I see they posted a FarceBook message about 13 hours ago saying "they found the problem and are 'fixing it' breaking it some more"

Yeah well, at least the BOINC part of WCG is still working.

If you described the problem as seen by an end user

How many of such users would be interested in better run time characteristics for a program like boinccmd?

and how the alternative you suggest might help, you might get a more positive response than …

The statement “RPC rpc(this);” can be searched in source files.
I got the impression then that the mentioned identifier is used only for a local variable in several member functions.
The shown function call trace excerpts indicated that multiple remote procedure calls would be performed so far.
Objects would be constructed for the variable “rpc” so often accordingly.
Can you imagine that the software implementation would benefit from the design adjustment to define such a variable only once and reference the corresponding object at the other source code places?

Tried to add a project and when I tried to sign in, it said NO SUCH USER! What happened??? I had to register again??? What happened to all my stats and stuff?? Now it states I'm a member since 2023 and I do remember that I have been a member since at least 2007 or before....

I am considering deleting BOINC and re-installing as a service so it will start to run without the need to be logged in, e.g.. if the machine recovers from a power interruption while I am not here. I am assuming that the manager will operate in the same way as communication with the core should be exactly the same. In case... (!)... there are things I am missing, are there things I am missing?

Boinc Client (Windows 10) Version 7.22.2 (x64)

Hello - Is there a way to set the Client process tasks by Deadline date priority?

I note that when a fresh batch of Asteroids@Home are downloaded, that no matter the Deadline date of Einstein@Home is set at, the newly downloaded Asteroids@Home tasks start immediately and places the existing Einstein@Home tasks on 'hold'.

I'd rather set the client to process tasks by Deadline date in chronological order to ensure that each tasks are completed at or before their Deadline date.

Thanks in advance, David

Hi,

I was wondering if anyone has had any experience running the small-form-factor NUC computers with BOINC projects.

The reason I am considering NUCs is that I'm trying to reduce my energy usage. Therefore, my primary concern is not so much the performance. It is the heat. With my past power-hungry desktop machines with GPUs I was always able to run them at 100%, 24/7 because I installed cooling systems as necessary to keep temps within limits.

With off-the-shelf NUCs I'm not so sure that their stock cooling systems and designs would handle BOINC 100%/24/7. I'd hate to have to throttle them to keeps temps in line.

I'm looking at keeping the cost at below $400 per NUC and I will be needing at least two.

Any guidance? Thank you very much!

Tomorrow is Friday again. Let's see if my suspicions about how they handle sending out "new" tasks during weekends proves correct.

Did someone make a mistake? My job-cache was still full after a weekend away.

Linux Mint
Made a change in the preferences and the GUI stopped being responsive, failing to display anything. I tried to recover the change I made so I could report it, but the GUI would not respond and kept blinking. Nothing I tried, even reinstall after complete delete, made any difference. Did a manual delete of every file that had "boinc" in the name after another complete delete as I no longer have the will to continue.

Apoligies for all the project tasks that were incomplete and abandoned.

In the old days, it BOINC was very stable and the linux version did not have any serious problems. BOINC has come a long way and I am unable to cope with the problems. I am uncomfortable using the CLI because of the lack of knowledge of the commands and how they are supposed to perform. The "man" instructions are for the experts. Except for the simplest CLI tasks, the results are from not good to disasterous. Therefore I must give up on BOINC because a single change I made in the GUI crashed so bad there was no way to recover.

Suspend when computer in use and Suspend GPU computing when computer is in use are both unchecked
Suspend when non-BOINC CPU usage is above 33.00% is checked

I am using BOINC 7.20. In the BOINC Manager Advanced view, Projects I have two projects where the Remove button is unavailable: World Community Grid and Rosetta@home. These two projects were added automatically when I installed the BOINC software. I already had accounts for them but they were not manually added. I thought maybe that was significant.

Can someone please tell me if there is a way to fix the Remove button. It's an issue because I keep getting a message in Notices This project seems to have changed its URL. When convenient, remove the project, then add http://www.rnaworld.de/rnaworld/

Thanks

Hi,

I have Ubuntu server 22.04 with boinc running as daemon, installed from git.

Boinc client reports 97.87GB disk size and 53.76GB free, but I have 2TB disk. So I am unable to get new tasks.
I have no idea why this is happening.
I don't know if it has anything to do with it, but I also installed cvmfs, go and singularity to run native ATLAS tasks.

Any help would be appreciated!

Kind regards,
Nejc



user@user:/$ df -h -T gives:
Filesystem Type Size Used Avail Use% Mounted on
tmpfs tmpfs 51G 60M 51G 1% /run
/dev/mapper/ubuntu--vg-ubuntu--lv ext4 98G 40G 54G 43% /
tmpfs tmpfs 252G 0 252G 0% /dev/shm
tmpfs tmpfs 5.0M 0 5.0M 0% /run/lock
/dev/nvme0n1p2 ext4 2.0G 430M 1.4G 24% /boot
/dev/nvme0n1p1 vfat 1.1G 6.1M 1.1G 1% /boot/efi
tmpfs tmpfs 51G 8.0K 51G 1% /run/user/1000
cvmfs2 fuse 40G 1.8G 39G 5% /cvmfs/cvmfs-config.cern.ch
cvmfs2 fuse 40G 1.8G 39G 5% /cvmfs/atlas.cern.ch
cvmfs2 fuse 40G 1.8G 39G 5% /cvmfs/sft.cern.ch
cvmfs2 fuse 40G 1.8G 39G 5% /cvmfs/atlas-condb.cern.ch
cvmfs2 fuse 40G 1.8G 39G 5% /cvmfs/sft-nightlies.cern.ch
cvmfs2 fuse 40G 1.8G 39G 5% /cvmfs/atlas-nightlies.cern.ch
cvmfs2 fuse 40G 1.8G 39G 5% /cvmfs/unpacked.cern.ch


user@user: sudo fdisk -l gives (shortened):
Disk /dev/loop0: 63.28 MiB, 66355200 bytes, 129600 sectors
Disk /dev/loop1: 63.34 MiB, 66412544 bytes, 129712 sectors
Disk /dev/loop2: 111.95 MiB, 117387264 bytes, 229272 sectors
Disk /dev/loop3: 49.84 MiB, 52260864 bytes, 102072 sectors
Disk /dev/loop4: 53.24 MiB, 55824384 bytes, 109032 sectors
Disk /dev/nvme0n1: 1.82 TiB, 2000398934016 bytes, 3907029168 sectors
Disk model: Samsung SSD 980 PRO 2TB
Disklabel type: gpt

Device Start End Sectors Size Type
/dev/nvme0n1p1 2048 2203647 2201600 1G EFI System
/dev/nvme0n1p2 2203648 6397951 4194304 2G Linux filesystem
/dev/nvme0n1p3 6397952 3907026943 3900628992 1.8T Linux filesystem

A couple of weeks ago WCG did make an announcement that the stats pages would be out of synch for some time as they were having to rebuild a lot of users & computers stats data from scratch.

IMHO, out of sync is not the same as missing devices. I have devices registered before 2019 that are crunching away but not listed. Users are posting their host ID's and the staff is forwarding the ID for manual entry into the database. That is NFG, and they need help.

After fighting with the repository (Kubuntu 22.04) version of BOINC Manager, I gave up and installed the flatpak version this past weekend. That's working just fine -- except I noticed that my CPU fan wasn't running at high level like it should when my Fx8350 is going full blast on all eight cores. A little poking, and I've determined that it runs fine per preferences when the BOINC Manager is open, but if I close the manager (either with the x button upper right, or with File, Close Window), computation stops.

I *think* this means the client service isn't running? The command that would tell the client to start on boot in the .deb install (sudo systemctl enable boinc-client) returns "Failed to enable unit: Unit file boinc-client.service does not exist." I presume this is because the flatpak doesn't offer a system service. There must be a flatpak command that I can put in my autostart so the client runs independent of the manager -- but I don't know what it is. flatpak run edu.berkeley.BOINC launches the manager (which starts the client), but the client ought to run in background whether the manager is running or not.

Or is this one of the ways in which the flatpak version isn't quite ready for prime time?

...it will take about 74 hours.

Sounds good, but don't assume that is around the clock crunching. Spread over 8 hour days and no weekends or a standard 40 hour week, that could be 2 more weeks. So maybe 4/10?