Can you show us a test script to reproduce this issue? On Mon, Aug 20, 2012 at 10:17 AM, Brook <jingyu_chen at hotmail.com> wrote: > Hi: > >> The 36th version have 6 nodes: 192.168.146[65-69,71], but in the 37th >> version you didn't start 192.168.146.71 node. >> > > Thank you for reply. > '192.168.146.71' and "192.168.146.69" are gateways, i've shut them down. now the cluster only contains 4 nodes. > 192.168.146.65:7000 192.168.146.66:7000 192.168.146.67:7000 192.168.146.68:7000 > I think there should be a "The 38th version" epoch which contains 4 nodes, but for some reasons which i don't know, the nodes can't make a whole cluster. > > > -------------------------------------------------- > From: "Yunkai Zhang" <yunkai.me at gmail.com> > Sent: Monday, August 20, 2012 10:04 AM > To: "Brook" <jingyu_chen at hotmail.com> > Cc: <sheepdog at lists.wpkg.org> > Subject: Re: [sheepdog] I got a"Waiting for other nodes to join cluster" > >> On Mon, Aug 20, 2012 at 9:57 AM, Brook <jingyu_chen at hotmail.com> wrote: >>> Hi All. >>> After a failure of switcher, my sheepdog cluster can't run. >>> I got the message below, what should i do ? >>> [root at 17-IDC-D-2115 ~]# collie vdi list >>> Name Id Size Used Shared Creation time VDI id Tag >>> Failed to read object 8083d2b800000000 Waiting for other nodes to join >>> cluster >>> Failed to read inode header >>> Failed to read object 8083d46b00000000 Waiting for other nodes to join >>> cluster >>> Failed to read inode header >>> Failed to read object 8083d7d100000000 Waiting for other nodes to join >>> cluster >>> Failed to read inode header >>> Failed to read object 8083d7d200000000 Waiting for other nodes to join >>> cluster >>> Failed to read inode header >>> Failed to read object 8083db3700000000 Waiting for other nodes to join >>> cluster >>> Failed to read inode header >>> Failed to read object 8083de9d00000000 Waiting for other nodes to join >>> cluster >>> Failed to read inode header >>> Failed to read object 809bee7c00000000 Waiting for other nodes to join >>> cluster >>> Failed to read inode header >>> Failed to read object 809bf02f00000000 Waiting for other nodes to join >>> cluster >>> ...... >>> >>> [root at 17-IDC-D-2115 ~]# collie vdi create vol1 1G >>> Failed to create VDI vol1: Waiting for other nodes to join cluster >>> >>> [root at 17-IDC-D-2115 ~]# corosync-cpgtool >>> Group Name PID Node ID >>> sheepdog >>> 12747 1100130496 (192.168.146.65) >>> 8228 1150462144 (192.168.146.68) >>> 15328 1133684928 (192.168.146.67) >>> 2076 1116907712 (192.168.146.66) >>> >>> [root at 17-IDC-D-2115 ~]# collie node list >>> M Id Host:Port V-Nodes Zone >>> - 0 192.168.146.65:7000 64 1100130496 >>> - 1 192.168.146.66:7000 64 1116907712 >>> - 2 192.168.146.67:7000 64 1133684928 >>> - 3 192.168.146.68:7000 64 1150462144 >>> >>> [root at 17-IDC-D-2115 ~]# collie node info >>> Id Size Used Use% >>> Cannot get information from any nodes >>> >>> [root at 17-IDC-D-2115 ~]# collie cluster info >>> Cluster status: Waiting for other nodes to join cluster >>> >>> Cluster created at Mon Jul 9 16:57:18 2012 >>> >>> Epoch Time Version >>> 2012-08-16 14:51:45 37 [192.168.146.65:7000, 192.168.146.66:7000, >>> 192.168.146.67:7000, 192.168.146.68:7000, 192.168.146.69:7000] >>> 2012-08-16 14:51:44 36 [192.168.146.65:7000, 192.168.146.66:7000, >>> 192.168.146.67:7000, 192.168.146.68:7000, 192.168.146.69:7000, >>> 192.168.146.71:7000] >> >> The 36th version have 6 nodes: 192.168.146[65-69,71], but in the 37th >> version you didn't start 192.168.146.71 node. >> >>> ...... >>> >>> [root at 17-IDC-D-2115 ~]# tail -n30 /data/sheepdog/sheep.log >>> Aug 20 09:28:11 [main] listen_handler(819) accepted a new connection: 13 >>> Aug 20 09:28:11 [main] client_rx_handler(577) connection from: 13, ::1:45306 >>> Aug 20 09:28:11 [main] queue_request(323) 82 >>> Aug 20 09:28:11 [io 18] do_process_work(990) 82, 0 , 37 >>> Aug 20 09:28:11 [main] client_tx_handler(663) connection from: 13, ::1:45306 >>> Aug 20 09:28:11 [main] client_handler(764) connection seems to be dead >>> Aug 20 09:28:11 [main] clear_client(703) refcnt:0, fd:13, ::1:45306 >>> Aug 20 09:28:11 [main] destroy_client(672) connection from: ::1:45306 >>> Aug 20 09:28:11 [main] listen_handler(819) accepted a new connection: 13 >>> Aug 20 09:28:11 [main] client_rx_handler(577) connection from: 13, ::1:45307 >>> Aug 20 09:28:11 [main] queue_request(323) 11 >>> Aug 20 09:28:11 [main] client_tx_handler(663) connection from: 13, ::1:45307 >>> Aug 20 09:28:11 [main] client_handler(764) connection seems to be dead >>> Aug 20 09:28:11 [main] clear_client(703) refcnt:0, fd:13, ::1:45307 >>> Aug 20 09:28:11 [main] destroy_client(672) connection from: ::1:45307 >>> Aug 20 09:28:13 [main] listen_handler(819) accepted a new connection: 13 >>> Aug 20 09:28:13 [main] client_rx_handler(577) connection from: 13, ::1:45308 >>> Aug 20 09:28:13 [main] queue_request(323) 82 >>> Aug 20 09:28:13 [io 19] do_process_work(990) 82, 0 , 37 >>> Aug 20 09:28:13 [main] client_tx_handler(663) connection from: 13, ::1:45308 >>> Aug 20 09:28:13 [main] client_handler(764) connection seems to be dead >>> Aug 20 09:28:13 [main] clear_client(703) refcnt:0, fd:13, ::1:45308 >>> Aug 20 09:28:13 [main] destroy_client(672) connection from: ::1:45308 >>> Aug 20 09:28:13 [main] listen_handler(819) accepted a new connection: 13 >>> Aug 20 09:28:13 [main] client_rx_handler(577) connection from: 13, ::1:45309 >>> Aug 20 09:28:13 [main] queue_request(323) 11 >>> Aug 20 09:28:13 [main] client_tx_handler(663) connection from: 13, ::1:45309 >>> Aug 20 09:28:13 [main] client_handler(764) connection seems to be dead >>> Aug 20 09:28:13 [main] clear_client(703) refcnt:0, fd:13, ::1:45309 >>> Aug 20 09:28:13 [main] destroy_client(672) connection from: ::1:45309 >>> >>> >>> -- >>> sheepdog mailing list >>> sheepdog at lists.wpkg.org >>> http://lists.wpkg.org/mailman/listinfo/sheepdog >>> >> >> >> >> -- >> Yunkai Zhang >> Work at Taobao >> -- Yunkai Zhang Work at Taobao |