Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Strange test failures

After finally resolving several build breaks as well as the breakage due to ksem module being unload, there is still something very broken in NetBSD-current. I'm suspecting some breakage in the file system code, but I really don't have a clue on how to further isolate the problem(s).

First of all, a large number of tests that were previously working are now failing:

Failed test cases:
    fs/nfs/t_rquotad:get_nfs_be_1_user, fs/nfs/t_rquotad:get_nfs_le_1_both,
    fs/nfs/t_rquotad:get_nfs_le_1_user, fs/vfs/t_renamerace:lfs_renamerace,
    lib/libc/stdlib/t_strtod:strtod_inf, lib/libevent/t_event:kqueue,
    lib/libevent/t_event:poll, lib/libm/t_infinity:infinity_long_double,
    lib/librumpclient/t_fd:sigio, lib/librumphijack/t_sh:redirect,
    lib/librumphijack/t_sh:runscript, lib/librumphijack/t_tcpip:http,
    lib/librumphijack/t_tcpip:nfs, lib/librumphijack/t_tcpip:ssh,
    lib/librumphijack/t_vfs:cpcopy, lib/librumphijack/t_vfs:mv_x,
    lib/librumphijack/t_vfs:paxcopy, lib/semaphore/pthread/t_sem_pth:named,

The nfs/t_rquotad test failures are semi-expected time-outs, although it is unusual for all six test cases to fail in a single pass. Similarly, the libevent/t_event failures are fairly common failures.

But the lib/librumpclient failures have previously passed successfully, so this is quite unexpected.

In addition, when running the atf tests via "anita interact" with the

        atf-run | tee /tmp/X | atf-report -o ticker:- -o xml:/tmp/Y

(which is essentially the same command that "anita test" uses), we get some strange interweaving of output. Note in the below example that output from a previous execution of t_grow_swapped (test #348) is displayed it the middle of the output from the current execution of tests t_mkdir (test #106)!

fs/tmpfs/t_mkdir (106/399): 5 test cases
    attrs: tee: /tmp/atf/test.tps: No space left on device
Expected failure: PR bin/44205: atf-check failed; see the outputPassed.
    kqueue:  of the test for details
    shrink_ffsv1_partial_cg: Passed.

sbin/resize_ffs/t_grow_swapped (348/400): 4 test cases
many: 65536: pid 24882 (rump_ffs): system write of 10920@0xffff800000984000 at 64 failed: 28

/: write failed, file system is full

/: write failed, file system is full

    nested: Passed.
    single: Passed.

fs/tmpfs/t_mknod (107/399): 6 test cases
    block: Passed.
    block_kqueue: Passed.

| Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:       |
| Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at    |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at |
| Kernel Developer |                          | pgoyette at  |

Home | Main Index | Thread Index | Old Index