Gyes & Girls, I've got a bit of a dilemma.

Message boards : Number crunching : Gyes & Girls, I've got a bit of a dilemma.
Message board moderation

To post messages, you must log in.

AuthorMessage
Spud1200
Avatar

Send message
Joined: 2 Aug 14
Posts: 115
Credit: 1,367,835
RAC: 0
United Kingdom
Message 1902803 - Posted: 25 Nov 2017, 3:11:24 UTC

At the moment I'm currently running Linux Mint across all my machines on my network.

I currently have a good few hundred tasks in line waiting to be processed but I need to do an OS Upgrade all to gether in total accross my machines.

The only thing I can think of is to dich all Tasks and stop crunching on all amchines asap because I want to get my Network switched over as at the moment I'm thinking of updating my OS on LM or jumping ship too another OS all togeather.

Its gonna take about 7 too 10 days to fully process my back log of tasks.

So hears the question.

Would you gyes be OK, with me dumping in a error out 300+ tasks and then taking a few days out . My OS is seriously in need of updating.
ID: 1902803 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 10004
Credit: 131,119,353
RAC: 83,727
Australia
Message 1902806 - Posted: 25 Nov 2017, 3:15:12 UTC

If you're going to wipe everything and start from scratch, then just detach from the project, and re-attach when done.
If you're just updating things, then do the updates & continue crunching your existing cache when you've finished updating.
Grant
Darwin NT
ID: 1902806 · Report as offensive
Spud1200
Avatar

Send message
Joined: 2 Aug 14
Posts: 115
Credit: 1,367,835
RAC: 0
United Kingdom
Message 1902807 - Posted: 25 Nov 2017, 3:19:48 UTC - in response to Message 1902806.  

If you're going to wipe everything and start from scratch, then just detach from the project, and re-attach when done .. .. ..


This ..


If I detach will this download the same tasks again as I'm going too completely wipe the HDDs for a fresh OS Installation.
ID: 1902807 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 10004
Credit: 131,119,353
RAC: 83,727
Australia
Message 1902810 - Posted: 25 Nov 2017, 3:38:12 UTC - in response to Message 1902807.  

If I detach will this download the same tasks again as I'm going too completely wipe the HDDs for a fresh OS Installation.

Nope, you will get new WUs.
But when you detach, all your current unprocessed WUs will be released again, so people don't have to wait a month or 2 for them to time out before they get resent to another system.

Starting from scratch this way will get you new computer IDs
It is possible to make a copy of your current programme & data directories, and copy them back over the new installation of BOINC on the newly installed OS to keep your existing computer IDs and work caches, but it involves a lot more mucking around than just detaching & re-attaching to the project.
Grant
Darwin NT
ID: 1902810 · Report as offensive
Spud1200
Avatar

Send message
Joined: 2 Aug 14
Posts: 115
Credit: 1,367,835
RAC: 0
United Kingdom
Message 1902811 - Posted: 25 Nov 2017, 3:51:16 UTC - in response to Message 1902810.  

Can I ask. By detach do you mean select seti & select remove from the options on the left hand side.
ID: 1902811 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 10004
Credit: 131,119,353
RAC: 83,727
Australia
Message 1902812 - Posted: 25 Nov 2017, 4:00:07 UTC - in response to Message 1902811.  

Can I ask. By detach do you mean select seti & select remove from the options on the left hand side.

Yep, Remove.
Set No new tasks, do an Update to report any finished but not yet reported work, then Remove to release all your current unprocessed work back to the project.
Grant
Darwin NT
ID: 1902812 · Report as offensive
Spud1200
Avatar

Send message
Joined: 2 Aug 14
Posts: 115
Credit: 1,367,835
RAC: 0
United Kingdom
Message 1902813 - Posted: 25 Nov 2017, 4:13:27 UTC - in response to Message 1902812.  

I've just checked & the tasks are still showing up on my main page as over 700.

What will happen too that.
ID: 1902813 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6530
Credit: 186,814,324
RAC: 33,039
United States
Message 1902816 - Posted: 25 Nov 2017, 4:21:40 UTC

My first choice would have been to backup the data directory. Then restore it afterward.
Otherwise I would just abort the tasks from BOINC. That nearly immediately puts them in the queue to go back out.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the BP6/VP6 User Group today!
ID: 1902816 · Report as offensive
Cameron
Avatar

Send message
Joined: 27 Nov 02
Posts: 99
Credit: 3,032,321
RAC: 3,456
Australia
Message 1902918 - Posted: 25 Nov 2017, 21:53:55 UTC

You could considering staggering your upgrade.

If it's just a version bump of the OS [Linux Mint] or a switch to another linux distro.
Upgrade your primary machine/s for example and then perform the upgrade on the rest of your machines.
And migrate the BOINC Data Directory across of each machine

If your also taking the opportunity to change the hardware I'd probably do a project detach of the machine.

finding a Second Sechedule window could be a problem though.
ID: 1902918 · Report as offensive

Message boards : Number crunching : Gyes & Girls, I've got a bit of a dilemma.


 
©2018 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.