Hi Juan,

On Sun, 2013-01-20 at 22:15 +0000, Juan Barry Manuel Canham wrote:
> Hi and thanks for the great fs.
> 
> I managed to corrupt a volume (my /var)
> 
> to do this I was running an upgrade and when I ran out of space I ran
>       while true; do sudo nilfs-clean -v -S 10/1 -p 1  /dev/sda4; df -h /var/ 
> | tail -n 1 ; sleep 1 ; done
> 
> I'm running 
>       Linux JCanhamLaptop 3.3.0-rc4 #1 SMP Fri Feb 24 19:12:32 GMT 2012 
> x86_64 x86_64 x86_64 GNU/Linux
> 
> I know it's an RC but I never got round to upgrading, hope this doesn't 
> invalidate my bug
> 
> I've attached what I think are the relevant dmesg and syslog but they are a 
> bit long,
> 
> I can make the system be remounted read-only by running apt-get autoclean
> 
> Jan 20 22:08:15 localhost kernel: [129306.872119] NILFS: bad btree node 
> (blocknr=185027): level = 0, flags = 0x0, nchildren = 768
> Jan 20 22:08:15 localhost kernel: [129306.872127] NILFS error (device sda4): 
> nilfs_bmap_last_key: broken bmap (inode number=11540)
> Jan 20 22:08:15 localhost kernel: [129306.872129] 
> Jan 20 22:08:15 localhost kernel: [129306.872131] Remounting filesystem 
> read-only
> Jan 20 22:08:15 localhost kernel: [129306.872135] NILFS warning (device 
> sda4): nilfs_truncate_bmap: failed to truncate bmap (ino=11540, err=-5)
> 
> Remounting rw works until I run autoclean
> Jan 20 22:08:44 localhost kernel: [129335.576471] segctord starting. 
> Construction interval = 5 seconds, CP frequency < 30 seconds
> Jan 20 22:08:44 localhost kernel: [129335.576483] NILFS warning: mounting fs 
> with errors
> Jan 20 22:08:44 localhost nilfs_cleanerd[16781]: start
> Jan 20 22:08:44 localhost nilfs_cleanerd[16781]: pause (clean check)
> Jan 20 22:08:45 localhost nilfs_cleanerd[16786]: start
> Jan 20 22:08:45 localhost nilfs_cleanerd[16786]: pause (clean check)
> 

Thank you for your report.

This issue is under investigation and fix. Currently, I think that the
reason of the reported by you issue is low value of vm.min_free_kbytes
and the issue with flush kernel thread. The issue with flush kernel
thread is near of fix. So, it needs to wait sometime. I hope that the
patch will be prepared soon. I need a time for testing and check the
fix.

With the best regards,
Vyacheslav Dubeyko.

> Thanks for any help and I hope I can help track down this bug
> 
> Juan


--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to