[sheepdog-users] Sheep crash after cluster shutdown on debian wheezy

[Valerio Pachera]

On a single host with no data

/usr/sbin/sheep -n
/var/lib/sheepdog,/mnt/sheep/0,/mnt/sheep/1,/mnt/sheep/2 --cluster
zookeeper:192.168.6.45:2181

cat /var/lib/sheepdog/sheep.log
Aug 19 16:25:28   INFO [main] md_add_disk(368) /mnt/sheep/0, vdisk nr
413, total disk 1
Aug 19 16:25:28   INFO [main] md_add_disk(368) /mnt/sheep/1, vdisk nr
413, total disk 2
Aug 19 16:25:28   INFO [main] md_add_disk(368) /mnt/sheep/2, vdisk nr
1741, total disk 3
Aug 19 16:25:28   INFO [main] zk_init(1498) version 3.3.5, address
192.168.6.45:2181/sheepdog, timeout 30000
Aug 19 16:25:28   INFO [main] zk_init(1503) the negociated session
timeout is 30000
Aug 19 16:25:28 NOTICE [main] get_local_addr(551) found IPv4 address
Aug 19 16:25:28   INFO [main] send_join_request(1091) IPv4
ip:192.168.5.45 port:7000 going to join the cluster
Aug 19 16:25:28 NOTICE [main] nfs_init(647) nfs server service is not compiled
Aug 19 16:25:28   INFO [main] main(1044) sheepdog daemon (version
0.9.0_244_gff8590b) started

dog cluster shutdown
sleep 30
/usr/sbin/sheep -n
/var/lib/sheepdog,/mnt/sheep/0,/mnt/sheep/1,/mnt/sheep/2 --cluster
zookeeper:192.168.6.45:2181

cat /var/lib/sheepdog/sheep.log
Aug 19 16:25:28   INFO [main] md_add_disk(368) /mnt/sheep/0, vdisk nr
413, total disk 1
Aug 19 16:25:28   INFO [main] md_add_disk(368) /mnt/sheep/1, vdisk nr
413, total disk 2
Aug 19 16:25:28   INFO [main] md_add_disk(368) /mnt/sheep/2, vdisk nr
1741, total disk 3
Aug 19 16:25:28   INFO [main] zk_init(1498) version 3.3.5, address
192.168.6.45:2181/sheepdog, timeout 30000
Aug 19 16:25:28   INFO [main] zk_init(1503) the negociated session
timeout is 30000
Aug 19 16:25:28 NOTICE [main] get_local_addr(551) found IPv4 address
Aug 19 16:25:28   INFO [main] send_join_request(1091) IPv4
ip:192.168.5.45 port:7000 going to join the cluster
Aug 19 16:25:28 NOTICE [main] nfs_init(647) nfs server service is not compiled
Aug 19 16:25:28   INFO [main] main(1044) sheepdog daemon (version
0.9.0_244_gff8590b) started
Aug 19 16:25:42   INFO [main] rx_main(836) req=0x26d1f70, fd=18,
client=127.0.0.1:58974, op=SHUTDOWN, data=(null)
Aug 19 16:25:42   INFO [main] tx_main(894) req=0x26d1f70, fd=18,
client=127.0.0.1:58974, op=SHUTDOWN, result=00
Aug 19 16:25:42   INFO [main] main(1052) shutdown
Aug 19 16:25:42   INFO [main] zk_leave(1036) leaving from cluster
Aug 19 16:26:26   INFO [main] md_add_disk(368) /mnt/sheep/0, vdisk nr
413, total disk 1
Aug 19 16:26:26   INFO [main] md_add_disk(368) /mnt/sheep/1, vdisk nr
413, total disk 2
Aug 19 16:26:26   INFO [main] md_add_disk(368) /mnt/sheep/2, vdisk nr
1741, total disk 3
Aug 19 16:26:26   INFO [main] zk_init(1498) version 3.3.5, address
192.168.6.45:2181/sheepdog, timeout 30000
Aug 19 16:26:26   INFO [main] zk_init(1503) the negociated session
timeout is 30000
Aug 19 16:26:26 NOTICE [main] get_local_addr(551) found IPv4 address
Aug 19 16:26:26   INFO [main] send_join_request(1091) IPv4
ip:192.168.5.45 port:7000 going to join the cluster
Aug 19 16:26:26 NOTICE [main] nfs_init(647) nfs server service is not compiled
Aug 19 16:26:26   INFO [main] main(1044) sheepdog daemon (version
0.9.0_244_gff8590b) started
Aug 19 16:26:26  EMERG [main] crash_handler(288) sheep exits
unexpectedly (Segmentation fault), si pid 0, uid 0, errno 0, code 1
Aug 19 16:26:26  EMERG [main] sd_backtrace(835) sheep.c:290: crash_handler
Aug 19 16:26:26  EMERG [main] sd_backtrace(849)
/lib/x86_64-linux-gnu/libpthread.so.0(+0xf09f) [0x7f9b8d68709f]
Aug 19 16:26:26  EMERG [main] sd_backtrace(835) list.h:76: __list_add
Aug 19 16:26:26  EMERG [main] sd_backtrace(835) group.c:1037: sd_notify_handler
Aug 19 16:26:26  EMERG [main] sd_backtrace(835) zookeeper.c:1310:
zk_event_handler
Aug 19 16:26:26  EMERG [main] sd_backtrace(835) event.c:220: do_event_loop
Aug 19 16:26:26  EMERG [main] sd_backtrace(835) sheep.c:1049: main
Aug 19 16:26:27  EMERG [main] sd_backtrace(849)
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xfc)
[0x7f9b8cc56eac]
Aug 19 16:26:27  EMERG [main] sd_backtrace(849) /usr/sbin/sheep() [0x4062b8]

It's repeatable.
It doesn't happen on debian jessie.

Sheepdog daemon version 0.9.0_244_gff8590b