Bug 175812
| Summary: | setxattr() to a file on NFS returns EIO | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 4 | Reporter: | Keiichi Mori <kmori> | ||||||
| Component: | kernel | Assignee: | Steve Dickson <steved> | ||||||
| Status: | CLOSED ERRATA | QA Contact: | Brian Brock <bbrock> | ||||||
| Severity: | medium | Docs Contact: | |||||||
| Priority: | medium | ||||||||
| Version: | 4.3 | CC: | jbaron | ||||||
| Target Milestone: | --- | Keywords: | Regression | ||||||
| Target Release: | --- | ||||||||
| Hardware: | All | ||||||||
| OS: | Linux | ||||||||
| Whiteboard: | |||||||||
| Fixed In Version: | RHSA-2006-0132 | Doc Type: | Bug Fix | ||||||
| Doc Text: | Story Points: | --- | |||||||
| Clone Of: | Environment: | ||||||||
| Last Closed: | 2006-03-07 21:05:41 UTC | Type: | --- | ||||||
| Regression: | --- | Mount Type: | --- | ||||||
| Documentation: | --- | CRM: | |||||||
| Verified Versions: | Category: | --- | |||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||
| Embargoed: | |||||||||
| Bug Depends On: | |||||||||
| Bug Blocks: | 168429 | ||||||||
| Attachments: |
|
||||||||
Created attachment 122274 [details]
strace log on NFS client
I'm having a hard time reproducing this problem.... I notice your doing the 'mv' as root, now is the filesystem exported with no_root_squash? Also is the local filesystem (I'm assuming its an ext3 fs) mounted with the acl mount flag? I tried this again, I have just seen this issue in the case that NFS server/client are running both 2.6.9-25.EL. So, following table should be correct: NFS client NFS Server Error of setxattr() 2.6.9-25.EL 2.6.9-25.EL EIO 2.6.9-25.EL 2.6.9-22.0.1.EL EOPNOTSUPP 2.6.9-22.0.1.EL 2.6.9-25.EL EOPNOTSUPP 2.6.9-22.0.1.EL 2.6.9-22.0.1.EL EOPNOTSUPP I'm sorry that I've been confused. > now is the filesystem exported with no_root_squash? Yes, in the attached strace case. But it could happen without no_root_squach. > Also is the local filesystem (I'm assuming > its an ext3 fs) mounted with the acl mount flag? The localfile system is ext3 but it does not have acl flag. On NFS server, SELinux is disabled: # cat /proc/mounts | grep exports /dev/sde1 /exports ext3 rw 0 0 # tune2fs -l /dev/sde1|grep feature Filesystem features: has_journal resize_inode filetype needs_recovery sparse_super large_file # cat /etc/exports /exports *(rw) # ls -ld /exports drwxrwxrwx 3 root root 4096 Dec 20 02:02 /exports # uname -r 2.6.9-25.ELsmp NFS Client, SELinux is disabled: $ cat /proc/mounts | grep exports sumida:/exports /mnt nfs rw,v3,rsize=32768,wsize=32768,hard,tcp,lock,addr=sumida 0 0 $ uname -r 2.6.9-25.ELsmp $ ls -l /tmp/log -rw-rw-r-- 1 kmori kmori 1659 12æ 20 02:06 /tmp/log $ mv /tmp/log /mnt mv: setting permissions for `/mnt/log': Input/output error $ ls -l /mnt/log -rw-rw-r-- 1 kmori kmori 1659 Dec 20 02:06 /mnt/log Created attachment 122422 [details]
Proposed Patch
Ok... I think I found the problem... it appears
the server is not returning the correct error.
The Proposed patch seems to take care of this
problem. Note this patch is a combo of two upstream
pacthes...
I confirmed that above patch solved this issue. (appling the patch into kernel-2.6.9-25.EL). Thanks. An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on the solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2006-0132.html |
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; ja-JP; rv:1.7.12) Gecko/20050921 Red Hat/1.0.7-1.4.1 Firefox/1.0.7 Description of problem: When moving a file to a directory on NFS filesystem, mv command shows EIO error. # ls -l /tmp/log -rw-r--r-- 1 root root 5569 Dec 15 18:48 /tmp/log # mount | grep /mnt sumida:/exports on /mnt type nfs (rw,addr=172.16.32.115) # mv /tmp/log /mnt mv: setting permissions for `/mnt/hoge': Input/output error I confirmed that kernel-2.6.9-25.EL returns EIO instead of EOPNOTSUPP. NFS client NFS Server Error of setxattr() 2.6.9-25.EL 2.6.9-25.EL EIO 2.6.9-25.EL 2.6.9-22.0.1.EL EOPNOTSUPP 2.6.9-22.0.1.EL 2.6.9-25.EL EIO 2.6.9-22.0.1.EL 2.6.9-22.0.1.EL EOPNOTSUPP Strace log for kernel-2.6.9-25.EL NFS server: getxattr("/tmp/log", "system.posix_acl_access", 0xbff94130, 132) = -1 EOPNOTSUPP (Operation not supported) setxattr("/mnt/log", "system.posix_acl_access", "\x02\x00\x00\x00\x01\x00\x06\x00\xff\xff\xff\xff\x04\x00\x04\x00\xff\xff\xff\xff \x00\x04\x00\xff\xff\xff\xff", 28, 0) = -1 EIO (Input/output error) Strace log for kernel-2.6.9-22.0.1.EL NFS server: getxattr("hoge", "system.posix_acl_access", 0xbfec0df0, 132) = -1 EOPNOTSUPP (Operation not supported) setxattr("issues/hoge", "system.posix_acl_access", "\x02\x00\x00\x00\x01\x00\x06\x00\xff\xff\xff\xff\x04\x00\x06\x00\xff\xff\xff\xff \x00\x04\x00\xff\xff\xff\xff", 28, 0) = -1 EOPNOTSUPP (Operation not supported) e log for kernel-2.6.9-22.0.1.EL NFS server: Version-Release number of selected component (if applicable): kernel-2.6.9-25.EL How reproducible: Always Steps to Reproduce: 1. mount NFS exported by kernel-2.6.9-25.EL 2. move a local file to NFS Actual Results: I/O error happens: mv: setting permissions for `/mnt/hoge': Input/output error Expected Results: No I/O error happens Additional info: This is indeed regression issue.