mirror of
https://github.com/s3fs-fuse/s3fs-fuse.git
synced 2026-04-25 13:26:00 +03:00
[GH-ISSUE #1822] "multipart_size" parameter seems to be invalid #934
Labels
No labels
bug
bug
dataloss
duplicate
enhancement
feature request
help wanted
invalid
need info
performance
pull-request
question
question
testing
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/s3fs-fuse#934
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @CRblog on GitHub (Dec 24, 2021).
Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/1822
Version of s3fs being used (s3fs --version)
V1.90 (commit:5de92e9)
Version of fuse being used (pkg-config --modversion fuse, rpm -qi fuse, dpkg -s fuse)
2.9.2
s3fs command line used, if applicable
s3fs testbucket /data/testbucket -o url=http://xxxx -o passwd_file=/root/.passwd-s3fs -o dbglevel=info -o curldbg,use_path_request_style,allow_other -o retries=1 -o multipart_size=8 -o multireq_max=8 -o parallel_count=32
Details about issue
[root]# cp testdata testbucket/
cp: failed to close ‘testbucket/testdata’: Input/output error
When I try to copy a file of 200MB in size, I feel that the "multipart_size" parameter seems to be invalid. The s3 service I use requires me to set the part size to 8MB, and the request sent to the server contains 23 8MB parts and a 16MB part, but theoretically all parts should be 8MB (except the last part may be less than 8MB)
I have tested other file size and it seems to be the same. I am not sure which part is the problem, but there is always one part that does not satisfy multipart_size (guess it is the last part)
What I learned is that the "multipart_size" parameter is effective in this scenario of version 1.86
@skepticalwaves commented on GitHub (Jan 2, 2022):
Maybe related to #1816
@CRblog commented on GitHub (Jan 4, 2022):
Thanks, but it doesn't seem to be the same problem
@skepticalwaves commented on GitHub (Jan 4, 2022):
You may want to go through the whole debug logging like I did then, to give more info.
@ggtakec commented on GitHub (Jan 7, 2023):
@CRblog I'm sorry for my late reply.
You can specify
nomixuploadoption to solve this problem.I believe this makes each part size of the multipart the fixed length you expect.
This issue will be closed, but if you still have problems, please reopen or post another issue.
Thanks in advance for your assistance.
NOTE
s3fs by default uses MIX multipart upload (COPY API is included on upload).
This respects each specified part size by trying to keep part sizes as coherent as possible, but in some situations the part size is variable.
Your issue is a problem caused by this.