mirror of
https://github.com/s3fs-fuse/s3fs-fuse.git
synced 2026-04-25 05:16:00 +03:00
[GH-ISSUE #1765] Receiving 417 Expectation Failed when querying for IMDSv2 token #909
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#909
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 @kamilJ96 on GitHub (Sep 8, 2021).
Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/1765
Additional Information
When using IMDSv2, the first request that occurs is a PUT request to
/latest/api/tokenin order to retrieve a valid access token. By default, curl adds anExpect: 100-continueheader in the request, and AWS responds with a417 Expectation Failed.Now, curl will automatically retry this request without the
Expectheader, however it only does this for requests that have a body, which this request does not (see HERE), thus this request fails and s3fs defaults to using IMDSv1. If IMDSv1 is disabled on the instance, then the whole thing will fail.Version of s3fs being used (s3fs --version)
V1.90 (commit:9d1552a)
Version of fuse being used (pkg-config --modversion fuse, rpm -qi fuse, dpkg -s fuse)
2.9.2
(curl: 7.76.1)
Kernel information (uname -r)
Amazon Linux 2
s3fs syslog messages (grep s3fs /var/log/syslog, journalctl | grep s3fs, or s3fs outputs)
@nmeyerhans commented on GitHub (Sep 8, 2021):
For the record, this seems to only impact Xen-based EC2 instances; I have tested on t4g instances (Nitro-based arm64 instances) and the IMDS web server does respond with a token as expected, and does not return a 417 error. However, testing on t2 instances (Xen-based), does reproduce the problem as reported.
@kamilJ96 commented on GitHub (Sep 8, 2021):
@nmeyerhans Thanks AWS - I was wondering why this problem wasn't more prevalent but I guess it just needed a perfect storm of conditions before coming up 😄
@gaul do you have an estimate for the next release?
Thanks to both of you as well for help/approval! The s3fs code is really beautiful
@gaul commented on GitHub (Sep 8, 2021):
Probably February since s3fs aims for a release every six months and we just released 1.90.