[lustre-discuss] Odd problem with new OSTs not being used

Carlson, Timothy S Timothy.Carlson at pnnl.gov
Thu Sep 1 14:00:14 PDT 2016


Running Lustre 2.5.3(ish) backed with ZFS. 

We’ve added a few OSTs and they show as being “UP” but aren’t taking any data

[root at lzfs01a ~]# lctl dl
  0 UP osd-zfs MGS-osd MGS-osd_UUID 5
  1 UP mgs MGS MGS 1085
  2 UP mgc MGC172.17.210.11 at o2ib9 77cf08da-86a4-7824-1878-84b540993c6d 5
  3 UP osd-zfs lzfs-MDT0000-osd lzfs-MDT0000-osd_UUID 42
  4 UP mds MDS MDS_uuid 3
  5 UP lod lzfs-MDT0000-mdtlov lzfs-MDT0000-mdtlov_UUID 4
  6 UP mdt lzfs-MDT0000 lzfs-MDT0000_UUID 1087
  7 UP mdd lzfs-MDD0000 lzfs-MDD0000_UUID 4
  8 UP qmt lzfs-QMT0000 lzfs-QMT0000_UUID 4
  9 UP osp lzfs-OST0008-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 10 UP osp lzfs-OST0003-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 11 UP osp lzfs-OST0006-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 12 UP osp lzfs-OST0007-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 13 UP osp lzfs-OST0004-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 14 UP osp lzfs-OST000a-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 15 UP osp lzfs-OST0000-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 16 UP osp lzfs-OST0002-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 17 UP osp lzfs-OST0001-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 18 UP osp lzfs-OST0005-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 19 UP osp lzfs-OST0009-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 20 UP osp lzfs-OST000b-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 21 UP osp lzfs-OST000c-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 22 UP osp lzfs-OST000d-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 23 UP osp lzfs-OST0010-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 24 UP osp lzfs-OST000f-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 25 UP osp lzfs-OST000e-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 26 UP osp lzfs-OST0011-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 27 UP osp lzfs-OST0015-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 28 UP osp lzfs-OST0016-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 29 UP osp lzfs-OST0017-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 30 UP osp lzfs-OST0018-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 31 UP osp lzfs-OST0019-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 32 UP osp lzfs-OST001b-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 33 UP osp lzfs-OST0013-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 34 UP osp lzfs-OST0014-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 35 UP lwp lzfs-MDT0000-lwp-MDT0000 lzfs-MDT0000-lwp-MDT0000_UUID 5
 36 UP osp lzfs-OST001c-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 37 UP osp lzfs-OST0012-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 38 UP osp lzfs-OST001a-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 39 UP osp lzfs-OST001d-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 40 UP osp lzfs-OST001e-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 41 UP osp lzfs-OST001f-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 42 UP osp lzfs-OST0020-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 43 UP osp lzfs-OST0021-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5
 44 UP osp lzfs-OST0022-osc-MDT0000 lzfs-MDT0000-mdtlov_UUID 5

Now if you look at devices 36 and higher, you’ll see that they don’t have much data even though they have been online for a few weeks and this is a fairly active file system.   The data that is in there is data that I have “forced” onto the OSTs for testing by setting the stripe to that specific OST.  

# lfs df
UUID                   1K-blocks        Used   Available Use% Mounted on
lzfs-MDT0000_UUID    60762585216   262897152 60499686016   0% /pic[MDT:0]
lzfs-OST0000_UUID    90996712832 82190600320  8805795072  90% /pic[OST:0]
lzfs-OST0001_UUID    90996823936 82773737088  8221323776  91% /pic[OST:1]
lzfs-OST0002_UUID    90996723840 82547420928  8448555520  91% /pic[OST:2]
lzfs-OST0003_UUID    90996780416 82570822400  8425071872  91% /pic[OST:3]
lzfs-OST0004_UUID    90996792320 83526260096  7466092288  92% /pic[OST:4]
lzfs-OST0005_UUID    90996764544 83071284864  7922972800  91% /pic[OST:5]
lzfs-OST0006_UUID    90996729600 83348930304  7643451520  92% /pic[OST:6]
lzfs-OST0007_UUID    90996800000 82677238272  8314902016  91% /pic[OST:7]
lzfs-OST0008_UUID    90996910208 83598099584  7396038656  92% /pic[OST:8]
lzfs-OST0009_UUID    90997091328 85659415424  5335623552  94% /pic[OST:9]
lzfs-OST000a_UUID    90996807680 83581871872  7410268800  92% /pic[OST:10]
lzfs-OST000b_UUID    90996676352 77512128000 13484523136  85% /pic[OST:11]
lzfs-OST000c_UUID    90996505984 86176576256  4819325824  95% /pic[OST:12]
lzfs-OST000d_UUID    90997104256 90339916032   656510208  99% /pic[OST:13]
lzfs-OST000e_UUID    90996660480 86856594560  4134641792  95% /pic[OST:14]
lzfs-OST000f_UUID    90996441472 82859149568  8134773888  91% /pic[OST:15]
lzfs-OST0010_UUID    90996592896 88961102592  2034770816  98% /pic[OST:16]
lzfs-OST0011_UUID    90996459264 83005755520  7989576448  91% /pic[OST:17]
lzfs-OST0012_UUID    90999916800     1073280 90998828928   0% /pic[OST:18]
lzfs-OST0013_UUID    90996418560 83272862336  7716835328  92% /pic[OST:19]
lzfs-OST0014_UUID    90996442496 84503368320  6486773504  93% /pic[OST:20]
lzfs-OST0015_UUID    90996476416 82157845376  8831992320  90% /pic[OST:21]
lzfs-OST0016_UUID    90996456960 83149106688  7844745088  91% /pic[OST:22]
lzfs-OST0017_UUID    90996518912 84959371648  6032033408  93% /pic[OST:23]
lzfs-OST0018_UUID    90996448000 84187752448  6806393088  93% /pic[OST:24]
lzfs-OST0019_UUID    90996425344 85975606784  5012946816  94% /pic[OST:25]
lzfs-OST001a_UUID    90999916032        1920 90999912064   0% /pic[OST:26]
lzfs-OST001b_UUID    90996397312 82557086592  8435064576  91% /pic[OST:27]
lzfs-OST001c_UUID    90999915008    44549120 90955355520   0% /pic[OST:28]
lzfs-OST001d_UUID    90999917184        1920 90999913088   0% /pic[OST:29]
lzfs-OST001e_UUID    90999917312        2048 90999913216   0% /pic[OST:30]
lzfs-OST001f_UUID    90999917056        1920 90999903104   0% /pic[OST:31]
lzfs-OST0020_UUID    90999917312        2048 90999913216   0% /pic[OST:32]
lzfs-OST0021_UUID    90999917056        1920 90999903104   0% /pic[OST:33]
lzfs-OST0022_UUID    90999917312        2048 90999913216   0% /pic[OST:34]

filesystem summary:  3184912212480 2182065540096 1002764557568  69% /pic


Any ideas of something I need to trigger on the MDS to get files to being landing on the empty OSTs? Was thinking about rebooting the MDS, but wanted to use a smaller hammer to get this working.

Thanks!

Tim


More information about the lustre-discuss mailing list