balena-labs-projects / rosetta-at-home Goto Github PK

I've next logs from a RPI 4 1Gb when I push the code from master:

19.04.20 15:28:37 (+0200)  boinc-client  19-Apr-2020 13:28:37 [Rosetta@home] Rosetta needs 1907.35 MB RAM but only 965.88 MB is available for use.
19.04.20 15:28:37 (+0200)  boinc-client  19-Apr-2020 13:28:37 [Rosetta@home] Rosetta needs 1716.61 MB RAM but only 965.88 MB is available for use.
19.04.20 15:29:05 (+0200)  boinc-client  19-Apr-2020 13:29:05 [---] Project communication failed: attempting access to reference site
19.04.20 15:29:07 (+0200)  boinc-client  19-Apr-2020 13:29:07 [---] Internet access OK - project servers may be temporarily down.

Has anybody the same problem?

It's recommended to use a PWM fan (like the Noctua nf-a4x20 5V PWM fan) with the Jetson Nano board to avoid damaging it under normal operation (especially if used for long period of time, like rosetta-at-home).

Having a PWM fan is of course optional, but when we have one, it'd be best to leverage it to keep the board cooler if possible.

In the case of balenaOS on the nano, I noticed at the momento it doesn't seem to activate the PWM fan (tested yesterday, FAN didn't run at all, not even slowly).

Some software logic need to be used to adjust the fan speed.

On one hand, the temperature of the board can be read via software (just open and read value from /sys/devices/virtual/thermal/thermal_zone0/temp).

On the other hand the PWM fan speed can be adjusted by software by writing an int between 0 (low speed) and 255 (full speed) to /sys/devices/pwm-fan/target_pwm.

For example, if temp is below 20ºC, keep at speed 0, if temp > 50ºC , keep at full speed (255). Between 20 and 50, map/ramp up speed from 0 to 255. See example code here (https://github.com/Pyrestone/jetson-fan-ctl/blob/master/fanctl.py).

PS: I didn't find any way to access the balenaOS with SSH so I couldn't try this below. But it should be doable as it works fine in Linux with the official Nvidia image (meaning that the linux kernel exposes the necesarry API to read/write the temp sensor and the PWM fan speed).

More info:

• https://docs.nvidia.com/jetson/l4t/index.html#page/Tegra%20Linux%20Driver%20Package%20Development%20Guide/power_management_nano.html#wwpID0E6HA
• https://www.jetsonhacks.com/2019/09/08/jetson-nano-add-a-fan/

If you give me instructions to enter the balenaOS via SSH for example and test command, don't hesitate to contact me (here or email)

In the boinctui I can see this message:

This project is using an old URL. When convenient, remove the project, then add https://boinc.bakerlab.org/rosetta/

If you can see the message, It appears if you go to the menu: Projects > Rosetta@home > Update project

I must change this by myself or not?
Thanks

I add an snapshot:

[thgreasi] That's on a fresh DWB installation w/ balenaOS 2.56.0. The boinc-client was running properly but the UI was in a restart loop.
Supervisor logs:

[debug]   Replacing container for service ui because of config changes:
[debug]     Non-array fields:  {"added":{},"deleted":{},"updated":{"memReservation":16777216}}
[debug]   Replacing container for service ui because of config changes:
[debug]     Non-array fields:  {"added":{},"deleted":{},"updated":{"memReservation":16777216}}
[warn]    Ignoring unsupported or unknown compose fields: blkioConfig
[event]   Event: Service install {"service":{"appId":1753819,"serviceId":728349,"serviceName":"ui","releaseId":1560674}}
[event]   Event: Service installed {"service":{"appId":1753819,"serviceId":728349,"serviceName":"ui","releaseId":1560674}}
[event]   Event: Service start {"service":{"appId":1753819,"serviceId":728349,"serviceName":"ui","releaseId":1560674}}
[event]   Event: Service started {"service":{"appId":1753819,"serviceId":728349,"serviceName":"ui","releaseId":1560674}}
[warn]    Ignoring unsupported or unknown compose fields: blkioConfig

This also kept happening even after stopping the bloinc-client and having some RAM being freed.

The email address on the Rosetta team was changed which has invalidated the account key.

Still getting this message in a loop:

Message from server: Rosseta needs 1716.61 MB RAM but only 708.30 MB is available for use.

Any clue?
There is info in project's readme that RPi 3 is between tested devices...

Thanx for help!

Setting the env var ACCOUNT_KEY give the error:
Invalid or missing account key. To fix, remove and add this project.

Hey, after this change in #25 my RPi4 2GB RAM goes well and never run out of task, but I think this board can manage 2 or 3 task simultaneously.

Is possible to add this intermediate config, isn't ?
Thanks in advance.

If a user defines their own account key that is currently put in place of the Fold for Covid key in global_prefs_override.xml, but we do not handle the situation to revert back to the Fold for Covid key if the user removes their own.

Right now, users don't have a way to remotely manage the device or even power cycle it.
Add an interface to manage device

Please integrate: https://github.com/Pyrestone/jetson-fan-ctl
or at least turn the fan at 255

sudo sh -c 'echo 255 > /sys/devices/pwm-fan/target_pwm'

Is there any way of getting ssh on the machine?

The new account_boinc.bakerlab.org_rosetta.xml introduced in #24 is ignored by boinc due to the old account_boinc.rosetta.bakerlab.org.xml persisting on the boinc volume:

This might have lower impact on foldforcovid.io devices for the time being, but account change attempts won't succeed until the old account file is deleted.

With the current implementation of power-management, there are cases where if a laptop is closed and that container stops (for reasons such as update), the device could fall into a sleep state.

To quote @rahul-thakoor "if the power-management service dies, the lock for the file descriptor is released and the action is removed!"

Hi!

I'm looking through the repository but I cannot find any sign of what license this might be using. Is this intentionally unlicensed?

In order to make forums-based debugging easier, we should expose the UUID in the UI and prompt/suggest users report that back to us when opening issues.

cc @chrisys @jtonello

Hi, I'm getting the above on a Raspberry Pi 4 with 4GB of RAM. How can I utilise the maximum amount of RAM available?

Use named volumes to store BOINC data to ensure progress towards tasks is maintained as much as possible.

I'm getting this error when trying to get tasks from Rosetta@Home. What can we recommend to users encountering the same?

Message from server: This project doesn't support computers of type arm-unknown-linux-gnueabihf

I am using a Raspberry Pi 4 4GB RAM.

Hardware        : BCM2835
Revision        : c03111
Serial          : 10000000c45ae232
Model           : Raspberry Pi 4 Model B Rev 1.1

Here is the screenshot of the error:

Hello,

I've successfully completed the setup & my Pi is up & running, however I'm getting messages similar to the following on my web interface:

Rosetta needs 1716.61 MB RAM but only 966.83 MB is available for use.

Any ideas?

which results in reboots.

I tried to investigate whether any client / server config overrides might have happened, but found none. I assume the param should override server settings at least, and below is the global override which contains only cpu usage settings. The file list is in case you needed any other content, I've got it tarballed on my laptop.

/media/piotr/resin-data/docker/volumes/1644552_boinc/_data# ls -l
total 180
-rw-r--r-- 1 root root   926 kwi 22 13:32 account_boinc.bakerlab.org_rosetta.xml
-rw-rw-r-- 1 root root   161 kwi 22 13:32 account_boinc.rosetta.bakerlab.org.xml
-rw-r--r-- 1 root root 53556 kwi 22 10:58 all_projects_list.xml
-rw-r--r-- 1 root root 25846 kwi 22 13:34 client_state_prev.xml
-rw-r--r-- 1 root root 25847 kwi 22 13:34 client_state.xml
-rw-r--r-- 1 root root   325 kwi 22 13:33 coproc_info.xml
-rw-r--r-- 1 root root   120 kwi 22 13:34 daily_xfer_history.xml
-rw-rw-r-- 1 root root   258 kwi 21 22:18 global_prefs_override.xml
-rw-rw-r-- 1 root root     0 kwi 21 22:18 gui_rpc_auth.cfg
drwxr-xr-x 2 root root  1024 kwi 22 10:58 locale
-rw-r--r-- 1 root root     0 kwi 22 10:58 lockfile
-rw-r--r-- 1 root root 13655 kwi 22 11:26 lookup_website.html
-rw-r--r-- 1 root root 23721 kwi 22 10:58 master_boinc.bakerlab.org_rosetta.xml
drwxrwx--x 2 root root  1024 kwi 22 13:33 notices
drwxrwx--x 3 root root  1024 kwi 22 10:58 projects
-rw-r--r-- 1 root root 17375 kwi 22 11:20 sched_reply_boinc.bakerlab.org_rosetta.xml
-rw-r--r-- 1 root root  5972 kwi 22 11:20 sched_request_boinc.bakerlab.org_rosetta.xml
drwxr-xr-x 6 root root  1024 kwi 22 11:46 slots
-rw-r--r-- 1 root root   425 kwi 22 11:20 statistics_boinc.bakerlab.org_rosetta.xml
-rw-r--r-- 1 root root     0 kwi 22 10:58 stderrgpudetect.txt
-rw-r--r-- 1 root root  1020 kwi 22 13:33 stdoutgpudetect.txt
-rw-r--r-- 1 root root  2542 kwi 22 13:34 time_stats_log

/media/piotr/resin-data/docker/volumes/1644552_boinc/_data# cat global_prefs_override.xml
<global_preferences>
   <max_ncpus_pct>100.000000</max_ncpus_pct>
   <ram_max_used_busy_pct>100.000000</ram_max_used_busy_pct>
   <ram_max_used_idle_pct>100.000000</ram_max_used_idle_pct>
   <cpu_usage_limit>100.000000</cpu_usage_limit>
</global_preferences>

When I click on "Deploy with Balena" button on this project home page, it will open my dashboard, then I create a new app, but no release is added. I have to clone in it and deploy it from my computer via balena push

When I plugged in the installation thumb drive, the balena logo came out and the computer shut down. After turning it on again, I pressed the up arrow to open the command prompt and saw that it was stuck on a particular stage.

Look for and rename any references of balenaCloud application to fleet

Due to the use of --fetch_minimal_work used for lower memory devices, after finishing the first task a new one is not requested until a service or device reboot.

We can resolve this with the use of --exit_when_idle which should cause the container to restart and grab a new job.

As the /usr/app/boinc folder became a volume, the changes fixing #25 do not come to life on existing instances:

Only one of the global override files was edited, but this seems to not matter.

Please can we replace the fold for covid banner on the marketing site to this one - https://app.zeplin.io/project/5d5e913bc8e2bb9b4eff9321/screen/5ea950d977920124cf85f104

I need to connect to a socks5 or http Proxy to download the projects

Users are still reporting that 4GB Pi 4s reset when working on 4 tasks. Reports say 3 tasks are OK. 2GB devices are now running stably with 1 task.

Reduce the number of Pi 4 allocated tasks to 3 by setting CPU usage percent to 75.