Hello all, We are seeing a strange issue where a new node gfs3 shows another node gfs2 as not connected on the "gluster volume heal" info:
[root@gfs3 bricks]# gluster volume heal gvol0 info Brick gfs1:/nodirectwritedata/gluster/gvol0 Status: Connected Number of entries: 0 Brick gfs2:/nodirectwritedata/gluster/gvol0 Status: Transport endpoint is not connected Number of entries: - Brick gfs3:/nodirectwritedata/gluster/gvol0 Status: Connected Number of entries: 0 However it does show the same node connected on "gluster peer status". Does anyone know why this would be? [root@gfs3 bricks]# gluster peer status Number of Peers: 2 Hostname: gfs2 Uuid: 91863102-23a8-43e1-b3d3-f0a1bd57f350 State: Peer in Cluster (Connected) Hostname: gfs1 Uuid: 32c99e7d-71f2-421c-86fc-b87c0f68ad1b State: Peer in Cluster (Connected) In nodirectwritedata-gluster-gvol0.log on gfs3 we see this logged with regards to gfs2: [2019-05-29 00:17:50.646360] I [MSGID: 115029] [server-handshake.c:537:server_setvolume] 0-gvol0-server: accepted client from CTX_ID:30d74196-fece-4380-adc0-338760188b81-GRAPH_ID:0-PID:7718-HOST:gfs2.xxx.com-PC_NAME:gvol0-client-2-RECON_NO:-0 (version: 5.6) [2019-05-29 00:17:50.761120] I [MSGID: 115036] [server.c:469:server_rpc_notify] 0-gvol0-server: disconnecting connection from CTX_ID:30d74196-fece-4380-adc0-338760188b81-GRAPH_ID:0-PID:7718-HOST:gfs2.xxx.com-PC_NAME:gvol0-client-2-RECON_NO:-0 [2019-05-29 00:17:50.761352] I [MSGID: 101055] [client_t.c:435:gf_client_unref] 0-gvol0-server: Shutting down connection CTX_ID:30d74196-fece-4380-adc0-338760188b81-GRAPH_ID:0-PID:7718-HOST:gfs2.xxx.com-PC_NAME:gvol0-client-2-RECON_NO:-0 Thanks in advance for any assistance. -- David Cunningham, Voisonics Limited http://voisonics.com/ USA: +1 213 221 1092 New Zealand: +64 (0)28 2558 3782
_______________________________________________ Gluster-users mailing list Gluster-users@gluster.org https://lists.gluster.org/mailman/listinfo/gluster-users
