Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
I Dont Exactly Know How To Ask For Help On This... Nor Have A Reproducible Minimal Example... I Downgraded Back To 1.15.1 From 1.16.2 And Have The Same Issue There. I Have A Pipeline That'S Repeatedly Failing To Complete. It Correctly Marks Things As Cach

i dont exactly know how to ask for help on this... nor have a reproducible minimal example...
I downgraded back to 1.15.1 from 1.16.2 and have the same issue there.
I have a pipeline that's repeatedly failing to complete. it correctly marks things as cached, and then just doesnt execute the last step. The task stays "Running" forever, but disappears - the worker just has a process that dies. CPU/RAM aren't running out or anything like that at all. Prior runs of this pipeline worked just fine (less cached then too).

Has anyone seen this kind of "disappearing / zombie task" state before? It's very perplexing to me.

  
  
Posted one year ago
Votes Newest

Answers 43


hoping this really is a 1.16.2 issue. fingers crossed. at this point more pipes are failing than not.

  
  
Posted one year ago

(the "magic" of the env detection is nice but man... it has its surprises)

  
  
Posted one year ago

the workers connect to the clearml server via ssh-tunnels, so they all talk to "localhost" despite being deployed in different places. each task creates artifacts and metrics that are used downstream

  
  
Posted one year ago

I think i've narrowed this down to the ssh connection approach.

regarding the container that runs the pipeline:

  • when I made it stop using autossh tunnels and instead put it on the same machine as the clearml server + used docker network host mode, suddenly the problematic pipeline started completing.
    it's just so odd that the pipeline controller task is the only one with an issue. the modeling / data-creation tasks really all seem to complete consistently just fine.

so yeah, best guess now is that its unrelated to clearml verison but rather to the connectivity of the pipeline controller task to the api server.

when I run this pipeline controller locally (also using the same ssh tunnel approach for comms), the pipeline completes just fine. so it's something specific about how its working inside the container vs on my machine, it seems.

  
  
Posted one year ago

but maybe here's a clue. after hanging like that for a while... it seems like the agent restarts (the container it runs in does not)
image

  
  
Posted one year ago

None here's how I'm establishing worker-server (and client-server) comms fwiw

  
  
Posted one year ago

Hi @<1689446563463565312:profile|SmallTurkey79> , when this happens, do you see anything in the API server logs? How is the agent running, on top of K8s or bare metal? Docker mode or venv?

  
  
Posted one year ago

worker thinks its in venv mode but is containerized .
apiserver is docker compose stack

ill check logs next time i see it .

currently rushing to ship a model out, so I've just been running smaller experiments slowly hoping to avoid the situation . fingers crossed .

  
  
Posted one year ago

image

  
  
Posted one year ago

that's the final screenshot. it just shows a bunch of normal "launching ..." steps, and then stops all the sudden.

  
  
Posted one year ago

it happens consistently with this one task that really should be all cache.
I disabled cache in the final step and it seems to run now.

  
  
Posted one year ago

clearml_agent v1.8.1
  
  
Posted one year ago

would it be on the pipeline task itself then, since that's what's disappearing? that likely the case

  
  
Posted one year ago

it's pretty reliably happening but the logs are just not informative. just stops midway
image

  
  
Posted one year ago

Hi @<1689446563463565312:profile|SmallTurkey79> !
Prior runs of this pipeline worked just fine What SDK version were you using for the prior runs? Does this still happen if you revert to that version?
Can you provide a script that imitates what you are doing?
In the pipeline you are running, are you creating new tasks/pipelines/datasets?

  
  
Posted one year ago

I really can't provide a script that matches exactly (though I do plan to publish something like this soon enough), but here's one that's quite close / similar in style:
None where I tried function-steps out instead, but it's a similar architecture for the pipeline (the point of the example was to show how to do a dynamic pipeline)

  
  
Posted one year ago

do you have the agent logs that is supposed to run your pipeline? Maybe there is a clue there. I would also suggest to try enqueuing the pipeline to some other queue, maybe even run the agent on your on machine if you do not already and see what happens

  
  
Posted one year ago

nothing came up in the logs. all 200's

  
  
Posted one year ago

ugh. again. it launched all these tasks and then just died. logs go silent.
image
image
image

  
  
Posted one year ago

can you share the logs of the controller?

  
  
Posted one year ago

would it be on the pipeline task itself then, since that's what's disappearing?
I will do some experiment comparisons and see if there are package diffs. thanks for the tip.

  
  
Posted one year ago

enqueuing. pipe.start("default") but I think it's picking up on my local clearml install instead of what I told it to use.

my tasks have this in them... what's the equivalent for pipeline controllers?
image

  
  
Posted one year ago

ah. a clue! it came right below that but i guess out of order...
that id is the pipeline that failed
image

  
  
Posted one year ago

its odd... I really dont see tasks except the controller one dying

  
  
Posted one year ago

default queue is served with (containerized + custom entrypoint) venv workers (agent services just wasn't working great for me, gave up)

  
  
Posted one year ago

yeah this problem seems to happen on 1.15.1 and 1.16.2 as well, prior runs were on the same version even. It just feels like it happens absolutely randomly (but often).
just happened again to me.

The pipeline is constructed from tasks, it basically does map/reduce. prepare data -> model training + evaluation -> backtesting performance summary.

It figures out how wide to go by parsing the date range supplied as input parameter. Been running stuff like this for months but only recently did things just start... vanishing like this.

Would appreciate any help. Really need this to be more robust to make the case for company-wide adoption.
image

  
  
Posted one year ago

let me downgrade my install of clearml and try again.

  
  
Posted one year ago

I will ask internally about this

  
  
Posted one year ago

I have tried other queues, they're all running the same container.
so far the only thing reliable is pipe.start_locally()

  
  
Posted one year ago

thank you

  
  
Posted one year ago
102K Views
43 Answers
one year ago
one year ago
Tags