Description of problem: ======================== With brick mux enabled,if i bring down the bricks by umounting the LV some of the volumes of still show brick as online and heal info shows the brick online instead of transport end point error Version-Release number of selected component (if applicable): ============================================================= glusterfs-server-3.12.2-29.el7rhgs.x86_64 How reproducible: ================= 1/1 Steps to Reproduce: 1.Enabled brick mux and created 9 volumes (3 volumes using 1 LV ) 2.now bring down bricks by using umount -l of the lv mount 3.The vol status still shows brick as online for 2 volumes and ofline for one volume Actual results: ============== [root@dhcp35-56 ~]# umount -l /gluster/brick3/ Status of volume: vol_-3-1 Gluster process TCP Port RDMA Port Online Pid ------------------------------------------------------------------------------ Brick 10.70.35.56:/gluster/brick3/vol_-3-1 N/A N/A N 2774 Brick 10.70.35.227:/gluster/brick3/vol_-3-1 49152 0 Y 3051 Brick 10.70.35.11:/gluster/brick3/vol_-3-1 49152 0 Y 11694 Self-heal Daemon on localhost N/A N/A Y 3750 Self-heal Daemon on dhcp35-227.lab.eng.blr. redhat.com N/A N/A Y 3827 Self-heal Daemon on dhcp35-11.lab.eng.blr.r edhat.com N/A N/A Y 12935 Task Status of Volume vol_-3-1 ------------------------------------------------------------------------------ There are no active volume tasks Status of volume: vol_-3-2 Gluster process TCP Port RDMA Port Online Pid ------------------------------------------------------------------------------ Brick 10.70.35.56:/gluster/brick3/vol_-3-2 N/A N/A N 2774 Brick 10.70.35.227:/gluster/brick3/vol_-3-2 49152 0 Y 3051 Brick 10.70.35.11:/gluster/brick3/vol_-3-2 49152 0 Y 11694 Self-heal Daemon on localhost N/A N/A Y 3750 Self-heal Daemon on dhcp35-227.lab.eng.blr. redhat.com N/A N/A Y 3827 Self-heal Daemon on dhcp35-11.lab.eng.blr.r edhat.com N/A N/A Y 12935 Task Status of Volume vol_-3-2 ------------------------------------------------------------------------------ There are no active volume tasks Status of volume: vol_-3-3 Gluster process TCP Port RDMA Port Online Pid ------------------------------------------------------------------------------ Brick 10.70.35.56:/gluster/brick3/vol_-3-3 N/A N/A N N/A Brick 10.70.35.227:/gluster/brick3/vol_-3-3 49152 0 Y 3051 Brick 10.70.35.11:/gluster/brick3/vol_-3-3 49152 0 Y 11694 Self-heal Daemon on localhost N/A N/A Y 3750 Self-heal Daemon on dhcp35-11.lab.eng.blr.r edhat.com N/A N/A Y 12935 Self-heal Daemon on dhcp35-227.lab.eng.blr. redhat.com N/A N/A Y 3827 Task Status of Volume vol_-3-3 ------------------------------------------------------------------------------ There are no active volume tasks Expected results: ================== The brick should be offline for all 3 volumes Additional info: ================= [root@dhcp35-56 ~]# df -h Filesystem Size Used Avail Use% Mounted on /dev/mapper/rhgs-root 26G 3.0G 24G 12% / devtmpfs 3.9G 0 3.9G 0% /dev tmpfs 3.9G 0 3.9G 0% /dev/shm tmpfs 3.9G 65M 3.8G 2% /run tmpfs 3.9G 0 3.9G 0% /sys/fs/cgroup /dev/mapper/GLUSTER_vg1-GLUSTER_lv1 50G 1.6G 49G 4% /gluster/brick1 /dev/mapper/GLUSTER_vg2-GLUSTER_lv2 50G 34M 50G 1% /gluster/brick2 /dev/sda1 1014M 163M 852M 17% /boot tmpfs 783M 0 783M 0% /run/user/0 [root@dhcp35-56 ~]# The heal info also shows brick as online instead of showing transport end point error on the 2 volumes and on the 3rd it is fine [root@dhcp35-56 ~]# gluster v heal vol_-3-1 info Brick 10.70.35.56:/gluster/brick3/vol_-3-1 Status: Connected Number of entries: 0 Brick 10.70.35.227:/gluster/brick3/vol_-3-1 /linux-4.19.6/Documentation/features/sched/membarrier-sync-core/arch-support.txt /linux-4.19.6/arch/arm/mach-dove/include/mach/bridge-regs.h /linux-4.19.6/arch/arm/mach-dove/include/mach/dove.h /linux-4.19.6/Documentation/features/sched/numa-balancing/arch-support.txt /linux-4.19.6/Documentation/w1/masters/ds2482 /linux-4.19.6/Documentation/w1/masters/ds2490 /linux-4.19.6/Documentation/w1/masters/00-INDEX /linux-4.19.6/Documentation/w1/masters/w1-gpio /linux-4.19.6/Documentation/features/seccomp/seccomp-filter/arch-support.txt /linux-4.19.6/Documentation/features/vm/ELF-ASLR/a [root@dhcp35-56 ~]# gluster v heal vol_-3-2 info Brick 10.70.35.56:/gluster/brick3/vol_-3-2 Status: Connected Number of entries: 0 Brick 10.70.35.227:/gluster/brick3/vol_-3-2 /linux-4.19.6/arch/arm/boot/compressed/.gitignore /linux-4.19.6/arch/x86/include/asm/uv/uv_bau.h /linux-4.19.6/arch/arc/plat-eznps/include/plat /linux-4.19.6/arch/x86/include/asm/uv/uv_hub.h /linux-4.19.6/arch/x86/include/asm/uv/uv_irq.h /linux-4.19.6/arch/x86/include/asm/uv/uv_mmrs.h /linux-4.19.6/Documentation/aoe/autoload.sh /linux-4.19.6/arch/x86/include/asm/vdso.h /linux-4.19.6/arch/arm/mach-pxa/palmtc.c /linux-4.19.6/Documentation/DMA-API.txt [root@dhcp35-56 ~]# gluster v heal vol_-3-3 info Brick 10.70.35.56:/gluster/brick3/vol_-3-3 Status: Transport endpoint is not connected Number of entries: - Brick 10.70.35.227:/gluster/brick3/vol_-3-3 Status: Connected Number of entries: 0 Brick 10.70.35.11:/gluster/brick3/vol_-3-3 Status: Connected Number of entries: 0 Very similar to Bug 1450806 - Brick Multiplexing: Brick process shows as online in vol status even when brick is offline and Bug 1450813 - Brick Multiplexing: heal info shows brick as online even when it is brought down both both the bugs have been closed.