I started posting in the 32 bit thread but figured I'd bring my issue here, since it's still not resolved.

Just got this, despite work being available according to server status. Followed the instructions in the prudent thread, typed in the following before installing Boinc 7.16.3:

sudo apt-get install lib32ncurses5 lib32z1 lib32stdc++-6-dev


12/19/2019 1:59:14 AM | climateprediction.net | Sending scheduler request: Project initialization.
12/19/2019 1:59:14 AM | climateprediction.net | Requesting new tasks for CPU
12/19/2019 1:59:15 AM | climateprediction.net | Scheduler request completed: got 0 new tasks
12/19/2019 1:59:15 AM | climateprediction.net | No tasks sent

This is on a fresh Ubuntu 18.04.3 install. All OS upgrades complete as well.
Same OS on 2 other machines having the same issues, one since September. I took it to mean no tasks available for Linux, but didn't really pay it much mind until I noticed this. This is on a machine, the 2600, with more log flags enabled. Does this help?
12/18/2019 11:39:54 PM | climateprediction.net | [sched_op] Starting scheduler request
12/18/2019 11:39:54 PM | climateprediction.net | Sending scheduler request: To fetch work.
12/18/2019 11:39:54 PM | climateprediction.net | Requesting new tasks for CPU
12/18/2019 11:39:54 PM | climateprediction.net | [sched_op] CPU work request: 13588114.08 seconds; 8.00 devices
12/18/2019 11:39:55 PM | climateprediction.net | Scheduler request completed: got 0 new tasks
12/18/2019 11:39:55 PM | climateprediction.net | [sched_op] Server version 713
12/18/2019 11:39:55 PM | climateprediction.net | No tasks sent
12/18/2019 11:39:55 PM | climateprediction.net | Project requested delay of 3636 seconds
12/18/2019 11:39:55 PM | climateprediction.net | [sched_op] Deferring communication for 01:00:36
12/18/2019 11:39:55 PM | climateprediction.net | [sched_op] Reason: requested by project



The machines in question are 1496269, 1496255, and 1493090.
What other information can I provide? I guess this Ryzen 1700x will be crunching another project for the time being. 64 GB of memory and running on an ssd complete this.
I'd love to contribute more to this project but for some reason something is failing to compute as it were on my Linux machines.

Truthfully it's a little frustrating that it could be as simple as a library not installed, but you would think tasks would at least download and fail. I'm not even getting that. I hope this project can get around to being 64 bit soon.

Any help appreciated.

I started posting in the 32 bit thread but figured I'd bring my issue here, since it's still not resolved.

Agreed, nothing to do with 32 bit libraries. There are machines out there that don't have them, download tasks and crash everything.

Your computer https://www.cpdn.org/show_host_detail.php?hostid=1496269 Is showing number of times client has contacted server as 0 but that doesn't explain the other two. I am running 7.16.3 so I don't see how that can be an issue.
Edit: lots of others are using it so any minor differences from having rolled my own from the source at Git Hub won't be the cause.


I am clutching at straws here. Are your windows machines at the same location? If not I wondered if it is something to do with your ip address being part of a block that has been shut off by CPDN/.

I don't really think this very likely but if these machines aren't chocker with work from other projects which I am pretty sure they aren't from what you have said, like yourself I am running out of ideas.

Truthfully it's a little frustrating that it could be as simple as a library not installed, but you would think tasks would at least download and fail. I'm not even getting that. I hope this project can get around to being 64 bit soon.

Very, very strange. I assume you still had all the Ubuntu updates enabled when you did the 32-bit install? I usually disable them at the end, and then if I try to install something it may not work.

Otherwise, I would detach from the project and try again, hopefully with a new computer ID. It seems that the server thinks it has just sent you something, and is doing the usual one-hour back-off.

Agreed, nothing to do with 32 bit libraries. There are machines out there that don't have them, download tasks and crash everything.

Your computer https://www.cpdn.org/show_host_detail.php?hostid=1496269 Is showing number of times client has contacted server as 0 but that doesn't explain the other two. I am running 7.16.3 so I don't see how that can be an issue.
Edit: lots of others are using it so any minor differences from having rolled my own from the source at Git Hub won't be the cause.


I am clutching at straws here. Are your windows machines at the same location? If not I wondered if it is something to do with your ip address being part of a block that has been shut off by CPDN/.

I don't really think this very likely but if these machines aren't chocker with work from other projects which I am pretty sure they aren't from what you have said, like yourself I am running out of ideas.

Nope, the Ryzen was freshly added as of 10 or so hours ago and CPDN was the first project to be thrown at it, with Universe at home at 0 resource and CPDN at 300.
These are machines running in a Hetzner datacenter, I'm not sure if this would make any difference in terms of IP as I do have a Hetzner Windows machine (though all the IPs are different blocks I think) but the Windows machine gets work without issues when work is available that is.

Very, very strange. I assume you still had all the Ubuntu updates enabled when you did the 32-bit install? I usually disable them at the end, and then if I try to install something it may not work.

Otherwise, I would detach from the project and try again, hopefully with a new computer ID. It seems that the server thinks it has just sent you something, and is doing the usual one-hour back-off.

Forgive the perhaps stupid question, but what do you mean by disable updates? I'm not sure how to go about doing that - I'm assuming you meant disable updates for Ubuntu itself after I install the 32 bit libraries?
Should I reset the project, then remove it? Maybe I'll detach from BAM too just to be on the safe side. Of course, my lowest end machine with a core 2 duo is sitting here getting tasks without issue on 7.9.3 (or whatever the latest Ubuntu has to offer), at least as of 18 or so hours ago, and I don't understand why.

Sorry for the double post - I can't find an edit option. Time to grab a beer and take a break from messing with this - literally just found the edit option after I posted this. Clearly my brain is dead.


So since crunching is all these are used for, I just completely wiped this machine [url]https://www.cpdn.org/show_host_detail.php?hostid=1493090-/url] and installed the libraries plus 7.9.3. Apparently when I attached again, this time without bam or any other projects, cpdn still sees it under the same ID as last time, despite resetting it before I removed it previously, so now I got a communication deferred for 58 minutes message and we'll see what happens next.
Hetzner offers 2 versions of Ubuntu 18.04 - one called Minimal, which is what I've been using and another that also includes nextcloud, which I'm trying now. I highly doubt this is the issue, but I have fingers crossed.
If it still doesn't fetch tasks, at this point I'm at a complete and utter loss. There are no firewalls installed, all ports are open, and any other project I attach to gets work, including Rosetta, WCG, Universe, even LHC provided everything is correctly installed.
I'm mainly renting these due to lack of physical space and for $28 a month for a 4770, it isn't horrible. Of course long term it will get expensive, but right now there is also a Ryzen 3600 on offer at $43.
Of course if I can't get these crunching here at all....
Would I break anything if I tried installing the needed libraries for Ubuntu 1604 as well? Just grasping at straws here at this point.
It also doesn't help I'm so very new to Linux and all it has to offer.

Edit 2: Looks like something I did forced it to start actually downloading tasks. This is fantastic! I wonder if starting from Boinc 7.9.3 forced it into action. I'll upgrade to 7.16.3 when these tasks finish and see if fetching continues, which I have no doubt it should. I'll do the same on the other two machines that are currently struggling to receive work.

Forgive the perhaps stupid question, but what do you mean by disable updates? I'm not sure how to go about doing that - I'm assuming you meant disable updates for Ubuntu itself after I install the 32 bit libraries?

It is done via the check boxes in "Software & Updates". It looks like you have not done it, so don't have to worry about it.

Should I reset the project, then remove it? Maybe I'll detach from BAM too just to be on the safe side.

Generally, yes that is the idea. But I don't use BAM, and don't know anything about it. Just make sure you fully detach, and try again.

Generally, yes that is the idea. But I don't use BAM, and don't know anything about it. Just make sure you fully detach, and try again.

I tried that before reinstalling completely and for some odd reason, it was still failing to download tasks.
I have a machine here locally that I need to get some ram for and I'll test that theory. Update Boinc to 7.16.3 before attaching to the project. I doubt it will make a bit of difference and am assuming something was off in the first Ubuntu installs.
The other difference is there is no bam this time and likely won't be in my future either. The interface is awful now and I'll just use boinc tasks to manage everything.

I am reading every post and will keep on reading. I have installed Linux on two laptops. Two instances of Mint OS and one of Zorin OS. Zorin on command line downloaded 32-bit lib but Mint has a sullen face in both the instances. I am not being able to get WU's on any VM? I will check again in the morning, past midnight here.
Any light that can be shed would be much appreciated.

It would also be worth a look at the special threads in Number crunching for the HADAM4 models, especially the N216 model.
They require LOTS of memory.

For sure. And they are not the worst: some Rosetta@hoe tasks are even bigger. I even doubled my RAM size from 32 Gigabytes to 64 Gigabytes to deal with those

top - 20:13:24 up 8 days,  3:02,  1 user,  load average: 3.47, 3.31, 3.22
Tasks: 451 total,   4 running, 446 sleeping,   1 stopped,   0 zombie
%Cpu(s):  0.3 us,  0.1 sy, 18.7 ni, 79.0 id,  1.8 wa,  0.1 hi,  0.0 si,  0.0 st
MiB Mem :  63944.0 total,    925.2 free,   8144.1 used,  54874.7 buff/cache
MiB Swap:  15992.0 total,  15934.2 free,     57.8 used.  54989.1 avail Mem 

    PID    PPID USER      PR  NI S    RES    SHR  %MEM  %CPU  P     TIME+ COMMAND                                                       
   5527    5508 boinc     39  19 R   1.4g  19872   2.2   6.2  3  11628:51 /var/lib/boinc/projects/climateprediction.net/hadam4_um_8.52+ 
  22591   22588 boinc     39  19 R   1.4g  19812   2.2   6.2  4  11017:32 /var/lib/boinc/projects/climateprediction.net/hadam4_um_8.52+ 
  25133   25130 boinc     39  19 R   1.4g  19912   2.2   6.2  7  10941:17 /var/lib/boinc/projects/climateprediction.net/hadam4_um_8.52+ 
  25164   25162 boinc     39  19 T   1.3g  19924   2.2   0.0  7  10103:13 /var/lib/boinc/projects/climateprediction.net/hadam4_um_8.52+ 
   2022       1 boinc     30  10 S  46492  17176   0.1   0.0 13  54224:53 /usr/bin/boinc                                                
  22588    2022 boinc     39  19 S  19468  17136   0.0   0.0 12   5:26.28 ../../projects/climateprediction.net/hadam4_8.52_i686-pc-lin+ 
   5508    2022 boinc     39  19 S  19356  17024   0.0   0.0  5   5:46.41 ../../projects/climateprediction.net/hadam4_8.52_i686-pc-lin+ 
  25162    2022 boinc     39  19 S  19308  16980   0.0   0.0  1   5:27.86 ../../projects/climateprediction.net/hadam4_8.52_i686-pc-lin+ 
  25130    2022 boinc     39  19 S  19304  16968   0.0   0.0 12   5:30.69 ../../projects/climateprediction.net/hadam4_8.52_i686-pc-lin+ 

Notice the RES is about 1.4 gIGig, but this is only near the very end. Most of the time it is "'ONLY" 1.3 Gig.

---

I am once again having this problem.
Machines 1524147 and 1524121 are not able to download work, despite work being available. Tried rebooting without success.
Ubuntu 20.04 fresh installs on both. The following libraries installed as per the 32 bit on 64 bit linux systems.
lib32ncurses6 lib32z1 lib32stdc++-7-dev.

I will have more information from the event log now that I've enabled more debug options, but right now I simply am seeing this. I do apologize if this is too much info.

11/15/2021 6:12:21 PM | climateprediction.net | Project requested delay of 3636 seconds
11/15/2021 7:13:00 PM | climateprediction.net | No tasks sent
11/15/2021 7:13:00 PM | climateprediction.net | Project requested delay of 3636 seconds
11/15/2021 8:13:40 PM | climateprediction.net | No tasks sent
11/15/2021 8:13:40 PM | climateprediction.net | Project requested delay of 3636 seconds
11/16/2021 3:44:55 PM | climateprediction.net | No tasks sent
11/16/2021 3:44:55 PM | climateprediction.net | Project requested delay of 3636 seconds
11/16/2021 11:39:31 PM | | Re-reading cc_config.xml
11/16/2021 11:39:31 PM | | Config: GUI RPC allowed from any host
11/16/2021 11:39:31 PM | | Config: GUI RPCs allowed from:
11/16/2021 11:39:31 PM | | log flags: file_xfer, sched_ops, task, cpu_sched, cpu_sched_debug, work_fetch_debug
11/16/2021 11:39:31 PM | | [cpu_sched_debug] Request CPU reschedule: Core client configuration
11/16/2021 11:39:31 PM | | [work_fetch] Request work fetch: Core client configuration
11/16/2021 11:39:31 PM | | [cpu_sched_debug] schedule_cpus(): start
11/16/2021 11:39:31 PM | | [cpu_sched_debug] enforce_run_list(): start
11/16/2021 11:39:31 PM | | [cpu_sched_debug] preliminary job list:
11/16/2021 11:39:31 PM | | [cpu_sched_debug] final job list:
11/16/2021 11:39:31 PM | | [cpu_sched_debug] enforce_run_list: end
11/16/2021 11:39:32 PM | | choose_project(): 1637127572.607807
11/16/2021 11:39:32 PM | | [work_fetch] ------- start work fetch state -------
11/16/2021 11:39:32 PM | | [work_fetch] target work buffer: 43200.00 + 43200.00 sec
11/16/2021 11:39:32 PM | | [work_fetch] --- project states ---
11/16/2021 11:39:32 PM | climateprediction.net | [work_fetch] REC 0.000 prio -0.000 can request work
11/16/2021 11:39:32 PM | | [work_fetch] --- state for CPU ---
11/16/2021 11:39:32 PM | | [work_fetch] shortfall 87971.46 nidle 0.00 saturated 68116.50 busy 0.00
11/16/2021 11:39:32 PM | climateprediction.net | [work_fetch] share 0.500
11/16/2021 11:39:32 PM | | [work_fetch] ------- end work fetch state -------
11/16/2021 11:39:32 PM | climateprediction.net | choose_project: scanning
11/16/2021 11:39:32 PM | climateprediction.net | can fetch CPU
11/16/2021 11:39:32 PM | | [work_fetch] No project chosen for work fetch
11/16/2021 11:40:31 PM | | [cpu_sched_debug] Request CPU reschedule: periodic CPU scheduling
11/16/2021 11:40:31 PM | | [cpu_sched_debug] schedule_cpus(): start
11/16/2021 11:40:31 PM | | [cpu_sched_debug] enforce_run_list(): start
11/16/2021 11:40:31 PM | | [cpu_sched_debug] preliminary job list:
11/16/2021 11:40:31 PM | | [cpu_sched_debug] final job list:
11/16/2021 11:40:31 PM | | [cpu_sched_debug] enforce_run_list: end
11/16/2021 11:40:32 PM | | choose_project(): 1637127632.758010
11/16/2021 11:40:32 PM | | [work_fetch] ------- start work fetch state -------
11/16/2021 11:40:32 PM | | [work_fetch] target work buffer: 43200.00 + 43200.00 sec
11/16/2021 11:40:32 PM | | [work_fetch] --- project states ---
11/16/2021 11:40:32 PM | climateprediction.net | [work_fetch] REC 0.000 prio -0.000 can request work
11/16/2021 11:40:32 PM | | [work_fetch] --- state for CPU ---
11/16/2021 11:40:32 PM | | [work_fetch] shortfall 88175.35 nidle 0.00 saturated 68075.60 busy 0.00
11/16/2021 11:40:32 PM | climateprediction.net | [work_fetch] share 0.500
11/16/2021 11:40:32 PM | | [work_fetch] ------- end work fetch state -------
11/16/2021 11:40:32 PM | climateprediction.net | choose_project: scanning
11/16/2021 11:40:32 PM | climateprediction.net | can fetch CPU
11/16/2021 11:40:32 PM | | [work_fetch] No project chosen for work fetch
11/16/2021 11:41:31 PM | | [cpu_sched_debug] Request CPU reschedule: periodic CPU scheduling
11/16/2021 11:41:31 PM | | [cpu_sched_debug] schedule_cpus(): start
11/16/2021 11:41:31 PM | | [cpu_sched_debug] enforce_run_list(): start
11/16/2021 11:41:31 PM | | [cpu_sched_debug] preliminary job list:
11/16/2021 11:41:31 PM | | [cpu_sched_debug] final job list:
11/16/2021 11:41:31 PM | | [cpu_sched_debug] enforce_run_list: end
11/16/2021 11:41:33 PM | | choose_project(): 1637127692.995801
11/16/2021 11:41:33 PM | | [work_fetch] ------- start work fetch state -------
11/16/2021 11:41:33 PM | | [work_fetch] target work buffer: 43200.00 + 43200.00 sec
11/16/2021 11:41:33 PM | | [work_fetch] --- project states ---
11/16/2021 11:41:33 PM | climateprediction.net | [work_fetch] REC 0.000 prio -0.000 can request work
11/16/2021 11:41:33 PM | | [work_fetch] --- state for CPU ---
11/16/2021 11:41:33 PM | | [work_fetch] shortfall 88362.31 nidle 0.00 saturated 68042.88 busy 0.00
11/16/2021 11:41:33 PM | climateprediction.net | [work_fetch] share 0.500
11/16/2021 11:41:33 PM | | [work_fetch] ------- end work fetch state -------
11/16/2021 11:41:33 PM | climateprediction.net | choose_project: scanning
11/16/2021 11:41:33 PM | climateprediction.net | can fetch CPU
11/16/2021 11:41:33 PM | | [work_fetch] No project chosen for work fetch

Edit: Since I for the first time installed minimal Ubuntu, I went ahead and installed the full desktop packages.
sudo apt install ubuntu-desktop
We will see what difference this brings, if any, over night. All other projects I have attached work without issue.

@wolfman1360

I'm not sure what is going on. First things first on another of your computers, your Core 2 8400 is requesting work and all tasks are failing. The stderr on the failed task webpages states a missing library. Try to reinstall the needed compatibility libraries for that version of Ubuntu.

On the i7 4770, the benchmark scores are 1 billion ops/sec for both floating point and integer. While this would not be causing your work fetch problem, it would give ridiculously large estimates for time to task completion in boinc manager. Run a boinc benchmark when you get a chance.

How many other projects are the two work fetch problem computers attached to? And did you attach to the other projects before cpdn?

Perhaps suspend all other projects in boinc manager and then allow boinc to try to request work from cpdn. I'm not sure if that will do anything, but it's worth a try.

@wolfman1360

I'm not sure what is going on. First things first on another of your computers, your Core 2 8400 is requesting work and all tasks are failing. The stderr on the failed task webpages states a missing library. Try to reinstall the needed compatibility libraries for that version of Ubuntu.

Libraries installed, my apologies. That's a machine I generally set and forget - last I checked it was getting work just fine, but I think I absently installed Ubuntu 20.04 and forgot to update those.

On the i7 4770, the benchmark scores are 1 billion ops/sec for both floating point and integer. While this would not be causing your work fetch problem, it would give ridiculously large estimates for time to task completion in boinc manager. Run a boinc benchmark when you get a chance.

Will do right now. Benchmarks are in progress.

How many other projects are the two work fetch problem computers attached to? And did you attach to the other projects before cpdn?

The i7-4790 is aalso attached to WCG. The 4770 is attached to nothing but CPDN and wcg was attached before CPDN. Oddly enough a tiny little i3-2130 that I just attached recently is getting work alongside WCG.

Perhaps suspend all other projects in boinc manager and then allow boinc to try to request work from cpdn. I'm not sure if that will do anything, but it's worth a try.

I can do this. Left both machines sitting over night - the 4770 is still completely idle with work fetches attempted throughout the remaining time.
I really don't understand what is going on or what I am missing to make this work.

I am no good at reading that debug log stuff unfortunately. I don't understand this line at all:

11/16/2021 11:41:33 PM | | [work_fetch] No project chosen for work fetch

but it doesn't sound logical. When programmers write the debug checks/logging options, I'm sure they understand the output, but too often it is meaningless to me.

This is my output with no debugging options enabled. This tells me nothing.
11/17/2021 11:15:05 AM | climateprediction.net | No tasks sent
11/17/2021 11:15:05 AM | climateprediction.net | Project requested delay of 3636 seconds
11/17/2021 12:15:42 PM | climateprediction.net | No tasks sent
11/17/2021 12:15:42 PM | climateprediction.net | Project requested delay of 3636 seconds
This is on a machine with no other projects running, all benchmarks ran, all libraries installed that I know of. It doesn't even try and fetch work to fail later.
I don't know how to fix this. There are over 4000 WUs available.

Update - just randomly installed nextcloud. The next check for updates work has downloaded - 1 N216 - and begun crunching with, so far, no errors.
Your guess is as good as mine, at this point. I will test on the 4790.

Hi,

I'm trying to run CPDN on a linux Debian 10 host with 4 Gb RAM only, CPDN is not sending any work for days.

CPDN freshly added to the host that has been crunching many other projects for years.

I tried

sudo apt-get install lib32ncurses5 lib32z1 lib32stdc++-6-dev

but it claims none of these can be found

Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Unable to locate package lib32ncurses5
E: Unable to locate package lib32stdc++-6-dev
E: Couldn't find any package by regex 'lib32stdc++-6-dev'

I see HadAM4 at N216 is supposed to be available and is linux only.

Note I had tried on a local linux VM on my iMac but no task sent neither.

ID de l'ordinateur	Nom	Modèle	Localisation	Crédit moyen	Crédit total	BOINC
version	CPU	GPU	Système d'exploitation	Dernier contact

ID: 1525909 
7.14.2 Intel Core Processor (Haswell, no TSX) [Family 6 Model 60 Stepping 1] (2 processeurs)
Linux Debian GNU/Linux 10 (buster) [4.19.171|libc 2.28 (Debian GLIBC 2.28-10)]	
20 Jan 2022, 11:17:44 UTC

ID: 1525888 
7.14.2 Intel(R) Core(TM) i9-10910 CPU @ 3.60GHz [Family 6 Model 165 Stepping 5] (8 processeurs)
Linux Debian GNU/Linux 10 (buster) [4.19.0-18-amd64|libc 2.28 (Debian GLIBC 2.28-10)]	
17 Jan 2022, 22:37:11 UTC

Any idea ?

Thanks.

In this post

https://www.cpdn.org/forum_thread.php?id=8916#62038

there are commands for installing the right 32 bit libraries in Debian. Whether that will cover whatever version you are using, I'm not sure. But give it a try.

Right now, the Macs can only run the hadcm3s type tasks and there are none in the queue, and may not be for awhile again.

I suspect the no work being sent is not down to missing 32 bit libraries as lots of machines without them are downloading tasks and trashing them. Make sure your manager is set to advanced view and in options>event log options> check work fetch debug then post what you get in the event log when your machine requests work.

Under projects tab trigger a manual update and then see if there are any clues in the event log. - If it doesn't help, post the output here and see if anyone else can find some clues there.

How many tasks are running from other projects? It may be that BOINC thinks you don't need any more work. To test that theory, suspend all the tasks from other projects before doing an update.

Also once you get this sorted, don't try running too many tasks at once. With only 8GB of RAM, running all 8 cores on your I9 is likely to make things go very slowly!

Thank you to both : I think I had to install the 32 bits libs so anyway your link is very useful @geophi, and @Dave regarding getting task I start to struggle with the beast : I have a non GUI command line only Debian hosted VM, I use boinctui easy small manager but not strong enough to change the log flags, I found the doc and could edit the cc_config, then I did fight to re-read the config because native boinccmd option would fail with a strange error message, so I had to find how to restart the boinc-client service, and then without doing anything else boinc suddenly decided to download a task when I force updated the project once again !!!

It has now been crunching for the past 10 mn without crashing so I think the lib installation is OK too, thanks a lot to both !

(I'm not using the linux VM in my Mac at the moment, I have a win10 VM using those threads instead for another win-only boinc project ;) )

FYI I now how 2 tasks happily running and the limited resources of this small VM seem sufficient to handle them :

              total        used        free      shared  buff/cache   available
Mem:        3945716     2866124      167744       10664      911848      828136
Swap:      10485756       92856    10392900
(unit kb)