Message boards :
Number crunching :
Haunted by SETI Past, How to Delete a Rogue Machine?
Message board moderation
Author | Message |
---|---|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
It's a Linux thing. A while back, when the repository version wasn't working, I had installed BOINC in Ubuntu. Since it wasn't working, I uninstalled BOINC and deleted the machine from my list of Computers. I recently booted to the system and ran the long overdue system updates. This is the machine with the new Windows 8 and I don't use Ubuntu on it. Apparently, the SETI files lurking deep within the system came to life, downloaded files and began working. When I discovered this, I searched for all BOINC/SETI files and deleted them, I don't intend to run SETI on this system. Now I find I can't delete this Computer from my list of Computers. Apparently if it has active tasks you can't delete it? So, how do I abort these tasks and delete this Computer? There isn't anything SETI installed in that system at present. Computer 6941666 |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
You can only delete hosts with 0 tasks. You will just have to wait until after May 22nd to do so. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
I see. So 57 people are going to be waiting on 57 tasks for quite a while. There should be some way an Owner can delete a machine if he desires. You would think. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
I see. So 57 people are going to be waiting on 57 tasks for quite a while. There should be some way an Owner can delete a machine if he desires. You would think. You would think that, but they decided not to have that option. If you had aborted the tasks they would have been resent out quicker. I don't see why it says it's a linux host if it was running 8. Had it populated the OS with the current one you could possibly install BOINC, merge hosts, & then abort them after getting the tasks as resends. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
It's a dual Boot machine. All my machines are at least dual Boot. BOINC wasn't installed when SETI decided to download files and start working. You can't delete files without BOINC, as far as I know. A simple delete button on my Computer page would solve everything... |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
I see. So 57 people are going to be waiting on 57 tasks for quite a while. There should be some way an Owner can delete a machine if he desires. You would think. If you restore Boinc to a host, set the 'Use CPU' to 'no' on your default venue, (so if it's a new id Boinc won't get work for it) attach the host to Seti, then change the host ID to the one above (if needed), (see FAQ: How to revert to an older HostID number?) set 'Use CPU' to 'yes', get the Wu's resent, set NNT, then abort them, and report them, then those 57 people wont have to wait, Claggy |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
I see. So 57 people are going to be waiting on 57 tasks for quite a while. There should be some way an Owner can delete a machine if he desires. You would think. The delete button, why did SETI decide to not include it? This machine existed a long time ago. After it repeatedly erred out every download it was deleted. It would solve a lot of problems if I could just delete it now. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
It's a dual Boot machine. All my machines are at least dual Boot. BOINC wasn't installed when SETI decided to download files and start working. You can't delete files without BOINC, as far as I know. A simple delete button on my Computer page would solve everything... OK the dual boot makes more sense to me now. It sounds like BOINC wasn't actually removed from the Linux installation. The SETI@Home apps can't decided to download work. They don't have the ability to do so. They only process data that BOINC requests from projects. I would guess they don't see a need to add the functionality to remove a host with active tasks. As the problem corrects itself. There probably have been several requests for something along these lines submitted to the BOINC Alpha list. I expect there would be a lot of code to do all of the clean up work for the tasks in order to delete a host. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
According to the Ubuntu repository, BOINC was not on my System when SETI decide to come to life. I checked. If BOINC was installed, there would have been a little green check and the option would have been to Uninstall. There wasn't a check and the option was to install. This system was a straight repository install back then. I had no idea how to install BOINC in Ubuntu any other way...back then. So somehow, on this system, SETI came to life without BOINC. All I want to do is Delete one of my Computers, seems simple enough. |
Bernie Vine Send message Joined: 26 May 99 Posts: 9954 Credit: 103,452,613 RAC: 328 |
And so it is but possibly not with 57 tasks still in progress. Seems like a sensible precaution to me. Why can't you wait till the tasks expire? |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
I can wait, no problem. I just thought I'd help out a few people that might have to wait two months for their tasks to validate. What is the difference of me stopping the work in Windows 8, booting to Ubuntu, going through all the trouble of reinstalling/downloading 57 tasks, aborting those tasks, and then deleting the machine; or, just deleting the machine? I tried the merge, it doesn't want to merge the tasks to the other working system on the same machine, "Host 6796475 has overlapping lifetime: (15 Mar 2013 - 15 Mar 2013), (13 Oct 2012 - 15 Mar 2013)". Ha ha, I was just trying to save some people a wait. If that is a bad thing, excuse me. |
Horacio Send message Joined: 14 Jan 00 Posts: 536 Credit: 75,967,266 RAC: 0 |
I think that a better control of the users over dettached or no longer used computers in their account is a good thing, not only to speed up the validations (which at the end is just about when credits will be awarded) but also because those tasks are adding to the amount of records in the DB tables... Anyway, Im sure that really few users will use that and for sure not very often... so it wont be an efficient use of the scarce human resources of the project to code and test that feature... |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
Ha ha, I was just trying to save some people a wait. If that is a bad thing, excuse me. It's not a bad thing, but not worth the efford at all. There are thousands of WUs, that will never be returned, your 57 does not change anything. The system is build to handle things like this by itself. Actually, considering the current bandwidth issues, letting the WUs time out is the best you can do, as that does not generate additional load on SETI's internet connection. |
SonicAgamemnon Send message Joined: 8 Apr 06 Posts: 33 Credit: 30,435,904 RAC: 7 |
Perhaps I should volunteer to code this feature, because after hours of processing time on these huge WUs it is really annoying to see them marked "inconclusive" for months. It's just my luck to expend huge crunch time on WUs that depend on other machines that rarely run and/or simply drop off the face of the Earth. It really screws your RAC but good... *** END OF RANT *** "History is a pack of lies about events that never happened told by people who weren't there." - Santayana |
Horacio Send message Joined: 14 Jan 00 Posts: 536 Credit: 75,967,266 RAC: 0 |
Perhaps I should volunteer to code this feature, because after hours of processing time on these huge WUs it is really annoying to see them marked "inconclusive" for months. It's just my luck to expend huge crunch time on WUs that depend on other machines that rarely run and/or simply drop off the face of the Earth. It really screws your RAC but good... But even, with that feature you will need to force those "anonymous" users to use it... BTW, most of the inconclusives have nothing to do with the "dettached" or no longer active hosts, and over time you will be having a good stock of pending tasks from all the times, which will keep the RAC stable... (as "stable" as the CreditNew allows, which probably is an oxymoron ;b ) |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
BTW, most of the inconclusives have nothing to do with the "dettached" or no longer active hosts, Most? All. Inconclusives have been returned by the wingmen, but the results are not similar enough. WUs, that have no longer active hosts as wingmen are "Completed, waiting for validation". |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
Meanwhile, one of my other machines is currently passing 190 AstroPulse "Validation pending". A few days ago I thought 170 was high... |
Horacio Send message Joined: 14 Jan 00 Posts: 536 Credit: 75,967,266 RAC: 0 |
BTW, most of the inconclusives have nothing to do with the "dettached" or no longer active hosts, Yes, but what Ive meant is that if the third wigman does never return it, that inconclusive remains as pending even longer, so there are some inconclusives that are still inconclusives due to this also... |
SonicAgamemnon Send message Joined: 8 Apr 06 Posts: 33 Credit: 30,435,904 RAC: 7 |
Meanwhile, one of my other machines is currently passing 190 AstroPulse "Validation pending". A few days ago I thought 170 was high... Hey, this situation presents yet another feature to add: The ability to designate your wingman or a group of wingmen. I would like to designate you as my wingman, because we could quickly confirm our mountain of AtroPulse units. For example, on one of my computers: http://setiathome.berkeley.edu/results.php?hostid=6857388&offset=0&show_names=0&state=2&appid= Among the 400+ WUs stuck in pending purgatory for what seems like forever are well over 200 big AstroPulse units collecting dust, queued forever and a day. We should be able to find and "friend" wingmen with similar queues in order to more effectively eliminate our hellish backlog of WUs... "History is a pack of lies about events that never happened told by people who weren't there." - Santayana |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
The situation appears to be getting progressively worse. When the AP tapes ran dry for a while there was just a slight improvement, then a quick reversal. Now the tapes are dry again. A different version of your 'features' could be to send out 'reassignments' when the tapes run dry. Tasks waiting validation a certain length of time would be resent to Hosts with quick turnaround times that after a couple of days are probably out of AP work anyway. Or, you could just shorten the deadlines... BTW, what does a server do when it is 'Not Running' or 'Disabled'. Are there any attempts to have it do other work? |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.