Repository navigation
test_os extended attribute setxattr tests can fail with ENOSPC (Errno 28) #66210
Description
Activity
Hibou57 commented
on Jul 19, 2014 Hibou57mannequinMannequinAuthormake testfails with Python 3.4.1 on Ubuntu 12.04 Intel 32 bits.The single
./configureoption used, was--prefix=$HOME/.local.As attached file, an archive containing both standard output and standard error outputs.
Not some important error notifications appears only in standard output, not in standard error: “OSError: [Errno 28] No space left on device”.
The latter is surprising as there was 11GiB free space.
(In the future, please just show the test failures and their tracebacks rather than uploading a tar file.)
The logs show three test failures in test_os:
======================================================================
ERROR: test_fds (test.test_os.ExtendedAttributeTests)
----------------------------------------------------------------------Traceback (most recent call last): File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2238, in test_fds self._check_xattrs(getxattr, setxattr, removexattr, listxattr) File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2213, in _check_xattrs self._check_xattrs_str(str, *args, **kwargs) File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2202, in _check_xattrs_str setxattr(fn, s("user.test"), b"a"*1024, **kwargs) File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2231, in setxattr os.setxattr(fp.fileno(), *args) OSError: [Errno 28] No space left on device: 4
======================================================================
ERROR: test_lpath (test.test_os.ExtendedAttributeTests)
----------------------------------------------------------------------Traceback (most recent call last): File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2223, in test_lpath os.listxattr, follow_symlinks=False) File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2213, in _check_xattrs self._check_xattrs_str(str, *args, **kwargs) File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2202, in _check_xattrs_str setxattr(fn, s("user.test"), b"a"*1024, **kwargs) OSError: [Errno 28] No space left on device: '@test_16527_tmp'
======================================================================
ERROR: test_simple (test.test_os.ExtendedAttributeTests)
----------------------------------------------------------------------Traceback (most recent call last): File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2219, in test_simple os.listxattr) File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2213, in _check_xattrs self._check_xattrs_str(str, *args, **kwargs) File "/home/yannick/c/src/Python-3.4.1/Lib/test/test_os.py", line 2202, in _check_xattrs_str setxattr(fn, s("user.test"), b"a"*1024, **kwargs) OSError: [Errno 28] No space left on device: '@test_16527_tmp'
Note that all three failures are on attempts to set a 1024-byte extended attributes on a /tmp file with os.setxattr(), a function which is just a wrapper around the Linux setxattr system call. "man 2 setxattr" documents that the call can fail with ENOSPC (Errno 28) if "there is insufficient space remaining to store the extended attribute". "man 5 attr", the overview of extended attribute support, notes that there are various kernel and file system specific limits on extended attributes, including length restrictions. So presumably there may be legitimate reasons why these tests could get ENOSPC on particular system configurations. It would be better if test_os anticipated such problems and passed those tests with a warning if ENOSPC is returned.
http://man7.org/linux/man-pages/man2/setxattr.2.html
http://man7.org/linux/man-pages/man5/attr.5.html- changed the title
[-]“make test” fails with 3.4.1 on Ubuntu 12.04[/-][+]test_os extended attribute setxattr tests can fail with ENOSPC (Errno 28)[/+]on Jul 20, 2014 If "insuffcient space" is insufficient disk space (ie: /tmp is full), I don't see any reason the test should cater to that possibility. None of our other tests do. If it means something else, then nevermind.
My reading of the man page is that in the xattr case it can refer to conditions other than just file system full, e.g. https://bugzilla.kernel.org/show_bug.cgi?id=12793. Also, Yannick reported that the volume was apparently not full.
We don't have enough info to decide if we should ignore the error or not. I just close this old issue as out of date.
#126909 continues this with more information.
Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.
Show more details
GitHub fields:
bugs.python.org fields: