Description of problem: ----------------------- In an IPV6 only setup, enabling the shared-storage, created the 'gluster_shared_storage' volume with IPV6 FQDNs, but while adding the fstab mount options there is no IPV6 specific mount options added and the volume fails to mount Version-Release number of selected component (if applicable): -------------------------------------------------------------- RHGS 3.5.2 How reproducible: ----------------- Always Steps to Reproduce: -------------------- 1. Create IPV6 only setup, no IPV4 2. Create geo-rep master volume using IPV6 FQDN 3. Enable shared-storage 4. Verify that the shared storage is mounted Actual results: ---------------- Shared storage fails to mount Expected results: ------------------ Shared storage should get mounted Additional info: ---------------- entry in fstab: --------------- rhsqa-grafton7-ipv6.lab.eng.blr.redhat.com:/gluster_shared_storage /run/gluster/shared_storage glusterfs defaults 0 0 Error in mounting ----------------- <snip> [2020-07-13 09:54:09.316039] I [MSGID: 100030] [glusterfsd.c:2858:main] 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 6.0 (args: /usr/sbin/glusterfs --process-name fuse --volfile-server=rhsqa-grafton7-ipv6.lab.eng.blr.redhat.com --volfile-id=/gluster_shared_storage /run/gluster/shared_storage) [2020-07-13 09:54:09.317134] I [glusterfsd.c:2567:daemonize] 0-glusterfs: Pid of current running process is 1879648 [2020-07-13 09:54:09.321866] E [MSGID: 101075] [common-utils.c:508:gf_resolve_ip6] 0-resolver: getaddrinfo failed (family:2) (Name or service not known) [2020-07-13 09:54:09.321891] E [name.c:266:af_inet_client_get_remote_sockaddr] 0-glusterfs: DNS resolution failed on host rhsqa-grafton7-ipv6.lab.eng.blr.redhat.com [2020-07-13 09:54:09.322047] I [glusterfsd-mgmt.c:2443:mgmt_rpc_notify] 0-glusterfsd-mgmt: disconnected from remote-host: rhsqa-grafton7-ipv6.lab.eng.blr.redhat.com [2020-07-13 09:54:09.322082] I [glusterfsd-mgmt.c:2463:mgmt_rpc_notify] 0-glusterfsd-mgmt: Exhausted all volfile servers [2020-07-13 09:54:09.322124] I [MSGID: 101190] [event-epoll.c:688:event_dispatch_epoll_worker] 0-epoll: Started thread with index 0 [2020-07-13 09:54:09.322149] I [MSGID: 101190] [event-epoll.c:688:event_dispatch_epoll_worker] 0-epoll: Started thread with index 1 [2020-07-13 09:54:09.322171] W [glusterfsd.c:1581:cleanup_and_exit] (-->/lib64/libgfrpc.so.0(+0xf5b3) [0x7fc5ab8935b3] -->/usr/sbin/glusterfs(+0x13c97) [0x564f0d191c97] -->/usr/sbin/glusterfs(cleanup_and_exit+0x58) [0x564f0d1866b8] ) 0-: received signum (1), shutting down [2020-07-13 09:54:09.322195] I [fuse-bridge.c:6917:fini] 0-fuse: Unmounting '/run/gluster/shared_storage'. [2020-07-13 09:54:09.334362] I [fuse-bridge.c:6922:fini] 0-fuse: Closing fuse connection to '/run/gluster/shared_storage'. [2020-07-13 09:54:09.334496] W [glusterfsd.c:1581:cleanup_and_exit] (-->/lib64/libpthread.so.0(+0x82de) [0x7fc5aa66c2de] -->/usr/sbin/glusterfs(glusterfs_sigwaiter+0xfd) [0x564f0d18686d] -->/usr/sbin/glusterfs(cleanup_and_exit+0x58) [0x564f0d1866b8] ) 0-: received signum (15), shutting down </snip>
Error is gone with adding fstab mount option as 'xlator-option=transport.address-family=inet6'
Tested with glusterfs-6.0-51.el8rhgs by enabling IPV6 and seeing 2 issues with IPV6 environment. 1. The shared storage volume is not even created. 2. No mount option is set with new entry in /etc/fstab
Upstream patch link: https://github.com/gluster/glusterfs/pull/1972
Verified with the RHGS 3.5.4 interim build - glusterfs-6.0-52.el8rhgs with the following steps repeated across RHEL 7.9 and RHEL 8.3 platforms 1. Create a new gluster cluster and volume using IPV4 FQDN 2. Enable shared storage using the command: # gluster volume set all cluster.enable-shared-storage enabled 3. Verify that the shared storage volume is created, and fuse mounted. Also has the appropriate entry in /etc/fstab 4. Reboot one of the hosts, and make sure that the shared storage is mounted Repeated the above test with cluster and volume created with IPV4 direct addresses Repeated the above test with cluster and volume created with IPV6 direct addresses Repeated the above test with cluster and volume created with IPV6 FQDNs All the cases works good
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 (glusterfs bug fix and enhancement update), 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://access.redhat.com/errata/RHBA-2021:1462