starred/s3fs-fuse

Fork 0

mirror of https://github.com/s3fs-fuse/s3fs-fuse.git synced 2026-04-25 05:16:00 +03:00

[GH-ISSUE #10] truncate error #4

New issue

Closed

opened 2026-03-04 01:41:04 +03:00 by kerem · 7 comments

kerem commented

2026-03-04 01:41:04 +03:00

Owner

Originally created by @timurb on GitHub (Jan 26, 2014).
Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/10

I'm using S3FS as a storage directory for Bacula backups and quite often I see the following error in bacula logs:

2014-01-25 04:30:10   xxxx-sd JobId 267: Fatal error: label.c:463 Truncate error on device "FileStorage" (/mnt/backups): ERR=dev.c:1831 Unable to truncate device "FileStorage" (/mnt/backups). ERR=Operation not permitted

I'm running s3fs v1.74 on Debian Wheezy x64 AWS instance with the following commandline:

s3fs xxxxxxxxxxx /mnt/backups -o allow_other,retries=10,connect_timeout=30,readwrite_timeout=30,use_cache=/mnt/cache

I've recently was observing the same error with nearly the same frequency with s3fs v1.61 on i686 Debian Squeeze instance.

Originally created by @timurb on GitHub (Jan 26, 2014). Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/10 I'm using S3FS as a storage directory for Bacula backups and quite often I see the following error in bacula logs: ``` 2014-01-25 04:30:10 xxxx-sd JobId 267: Fatal error: label.c:463 Truncate error on device "FileStorage" (/mnt/backups): ERR=dev.c:1831 Unable to truncate device "FileStorage" (/mnt/backups). ERR=Operation not permitted ``` I'm running s3fs v1.74 on Debian Wheezy x64 AWS instance with the following commandline: ``` s3fs xxxxxxxxxxx /mnt/backups -o allow_other,retries=10,connect_timeout=30,readwrite_timeout=30,use_cache=/mnt/cache ``` I've recently was observing the same error with nearly the same frequency with s3fs v1.61 on i686 Debian Squeeze instance.

kerem closed this issue

2026-03-04 01:41:05 +03:00

kerem commented

2026-03-04 01:41:05 +03:00

Author

Owner

@ggtakec commented on GitHub (Jan 26, 2014):

Thanks for reporting this problem.
Can you reproduce the problem?
If you can, please run s3fs with "-d" option, it puts debugging messages into /var/log/messages(ex: s3fs prefix).
I think it helps us for solving this problem.

Thanks for advance for your help.

@ggtakec commented on GitHub (Jan 26, 2014): Thanks for reporting this problem. Can you reproduce the problem? If you can, please run s3fs with "-d" option, it puts debugging messages into /var/log/messages(ex: s3fs prefix). I think it helps us for solving this problem. Thanks for advance for your help.

kerem commented

2026-03-04 01:41:05 +03:00

Author

Owner

@timurb commented on GitHub (Jan 26, 2014):

I see that quite often (once or twice a day) but not always.

I'll try running s3fs with -d and post its output here when I see the next error.

@timurb commented on GitHub (Jan 26, 2014): I see that quite often (once or twice a day) but not always. I'll try running s3fs with `-d` and post its output here when I see the next error.

kerem commented

2026-03-04 01:41:06 +03:00

Author

Owner

@timurb commented on GitHub (Jan 26, 2014):

@ggtakec Do you think there could be sense in upgrading to s3fs 1.76 prior to that?

@timurb commented on GitHub (Jan 26, 2014): @ggtakec Do you think there could be sense in upgrading to s3fs 1.76 prior to that?

kerem commented

2026-03-04 01:41:06 +03:00

Author

Owner

@ggtakec commented on GitHub (Jan 26, 2014):

Changing v1.73 to v1.74, s3fs_truncate function in s3fs.cpp is changed r487(on goolgecodes).
It fixed issue371(https://code.google.com/p/s3fs/issues/detail?id=371)
But I think probably that changes is not the reason for this problem.

I hope s3fs puts something information in your messages log.

Thanks a lot.

@ggtakec commented on GitHub (Jan 26, 2014): Changing v1.73 to v1.74, s3fs_truncate function in s3fs.cpp is changed r487(on goolgecodes). It fixed issue371(https://code.google.com/p/s3fs/issues/detail?id=371) But I think probably that changes is not the reason for this problem. I hope s3fs puts something information in your messages log. Thanks a lot.

kerem commented

2026-03-04 01:41:06 +03:00

Author

Owner

@timurb commented on GitHub (Feb 10, 2014):

I've recently changed bacula configuration a bit and I no longer see errors like that since that time.

However I'm not quite sure if the issue has nothing to do with s3fs:

During the backup Bacula writes a big file to s3fs
I've had a post-action which deleted the written file from s3fs cache right after the job has finished
I've removed that post-action and the truncate errors have disappeared

Is it really safe to delete files from s3fs cache?

@timurb commented on GitHub (Feb 10, 2014): I've recently changed bacula configuration a bit and I no longer see errors like that since that time. However I'm not quite sure if the issue has nothing to do with s3fs: - During the backup Bacula writes a big file to s3fs - I've had a post-action which deleted the written file from s3fs cache right after the job has finished - I've removed that post-action and the truncate errors have disappeared Is it really safe to delete files from s3fs cache?

kerem commented

2026-03-04 01:41:06 +03:00

Author

Owner

@ggtakec commented on GitHub (Mar 17, 2014):

Sorry for responce Issue #16
I can understand about #16 log, I think this issue is as same as question/problem.
It seems that s3fs did not finish post-action when you removed it, by #16 log.
I want to know how to know about finishing post-action.

Thanks in advance for your help.

@ggtakec commented on GitHub (Mar 17, 2014): Sorry for responce Issue #16 I can understand about #16 log, I think this issue is as same as question/problem. It seems that s3fs did not finish post-action when you removed it, by #16 log. I want to know how to know about finishing post-action. Thanks in advance for your help.

kerem commented

2026-03-04 01:41:06 +03:00

Author

Owner

@timurb commented on GitHub (Mar 17, 2014):

Yes, that looks like that, therefore I'm closing this.
If you feel that s3fs needs some explicit error messages for that please reopen.

@timurb commented on GitHub (Mar 17, 2014): Yes, that looks like that, therefore I'm closing this. If you feel that s3fs needs some explicit error messages for that please reopen.

kerem referenced this issue

2026-03-04 01:47:21 +03:00

[GH-ISSUE #1191] s3fs crash during cache read #630

kerem referenced this issue

2026-03-04 01:53:18 +03:00

[PR #4] [MERGED] Fix compilation error on MacOSX with missing const #1308

kerem referenced this issue

2026-03-04 02:01:18 +03:00

[PR #697] [MERGED] Fixing race condition in FdEntity::GetStats #1610