mirror of
https://github.com/s3fs-fuse/s3fs-fuse.git
synced 2026-04-25 21:35:58 +03:00
[GH-ISSUE #1890] s3fs 1.90 terminates with segfault #960
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#960
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 @nkkashyap on GitHub (Feb 15, 2022).
Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/1890
Additional Information
The following information is very important in order to help us to help you. Omission of the following details may delay your support request or receive no attention at all.
Keep in mind that the commands we provide to retrieve information are oriented to GNU/Linux Distributions, so you could need to use others if you use s3fs on macOS or BSD
Version of s3fs being used (s3fs --version)
Version: V1.90 (commit:cd466eb)
Version of fuse being used (pkg-config --modversion fuse, rpm -qi fuse, dpkg -s fuse)
Fuse Version: 2.9.7-1ubuntu1
Kernel information (uname -r)
Kernel: 4.15.0-163-generic
GNU/Linux Distribution, if applicable (cat /etc/os-release)
Distro: Ubuntu 18.04.6 LTS
s3fs command line used, if applicable
/etc/fstab entry, if applicable
s3fs syslog messages (grep s3fs /var/log/syslog, journalctl | grep s3fs, or s3fs outputs)
if you execute s3fs with dbglevel, curldbg option, you can get detail debug messages
Details about issue
This issue is occurring in IBM Cloud Kubernetes Service.
With the help of K8S FlexVolume driver
ibmc-s3fs, the bucket is mounted on a worker node and then bind mounted to POD.S3FS is getting
segfaultwhen it is retrying post500 Server Error@gaul commented on GitHub (Feb 15, 2022):
Could you try running s3fs under
valgrind? This should reveal the cause of the segfault. Alternatively you could try using AddressSanitizer:@ggtakec commented on GitHub (Feb 23, 2022):
@nkkashyap
I want a little more information such as debug logs, too(as @gaul said).
And there are some things I would like to know below.
@ggtakec commented on GitHub (Mar 5, 2022):
@nkkashyap
#1917 has been merged.
If the cause of this issue is the same as #1917, you can solve it using the master branch.
If you can, please try to use master branch and we want to know the result.
Thanks in advance for your kindness.
@nkkashyap commented on GitHub (Mar 18, 2022):
Hi @ggtakec @gaul
Sorry for the delay.
This issue is occurring very frequently in the Customer's cluster. Unfortunately I am not able to recreate the issue in my test environment.
Finally, I managed to get the
AddressSanitizertrace with Customer's helpSomehow with
clangnot able to stack details. I usedgcc (g++)compiler.@ggtakec commented on GitHub (Mar 18, 2022):
@nkkashyap Thanks for your reply.
We have released a new version, which includes #1917.
Is it still happening with using new version?
In the first log you received, you know the following:
And your content of the trace shows the same thing as follows:
What I can see from this log is that your s3fs process received a 500 HTTP status from the S3 server(IBM?), and s3fs is trying to resend the request and is getting segfault.
I didn't know why your s3fs got the 500 status, but assuming from the log output, I think that it happened when s3fs set the token string to the buffer in resend request.
In other words, I imagined that a segfault had occurred for the address of that token buffer.
Then I checked the source and noticed that the buffer was not locked when the token was refreshing.
Then this bug was corrected at #1917.
So far, no other cause has been found yet.
Maybe your problem has been fixed in new v1.91(#1917), but are you still getting the error with this version?
@nkkashyap commented on GitHub (Mar 18, 2022):
@ggtakec Thanks
Updated s3fs to
v1.191on the cluster. Will monitor it for couple of days.@nkkashyap commented on GitHub (Mar 24, 2022):
With
s3fs v1.191thesegfaultissue is not occurring.I am closing this issue.
Thanks @ggtakec @gaul
@ggtakec commented on GitHub (Mar 26, 2022):
@nkkashyap Thank you for confirmation.
If you run into problems, please reopen or post new issue.