Eu resolvi o problema com a ajuda de JoeJulian da Freenode # Gluster. Quando examinei a lista de processos, havia alguns processos usando a configuração antiga: /usr/sbin/glusterfsd -s localhost --volfile-id storage.11.111.111.11.storage-2b...
. Depois de executar este comando:
killall glusterfsd ; killall -9 glusterfsd ; killall glusterd ; glusterd
A situação está resolvida - todos os blocos estão online:
# gluster volume status
Status of volume: storage
Gluster process Port Online Pid
------------------------------------------------------------------------------
Brick gluster-1a:/storage/1a 24009 Y 17302
Brick gluster-1b:/storage/1b 24009 Y 12188
Brick gluster-2a:/storage/2a 24009 Y 10863
Brick gluster-2b:/storage/2b 24009 Y 13486
Brick gluster-1a:/storage/3a 24011 Y 17308
Brick gluster-1b:/storage/3b 24011 Y 12194
Brick gluster-2a:/storage/4a 24011 Y 10869
Brick gluster-2b:/storage/4b 24011 Y 13492
NFS Server on localhost 38467 Y 17314
Self-heal Daemon on localhost N/A Y 17320
NFS Server on gluster-2a 38467 Y 10879
Self-heal Daemon on gluster-2a N/A Y 10885
NFS Server on gluster-2b 38467 Y 13503
Self-heal Daemon on gluster-2b N/A Y 13509
NFS Server on gluster-1b 38467 Y 12200
Self-heal Daemon on gluster-1b N/A Y 12206