Hide Forgot
if master volume of a geo-replication setup is down, defunct auxiliary mounts will be created in a series. How-to-repeat: kill the brick server(s) backing the master, then ls /tmp/gsyncd-aux-mount*
PATCH: http://patches.gluster.com/patch/7026 in master (syncdaemon: handle errors in glusterfs mount more carefully)
verified with 3.2.3 killed master bricks and temporary mount are not available.(running df or ls -l /tmp/gsyncd-aux-mount* continuously shows some tmp mount from time to time ..but they are removed soon) [root@Centos-kaushal2 ~]# ls -l /tmp/gsyncd-aux-mount* ls: /tmp/gsyncd-aux-mount-Ktmcu2: Transport endpoint is not connected [root@Centos-kaushal2 ~]# ls -l /tmp/gsyncd-aux-mount* ls: /tmp/gsyncd-aux-mount*: No such file or directory [root@Centos-kaushal2 ~]# df Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda2 17438736 4562932 11945644 28% / /dev/sda1 256666 16888 226526 7% /boot tmpfs 190260 0 190260 0% /dev/shm /root/swift-disk 995136 610244 384892 62% /mnt/sdb1 df: `/tmp/gsyncd-aux-mount-9_sRML': Transport endpoint is not connected [root@Centos-kaushal2 ~]# df Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda2 17438736 4562928 11945648 28% / /dev/sda1 256666 16888 226526 7% /boot tmpfs 190260 0 190260 0% /dev/shm /root/swift-disk 995136 610244 384892 62% /mnt/sdb1 ... after some time .... [root@Centos-kaushal2 ~]# df Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda2 17438736 4562940 11945636 28% / /dev/sda1 256666 16888 226526 7% /boot tmpfs 190260 0 190260 0% /dev/shm /root/swift-disk 995136 610244 384892 62% /mnt/sdb1 df: `/tmp/gsyncd-aux-mount-Qj4hwz': Transport endpoint is not connected [root@Centos-kaushal2 ~]# df Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda2 17438736 4562936 11945640 28% / /dev/sda1 256666 16888 226526 7% /boot tmpfs 190260 0 190260 0% /dev/shm /root/swift-disk 995136 610244 384892 62% /mnt/sdb1