Re: [RFC PATCH 5/5] Btrfs: fix broken free space cache after the system crashed

2014-03-04 Thread Miao Xie
On Tue, 4 Mar 2014 10:19:20 -0500, Josef Bacik wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > On 01/15/2014 07:00 AM, Miao Xie wrote: >> When we mounted the filesystem after the crash, we got the >> following message: BTRFS error (device xxx): block group 4315938816 >> has wrong amou

Re: ENOSPC errors during raid1 rebalance

2014-03-04 Thread Chris Murphy
On Mar 4, 2014, at 5:27 PM, Mike Russo wrote: > I'm sure this is due to the ext4 conversion, but that means the utility is > making a btrfs filesystem that later can't be converted to another profile > for some reason. https://btrfs.wiki.kernel.org/index.php/Conversion_from_Ext3 "But still,

Re: ENOSPC errors during raid1 rebalance

2014-03-04 Thread Chris Murphy
On Mar 4, 2014, at 5:27 PM, Mike Russo wrote: > Chris Murphy colorremedies.com> writes: > >>> How can I find out what file is on the block group that >>> it's having a problem with? >> >> I think that's btrfs-debug-tree -b ? But don't hold me to that. >> I haven't done enough debugging to f

[PATCH v3] Btrfs: add readahead for send_write

2014-03-04 Thread Liu Bo
Btrfs send reads data from disk and then writes to a stream via pipe or a file via flush. Currently we're going to read each page a time, so every page results in a disk read, which is not friendly to disks, esp. HDD. Given that, the performance can be gained by adding readahead for those pages.

RE: ENOSPC errors during raid1 rebalance

2014-03-04 Thread Mike Russo
Chris Murphy colorremedies.com> writes: > > How can I find out what file is on the block group that > > it's having a problem with? > > I think that's btrfs-debug-tree -b ? But don't hold me to that. > I haven't done enough debugging to find files > from block numbers. Another that might be re

[PATCH] Btrfs-progs: remove unused variable and update btrfs-image man page

2014-03-04 Thread Rakesh Pandit
Remove unused variable in btrfs-image.c (update_super) and update man page documentation about -r option. Running btrfsck on a restored image produces missing chunk information. This is because by default, btrfs-image fixes up chunk tree to use 1 stripe pointing to the primary device. This in turns

Getting profile conversion to finish

2014-03-04 Thread E V
I'm having difficulty getting a profile conversion from raid0 -> single to finish. Is there a way to tell the filesystem to stop all new allocations for the raid0 profile, before the full balance finishes? Otherwise I feel like I'm in a sisyphian push of blocks into higher block numbers. Since the

Re: ENOSPC errors during raid1 rebalance

2014-03-04 Thread Chris Murphy
On Mar 4, 2014, at 11:54 AM, Michael Russo wrote: > Chris Murphy colorremedies.com> writes: >> Based on my reading of the man page, I think it's expected. >> You either need -s -l or -t. > > Ok, although the man page uses [ ] instead of < > and something > does happen if I don't add them. But

Re: ENOSPC errors during raid1 rebalance

2014-03-04 Thread Michael Russo
Chris Murphy colorremedies.com> writes: > Based on my reading of the man page, I think it's expected. >You either need -s -l or -t. Ok, although the man page uses [ ] instead of < > and something does happen if I don't add them. But if I use "-t 1" wouldn't that get everything? > > > Now I've

Re: ENOSPC errors during raid1 rebalance

2014-03-04 Thread Chris Murphy
On Mar 4, 2014, at 8:55 AM, Michael Russo wrote: > Hugo Mills carfax.org.uk> writes: > >> This is just a guess, but you might have some large (>1GB) > extents >> in there that span across multiple chunks. I'd suggest running a > btrfs >> defrag on any particularly big files and see if that

Re: btrfsck does not fix

2014-03-04 Thread Chris Murphy
On Mar 3, 2014, at 11:42 PM, Hendrik Friedel wrote: > Hi Chris, > > >> It might be worth finding large files to defragment. See the ENOSPC errors >> during raid1 rebalance thread. It sounds like it might be possible for some >> fragmented files to be stuck across multiple chunks, preventing

Re: ENOSPC errors during raid1 rebalance

2014-03-04 Thread Michael Russo
Hugo Mills carfax.org.uk> writes: >This is just a guess, but you might have some large (>1GB) extents > in there that span across multiple chunks. I'd suggest running a btrfs > defrag on any particularly big files and see if that helps the situation. > Doing this is definitely helping, b

Re: [RFC PATCH 5/5] Btrfs: fix broken free space cache after the system crashed

2014-03-04 Thread Josef Bacik
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 01/15/2014 07:00 AM, Miao Xie wrote: > When we mounted the filesystem after the crash, we got the > following message: BTRFS error (device xxx): block group 4315938816 > has wrong amount of free space BTRFS error (device xxx): failed to > load free

Re: Task Hang

2014-03-04 Thread Josef Bacik
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 03/04/2014 09:19 AM, Mark Murawski wrote: > I have btrfs as the fs for a backuppc box. updatedb was running at > the same time as a massive rsync. > > Mar 4 08:31:00 office-backup kernel: INFO: task > updatedb.mlocat:903 blocked for more than 120

Task Hang

2014-03-04 Thread Mark Murawski
I have btrfs as the fs for a backuppc box. updatedb was running at the same time as a massive rsync. Mar 4 08:31:00 office-backup kernel: INFO: task updatedb.mlocat:903 blocked for more than 120 seconds. Mar 4 08:31:00 office-backup kernel: Not tainted 3.13.2 #3 Mar 4 08:31:00 office-

Re: [PATCH v2] Btrfs: add readahead for send_write

2014-03-04 Thread David Sterba
On Tue, Mar 04, 2014 at 10:49:52AM +0800, Liu Bo wrote: > On Mon, Mar 03, 2014 at 03:25:28PM +0100, David Sterba wrote: > > This should not be a hard failure, it can continue without RA for this > > buffer. Besides, the RA buffer can be allocated at the beginning of send > > operation and pointer s