UFS panics

[Sean Bruno]

On 10/24/18 9:22 AM, Leandro wrote:
> Hello,
> 
> I'm seeing a kernel panic when trying to move a specific file.
> 
> panic: Bad effnlink fip 0xc0000004a2c69a00, fdp 0xc000000497093be0,
> tdp 0xc0000004be6295a0
> cpuid = 72
> time = 1540283798
> KDB: stack backtrace:
> 0xe0000000adb8dcc0: at .kdb_backtrace+0x5c
> 0xe0000000adb8ddf0: at .vpanic+0x1b4
> 0xe0000000adb8deb0: at .panic+0x38
> 0xe0000000adb8df40: at .ufs_readdir+0x2f24
> 0xe0000000adb8e1b0: at .VOP_RENAME_APV+0x190
> 0xe0000000adb8e240: at .kern_renameat+0x3c0
> 0xe0000000adb8e540: at .sys_rename+0x2c
> 0xe0000000adb8e5c0: at .trap+0x65c
> 0xe0000000adb8e780: at .powerpc_interrupt+0x290
> 0xe0000000adb8e820: user SC trap by 0x81010b7b8: srr1=0x900000000000f032
>             r1=0x3fffffffffffc490 cr=0x34000028 xer=0x20000000
> ctr=0x81010b7b0 r2=0x8102c5950
> KDB: enter: panic
> 
> Using 'ls' or 'rm' on the file gives a "Bad file descriptor" error.
> Using 'cat', I get another panic, but now during open.
> Everything indicates that the file system is in an inconsistent state.
> 
> Therefore, I would just like to ask the following: is it expected for
> kernel panics to happen when there are errors in the file system?
> 
> Thanks,
> Leandro
> _________________________

I can't make much sense of the ppc backtrace here, but I assume this is UFS?

If it is, UFS tends to panic to protect data in an error case and ZFS
tends to go read-only in an error case.  (over generalization here, but
its been my experience).

sean

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 618 bytes
Desc: OpenPGP digital signature
URL: <http://lists.freebsd.org/pipermail/freebsd-current/attachments/20181024/c5a9261d/attachment.sig>