-
Notifications
You must be signed in to change notification settings - Fork 615
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
bug: HDFS Storage flush error #9147
Comments
I believe this is some hdfs config issue, will check it later. |
Hi, can this issue be reproduced consistently or is it random? Just a quick note, OpenDAL has built-in support for retrying operations in case of temporary errors. You can find more information about it at this link: https://docs.rs/opendal/latest/opendal/layers/struct.RetryLayer.html. |
May due to the hdfs classpath is incorrect.
|
I think in the test case I mentioned, the database has already flushed part of the data into the system, which means the program should have already found the correct HDFS classpath. |
I think the problem has been resolved with apache/opendal#2114. Kindly upgrade to opendal |
Thanks, let me try it later. |
After a quick verification, I think hdfs works now. I will recheck this later, thanks for your work~ @Xuanwo |
Describe the bug
When trying to use

HDFS
as the storage object and flush content, hummock will meet some problems. And also, when selecting HDFS as the storage, some e2e tests will fail.In my environment, when launched by
./risedev d
, flush is ok, e2e test is also ok. When opening HDFS feature, build,./risedev d hdfs
, flush and e2e test will meet the problem above.When I check the HDFS, there are some data already storage successfully.

To Reproduce
wcy/push_for_yufan
which already enables the hdfs feature../risedev d hdfs
Expected behavior
No response
Additional context
No response
The text was updated successfully, but these errors were encountered: