 400, it takes a while. Check your machine to ensure that you have enough memory and are not overclocking, etc. Otherwise, it is something they have to fix. |
Send message Joined: 24 Apr 19 Posts: 53 Credit: 114,639 RAC: 0 |
They have bad batches, due to the large number of apps they are running from different users. I guess that whoever submits WUs should pay a closer look at what is broken and what needs fixing. In the end, they want good results not a pile of s..t. Plus adding an option for server to abort those dodgy tasks would be great. |
Send message Joined: 11 Jan 17 Posts: 99 Credit: 224,673 RAC: 0 |
A server abort is a good idea. But I don't think it is a problem with the apps, they just have to be adapted to run in VirtualBox (and BOINC). They don't normally run them that way, so they have no way to know what works except send them to us. We are the testers. |
©2025 COPYRIGHT 2017-2018 NCN