Happy to and thanks!

When running on our bigger research repository which includes saving checkpoints and uploading to S3, the training ends with errors as shown below and a Killed message for the main process (I do not abort the main process manually):

2023-01-26 17:37:17,527 INFO: Save the latest model.
2023-01-26 17:37:19,158 - clearml.storage - INFO - Starting upload: /tmp/.clearml.upload_model_cvqpor8r.tmp => glass-clearml/RealESR/Glass-ClearML Demo/[Lambda] FMEN distributed check, v10 fileserver u...

Although, some correction here: While the secret is indeed hidden in the logs, it is still visible in the “execution” tab of the experiment, see two screenshots below.
One again I set them with
task.set_base_docker(docker_arguments=["..."])

Sorry that these issues go quite deep and chaotic - we would appreciate any help or ideas you can think of!

Yes for example, or some other way to get credentials over to the container safely without them showing up in the checked-in code or web UI

Results of a bit more investigation:

The ClearML example does use the Pytorch dist package but none of the DistributedDataParallel functionality, instead, it reduces gradients “manually”. This script is also not prepared for torchrun as it launches more processes itself (w/o using the multiprocessing of Python or Pytorch.)

When running a simple example (code attached...

Hi John, thanks for getting back to me!
So it shows up in the UI like shown below. It happens both when “recording” the local run on Mac and on Linux.

Hi AgitatedDove14 , so it took some time but I’ve finally managed to reproduce. The issue seems to be related to writing images via Tensorboard:
` from torch.utils.tensorboard import SummaryWriter
import torch
from clearml import Task, Logger

if name == "main":
task = Task.init(project_name="ClearML-Debug", task_name="[Mac] TB Logger, offline")
tb_logger = SummaryWriter(log_dir="tb_logger/demo/")
image_tensor = torch.rand(256, 256, 3)
for iter in range(10):
t...

Hi @<1523701087100473344:profile|SuccessfulKoala55> , sorry there was a mistake on my end - clearml.conf pointed to the wrong URL 🙈

By the way, if we don’t wrap other calls in is_offline() we get errors like “DateTime object is not serializable”, but that’s a secondary issue.

Hi SuccessfulKoala55 , thanks for getting back to me!
In the docs of Task.set_base_docker() it says “When running remotely the call is ignored”. Does that mean that this function call is executed when running locally to “record” the arguments and then when I duplicate the experiment and clone it remote, the call is ignored and the recorded values are used?

AgitatedDove14 maybe to come at this from a broader angle:
Is ClearML combined with DataParallel or DistributedDataParallel officially supported / should that work without many adjustments? If so, would it be started via python ... or via torchrun ... ? What about remote runs, how will they support the parallel execution? To go even deeper, what about the machines started via ClearML Autoscaler? Can they either run multiple agents on them and/or start remote distribu...

@<1523701070390366208:profile|CostlyOstrich36> thank you, now everything works so far!
Last thing: Is there any way to change all the links in the new ClearML server such that an artifact that was previous under s3://… is now taken from gs://… ? The actual data is already available under the gs:// link of course

SuccessfulKoala55 AgitatedDove14 So I’ve tried the approach and it does work, however, this of course results in the credentials being visible in the ClearML web interface output, which comes close to just hard-coding them in…
Is there any way to send the secrets safely?
Is there any way to access the clearml.conf file contents from within code? (afaik, the file does not get send over to the container - otherwise I could just yml-read it myself…)

Is there for example a good way to start the new version without any legacy data base and then migrate the data more or less manually? Is it enough to migrate the Mongo DB to see our old tasks again and are the schematics still compatible?

Hi Jake, yes I’d love to! Just a question: how clean and complete does the example need to be? For example, this code relies on you building a correct Machine Image on GCP (which is somewhat unrelated to ClearML) and it does not get the logs from the agent instances - is that still good enough?

So AgitatedDove14 if we use the CLEARML_OFFLINE_MODE environment variable instead the program runs through again.
The only thing is that now we get errors of the form
` 0%| | 0/18 [00:00<?, ?image/s]ClearML running in offline mode, session stored in /home/manuel/.clearml/cache/offline/offline-167ceb1cd3c946df8abc7206b781b486
2022-11-07 07:49:06,986 - clearml.metrics - WARNING - Failed uploading to /home/manuel/.clearml/cache/offline/offline-167ceb1cd3c946df8abc7206b781b486/...

New state: After starting with the old YML again, the web app looks new (presumably because the image allegroai/clearml:latest is used), but the server version still lists WebApp: 2.1.0-664 • Server: 1.9.2-317 • API: 2.23 .

Creating tasks and reporting things works again, but I regularly see the UI error shown attached. Any way to resolve things?

Sorry to ask again, but the values are still showing up in the WebUI console logs this way (see screenshot.)
Here is the config that I paste into the EC2 Autoscaler Setup:
` agent {
extra_docker_arguments: ["-e AWS_ACCESS_KEY_ID=XXXXXX", "-e AWS_SECRET_ACCESS_KEY=XXXXXX"]

hide_docker_command_env_vars {
    enabled: true
    extra_keys: ["AWS_SECRET_ACCESS_KEY"]
    parse_embedded_urls: true
}    

} `Never mind, it came from setting the options wrong, it has to be ...

@<1523701070390366208:profile|CostlyOstrich36> , you mean the ClearML server needs access to Cloud Storage in its clearml.conf file?
Just tried it by creating a ~/clearml.conf file and setting the entry as below - unfortunately the same result. I’ve re-started the docker-compose of course.

Did I miss something here?

    google.storage {
        credentials_json: "/home/.../my-crendetials.json"
    }

That was the missing piece - thank you!
Awesome to all the details you have considered in ClearML 😉

Hi @<1523701070390366208:profile|CostlyOstrich36> , thank you for answering!
We are upgrading from v. 1.9 or so (I think) to the most recent one.

Attached below are 3 logs from api server, elastic search and file server - does this help to debug?

Also as an update, I tried to start the containers one by one and resolve the errors that came up. The only real one I found was that Redis crashed when loading the previous data base. Since I figured I wouldn't necessarily need the cache, I cleared the dump file, then all the services started - this all refers to the old server version.

When trying the same with the most recent docker-compose.yml , the services all started by themselves, but when I start the full docker compose, the das...

Ok great! I will debug starting with a simpler training script.
Just as a last question, is torchrun also supported rather than the (now deprecated but still usable) torch.distributed.launch ?

To recap, the server started up on GCP as expected before migrating the data over. The migration was done by

• deleting the current data sudo rm -fR /opt/clearml/data/*
• unpacking the backup sudo tar -xzf ~/clearml_backup_data.tgz -C /opt/clearml/data
• setting permissions sudo chown -R 1000:1000 /opt/clearml

Unfortunately not, task.data.output just contains <tasks.Output: { "destination": " s3://some_bucket " }> and when I convert task.data to a string and search for the desired uri, I cannot find it either.
But on the other hand, putting the url together from its name, id, etc. seems to work - it might be a little unsafe if the task gets re-named or something, but otherwise it should be fine.

Yes and yes - is that the issue and it might likely go away if we host it via HTTPS?

Ahhh, ok got it! Thanks 👍

SuccessfulKoala55 just in case you have any more thoughts, but we could also continue as is 😊

Yes totally, but we’ve been having problems of getting these GPUs specifically (even manually in the EC2 console and across regions), so I thought maybe it’s easier to get one big one than many small ones, but I’ve never actually checked if that is true 🙂 Thanks anyhow!