mirror of
https://github.com/s3fs-fuse/s3fs-fuse.git
synced 2026-04-25 05:16:00 +03:00
[GH-ISSUE #2098] PUT /latest/api/token hangs without -o imdsv1only #1068
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#1068
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 @daverysyn on GitHub (Jan 25, 2023).
Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/2098
Additional Information
Version of s3fs being used (
s3fs --version)I built s3fs from source but the behavior in the yum 1.91-1 package was identical.
Version of fuse being used (
pkg-config --modversion fuse,rpm -qi fuseordpkg -s fuse)Kernel information (
uname -r)GNU/Linux Distribution, if applicable (
cat /etc/os-release)Curl info
How to run s3fs, if applicable
log messages (with my comments)
Details about issue
I can't seem to get the
PUT /latest/api/tokenrequests in s3fs to return without using (CTRL+C). I think this has something to do with the requests being "chunked", given that pressing CTRL+C yields this message:I've tried recreating the curl request that seems to be hanging. In the first snippet, I've kept the
Transfer-Encoding: chunkedheader, and I've removed it in the second snippet:Yes, both requests succeeded. However, the chunked request took 30 seconds! Does that sound like a timeout threshold to anyone else?
I suspect this is a duplicate of https://github.com/s3fs-fuse/s3fs-fuse/issues/2029, as adding the options
-o imdsv1only -o compat_dirseemed to allow s3fs to mount the bucket and actually see the objects inside of it (-o imdsv1onlywould mount the bucket, but the directory was empty and the logs showed ENOENT/404 errors until I added-o compat_dir). If I could stop s3fs from sending the chunked encoding header, perhaps-o imdsv1onlywould not be necessary. Thoughts?@ggtakec commented on GitHub (Jan 29, 2023):
@daverysyn Thanks for the details of the bug about deadlock.
I found the cause of this and fixed it.
I'm posting a PR #2103 for the fix, pelase wait until it is merged.
@ggtakec commented on GitHub (Feb 1, 2023):
#2103 has been merged.
I believe this bug has been fixed.
I will close this issue, but if you still have problems please reopen or post a new issue.