Channels
dagster-de
topic-async-execution
dagster-jp
dagster-cube
gigs-freelance
dagster-wandb
dagster-feedback
dagster-machine-learning
dagster-bigquery
dagster-support
dagster-dask
dagster-snowflake
dagster-releases
dagster-dbt
faq-read-me-before-posting
tools
dagster-pipes
project-flexible-range-backfills
data-platform-design
events
dagster-serverless
dagster-noteable
introductions
dagster-airflow
dagster-cloud
random
dagster-airbyte
dagster-spatial
announcements
github-discussions
豆瓣酱帮
dagstereo
dagster-ecs
dagster-kubernetes
jobs
dagster-showcase
data-quality-asset-checks
Powered by
#announcements
Title
n
Noah K
12/23/2020, 2:40 PM
This was in the middle of a flurry of errors so I'm guessing it's some weird timing issue?
d
daniel
12/23/2020, 3:09 PM
Is it possible to post the other errors? This was the launcher that starts up runs in their own threads but all in the same process?
n
Noah K
12/23/2020, 3:10 PM
The other errors were legit pipeline failures, solid raised an exception.
It seems like about 50% failed with that error and 50% with the solid exception
This was with about 700 pipeline runs launched simultaneously
And yeah, each run in a thread, inside a celery worker process
Not something I've seen before hence the guess that's related to hundreds of runs failing almost immediately 😄
Might have been a connection overload on the gRPC daemon?
It doesn't have any logging output so hard to tell if it was sad from the rush of connections.
Post