From: sten@debian.org 
  
 Hi Leszek, 
  
 Sorry it took me so long to follow-up on this bug. 
  
 Leszek Dubiel  writes: 
  
 >> I made further tests. 
 >> 
 >> This system runs since 2018-02, so maybe it is spoiled from some bugs in 
 >> debian stretch (ver 9). 
  
 For i386 (i686), maybe!  If I remember correctly there are still 32bit 
 bugs.  Amd64/intel64 has been good since linux-4.4, and I have two 
 heavily used systems that have been fine since then.  But before for 
 that?  Ouf... 
  
 Leszek Dubiel  writes: 
  
 > Another test with "--repair" make BUG_ON: 
 > 
 > time btrfs check --mode=lowmem --progress --repair /dev/sda1 
 > enabling repair mode 
 > WARNING: low-memory mode repair support is only partial 
 > Opening filesystem to check... 
 > Checking filesystem on /dev/sda1 
 > UUID: 666a7089-d716-44ff-8081-56b969b58eff 
 > [1/7] checking root items€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€€ 
 (0:04:08 
 elapsed, 
 > 10279825 items checked) 
 > Fixed 0 roots. 
 > ERROR: extent[1198869807104 16384] backref lost (owner: 2, level: 0) 
 > root 2s checked) 
 > ERROR: fail to allocate new chunk No space left on device 
  
 Do you remember if you ever balanced metadata? 
  
 > Try to exclude all metadata blcoks and extents, it may be slowd, 81140 
 > items checked) 
 > Added an extent item [1198869807104 16384]€€€€€€€€ (2:16:04 elapsed, 81140 
 > items checked) 
 > Added one tree block ref start 1198869807104 root 26:05 elapsed, 81140 
 > items checked) 
 > ERROR: extent[81379328 16384] backref lost (owner: 2, level: 0) root 2 
 > items checked) 
 > Added an extent item [81379328 16384] 
 > transaction.c:168: btrfs_commit_transaction: BUG_ON `ret` triggered, 
 > value -17 
 > btrfs(+0x3c0ad)[0x5320ad] 
 > btrfs(btrfs_commit_transaction+0x68)[0x5324bb] 
 > btrfs(+0x78c24)[0x56ec24] 
 > btrfs(check_chunks_and_extents_lowmem+0x198)[0x572060] 
 > btrfs(cmd_check+0x16e2)[0x554da5] 
 > btrfs(main+0x22c)[0x502edc] 
 > /lib/i386-linux-gnu/libc.so.6(__libc_start_main+0xf1)[0xb7badb41] 
 > btrfs(_start+0x31)[0x502f21] 
 > Przerwane 
 > 
 > real€€€€€€ 140m42,553s 
 > user€€€€€€ 6m14,394s 
 > sys€€€€€€€€ 2m21,156s 
  
 I guess the upside is it failed faster this time?  (2h20min vs 3days) 
  
 The lowmem mode was practically brand new back then, and I wonder how 
 useful it is to keep this bug open.  Do you remember if you contacted 
 upstream linux-btrfs?  Often they'll expand he scope of btrfs-check to 
 accommodate new cases. 
  
 Cheers! 
 Nicholas 
  
 --=-=-Content-Type: application/pgp-signature; name="signature.asc" 
  
 -----BEGIN PGP SIGNATURE----- 
  
 iQJEBAEBCgAuFiEE4qYmHjkArtfNxmcIWogwR199EGEFAmjrBzoQHHN0ZW5AZGVi 
 aWFuLm9yZwAKCRBaiDBHX30QYY+wD/9oUnULW0OjoSM+91p9J3ALVWMfVruKzWY2 
 wpEC0t1Bv2FwA8dT4W0c+k2hDEAc8qWImnD56Qt5VPFsbIvoGyfEO5QosieghlWP 
 dJsrWlkP5Jvr26KcCgnHcZre0C8RV9z/sNiPSBeC3rwlD/N4ERPsIMwffdZTQE2u 
 9DrbCkBQGyWcWZBFdIT7fnXr4fbpzXGznwmNV2Iyf739oZWQerU+X1xsZgVKeZC/ 
 jRlPE2Ek/0oJ9PpzRYUVlXD3bk8xHYs5z48KjXHmf0Mq0JI0YwGRM0qkZozTmoy9 
 1molv7cmzpuQtnj9Qs1Gnak9VvJ4okR+cm2Lru34YBi1L19LmIJ8FR3DajX1eken 
 0fspJxoymG7dL6Q8SCVtjy8AoW2J1Dy+ozBJKJTwzCIDtri+2hfNnKruLqZW/rBf 
 tthaD9gt3scoct5IPw4+rozmtcmLl2vPtFLYTndyxWxzsWTxMUf1MuIftJG2thpR 
 Rq4NwZ7dpH/oIj0L+g+jCZDnPnzngwJ3TOUt6X7VQ5+FyAkqpSBQ7LoSRlKRMSEe 
 qwPgUElYnTloVRCt2X6mz+nXiIOtBbDxzK54bT59XXE/30AE+H5y6/0SIy0cIHjK 
 tCUcOFKQCpJLxEiBw5bwnhgVK72N7AGDkExGPqlyy3AsG9EGlwIN4m/UjGA3ruMn 
 N4KnBKIRiA==DmYk 
 -----END PGP SIGNATURE----- 
  
 --- SoupGate-Win32 v1.05 
  * Origin: you cannot sedate... all the things you hate (1:229/2)