Description of problem: During performance testing I see the following messages spamming the client logs: [2015-02-23 05:06:06.355467] W [dict.c:329:dict_set] (-->/usr/lib64/glusterfs/3.6.0.46/xlator/cluster/distribute.so(dht_selfheal_layout_lock_cbk+0x10) [0x7fc76447b110] (-->/usr/lib64/glusterfs/3.6.0.46/xlator/cluster/distribute.so(dht_refresh_layout+0x30e) [0x7fc76447b08e] (-->/usr/lib64/libglusterfs.so.0(dict_set_uint32+0x3b) [0x7fc76ef32ceb]))) 0-dict: !this || !value for key=trusted.glusterfs.dht [2015-02-23 05:06:06.355478] W [MSGID: 109003] [dht-selfheal.c:270:dht_refresh_layout] 0-testvol-dht: /file_dstdir/192.168.52.107/thrd_07/d_008/d_009: Failed to set dictionary value:key = trusted.glusterfs.dht [2015-02-23 05:06:06.356455] I [dict.c:371:dict_get] (-->/usr/lib64/glusterfs/3.6.0.46/xlator/cluster/distribute.so(dht_inodelk_done+0x4e) [0x7fc7644702ce] (-->/usr/lib64/glusterfs/3.6.0.46/xlator/cluster/distribute.so(dht_selfheal_layout_lock_cbk+0x10) [0x7fc76447b110] (-->/usr/lib64/glusterfs/3.6.0.46/xlator/cluster/distribute.so(dht_refresh_layout+0xb3) [0x7fc76447ae33]))) 0-dict: !this || key=trusted.glusterfs.dht Version-Release number of selected component (if applicable): glusterfs-3.6.0.46-1.el6.x86_64 How reproducible: Every run. Steps to Reproduce: 1. Run gluster benchmark kit on 2x1 volume. 2. Check logs Actual results: The mount logs are spammed with the above messages, also note that performance is poor. Expected results: Normal operation / performance. Additional info:
https://code.engineering.redhat.com/gerrit/43335
Verified on glusterfs-3.6.0.50-1.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-0682.html