Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

PyDBEngine for OmniSci engine and pyarrow.gandiva for pyarrow storage format are incompatible #3865

Closed
dchigarev opened this issue Dec 16, 2021 · 3 comments
Labels
HDK Related to HDK (OmniSci successor) engine or backend P2 Minor bugs or low-priority feature requests

Comments

@dchigarev
Copy link
Collaborator

dchigarev commented Dec 16, 2021

PyDBEngine that is being imported when using Modin with STORAGE_FORMAT=Omnisci and pyarrow.gandiva - when using modin with STORAGE_FORMAT=Pyarrow are incompatible:
If ganvida is imported first everything is alright:

import pyarrow.gandiva
from modin.experimental.core.execution.native.implementations.omnisci_on_native.utils import (
    PyDbEngine,
)

But vice-versa it fails with segfault:

from modin.experimental.core.execution.native.implementations.omnisci_on_native.utils import (
    PyDbEngine,
)
import pyarrow.gandiva
# Output:
# : CommandLine Error: Option 'enable-vfe' registered more than once!
# LLVM ERROR: inconsistency in registered CommandLine options
# Aborted (core dumped)

Packages versions:

omniscidb-common          5.8.0                h1234567_1    conda-forge
omniscidbe                5.8.0            h1234567_1_cpu    conda-forge
pyomniscidbe              5.8.0           py38h1234567_1_cpu    conda-forge

arrow-cpp                 5.0.0           py38h43b8b71_5_cpu    conda-forge
pyarrow                   5.0.0           py38h1bc9799_5_cpu    conda-forge

Ideally, there shouldn't be scenarios when both of these modules are imported, however, as the recent #3783 showed, it still could happen sometimes.

The role of this issue is to track progress (if any) of resolving the incompatibility of these packages.

@dchigarev dchigarev added the HDK Related to HDK (OmniSci successor) engine or backend label Dec 16, 2021
@vnlitvinov
Copy link
Collaborator

I wonder if #3783 fixed that, could someone from @modin-project/modin-omnisci please re-check?

@vnlitvinov vnlitvinov added the P2 Minor bugs or low-priority feature requests label Aug 26, 2022
anmyachev added a commit to anmyachev/modin that referenced this issue Jul 2, 2023
…ng HDK with pyarrow.gandiva

Signed-off-by: Anatoly Myachev <anatoly.myachev@intel.com>
@anmyachev
Copy link
Collaborator

anmyachev commented Jul 2, 2023

Looks very outdated. The issue could not be reproduced as follows:

from modin.experimental.core.execution.native.implementations.hdk_on_native.db_worker import DbWorker
import pyarrow.gandiva

If I have not missed anything, then I suggest merging #6343 to close this issue.

UPD: I tested on windows so it worked, but on linux the error is still there

@YarShev
Copy link
Collaborator

YarShev commented May 16, 2024

HDK engine is deprecated and will be removed in a future version.

@YarShev YarShev closed this as completed May 16, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
HDK Related to HDK (OmniSci successor) engine or backend P2 Minor bugs or low-priority feature requests
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants