ENOSPC on mostly empty file system

2014-09-09 Thread Arnd Bergmann
Hi Chris, As I mentioned at the kernel summit, I have a file system that I use mostly for storing my one kernel git tree and occasionally some build trees (those are normally on a tmpfs), and I have again run into the problem where the file system is only partially full (I think 18% in this case)

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Holger Hoffstätte
On Tue, 09 Sep 2014 16:29:05 +0200, Arnd Bergmann wrote: As I mentioned at the kernel summit, I have a file system that I use mostly for storing my one kernel git tree and occasionally some build trees (those are normally on a tmpfs), and I have again run into the problem where the file

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Holger Hoffstätte
On Tue, 09 Sep 2014 15:06:19 +, Holger Hoffstätte wrote: You didn't give us any btrfs fs show or btrfs fs df outpu to look at, Of course fi, not fs.. -h -- To unsubscribe from this list: send the line unsubscribe linux-btrfs in the body of a message to majord...@vger.kernel.org More

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Arnd Bergmann
On Tuesday 09 September 2014 16:29:05 Arnd Bergmann wrote: I also played around with it some more. After removing a few small files, I could create new files with up to 20-60MB again before hitting ENOSPC. I then did a 'make clean' in all the object directories I had and after that could

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Clemens Eisserer
Hi Arnd, Ok, one more data point: Why don't you provide the data point you were specifically asked for, btrfs fi df ;) Regards, Clemens -- To unsubscribe from this list: send the line unsubscribe linux-btrfs in the body of a message to majord...@vger.kernel.org More majordomo info at

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Hugo Mills
On Tue, Sep 09, 2014 at 09:49:12PM +0200, Clemens Eisserer wrote: Hi Arnd, Ok, one more data point: Why don't you provide the data point you were specifically asked for, btrfs fi df ;) btrfs fi show is important as well -- it's hard to work out the state of the FS from just one of

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Arnd Bergmann
On Tuesday 09 September 2014 21:49:12 Clemens Eisserer wrote: Hi Arnd, Ok, one more data point: Why don't you provide the data point you were specifically asked for, btrfs fi df ;) I've cleaned it up again already. At the moment, it's working fine, with this data: Data: total=65.11GB,

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Hugo Mills
On Tue, Sep 09, 2014 at 11:49:10PM +0200, Arnd Bergmann wrote: Ok, now I'm in the bad state again (after running a 'make allmodconfig' kernel build: Label: none uuid: 1d88cccb-3d0e-42d9-8252-a226dc5c2e47 Total devices 1 FS bytes used 8.79GB devid1 size 67.14GB used

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Arnd Bergmann
On Tuesday 09 September 2014 22:57:25 Hugo Mills wrote: On Tue, Sep 09, 2014 at 11:49:10PM +0200, Arnd Bergmann wrote: Ok, now I'm in the bad state again (after running a 'make allmodconfig' kernel build: Label: none uuid: 1d88cccb-3d0e-42d9-8252-a226dc5c2e47 Total devices 1

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Roman Mamedov
On Tue, 9 Sep 2014 22:57:25 +0100 Hugo Mills h...@carfax.org.uk wrote: On Tue, Sep 09, 2014 at 11:49:10PM +0200, Arnd Bergmann wrote: Ok, now I'm in the bad state again (after running a 'make allmodconfig' kernel build: Label: none uuid: 1d88cccb-3d0e-42d9-8252-a226dc5c2e47

Re: ENOSPC on mostly empty file system

2014-09-09 Thread Holger Hoffstätte
On Tue, 09 Sep 2014 23:49:10 +0200, Arnd Bergmann wrote: Label: none uuid: 1d88cccb-3d0e-42d9-8252-a226dc5c2e47 Total devices 1 FS bytes used 8.79GB devid1 size 67.14GB used 67.14GB path /dev/sdc6 *ding ding ding* The term