[sheepdog-users] collie node list There are no active sheep daemons

[Valerio Pachera]

2013/3/17 Hitoshi Mitake <mitake.hitoshi at gmail.com>:
> I'm grad to hear the good result :)

Hi, I'm sorry, but I still have problems :-(
I'm trying to start the cluster from scratch (the sheep folders are empty).

But I have strange behavior:
- I get 'no active sheep daemons' or
- 'collie node list' hangs up.
- I can't kill corosync, only by kill -9
- Hosts do not shutdown by 'halt' or 'shutdown'

I think /etc/hosts might be related.

In corosync.conf I set the actual ip address (no host name).
So dns should not matter, or does sheepdog/corosync use dns for some reason?

If it does, wich is the correct ip to write (loopback or ethX?)

127.0.0.1       localhost
#127.0.1.1      sheepdog002
192.168.6.42    sheepdog002

(I already tried both, with no success).

I attach some log but they don't seem very helpful.
I'm probably stuck on something really stupid.
-------------- next part --------------
syslog
Mar 18 15:51:39 sheepdog002 corosync[3368]:   [MAIN  ] Corosync Cluster Engine ('1.4.2'): started and ready to provide service.
Mar 18 15:51:39 sheepdog002 corosync[3368]:   [MAIN  ] Corosync built-in features: nss
Mar 18 15:51:39 sheepdog002 corosync[3368]:   [MAIN  ] Successfully read main configuration file '/etc/corosync/corosync.conf'.
Mar 18 15:51:39 sheepdog002 corosync[3368]:   [TOTEM ] Initializing transport (UDP/IP Multicast).
Mar 18 15:51:39 sheepdog002 corosync[3368]:   [TOTEM ] Initializing transmit/receive security: libtomcrypt SOBER128/SHA1HMAC (mode 0).
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [TOTEM ] The network interface [192.168.6.42] is now up.
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [SERV  ] Service failed to load 'pacemaker'.
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [SERV  ] Service engine loaded: corosync extended virtual synchrony service
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [SERV  ] Service engine loaded: corosync configuration service
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [SERV  ] Service engine loaded: corosync cluster closed process group service v1.01
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [SERV  ] Service engine loaded: corosync cluster config database access v1.01
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [SERV  ] Service engine loaded: corosync profile loading service
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [SERV  ] Service engine loaded: corosync cluster quorum service v0.1
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [MAIN  ] Compatibility mode set to whitetank.  Using V1 and V2 of the synchronization engine.
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [TOTEM ] A processor joined or left the membership and a new membership was formed.
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [CPG   ] chosen downlist: sender r(0) ip(192.168.6.42) ; members(old:0 left:0)
Mar 18 15:51:40 sheepdog002 corosync[3368]:   [MAIN  ] Completed service synchronization, ready to provide service.

sheep.log
Mar 18 15:48:22 [main] send_join_request(1060) IPv4 ip:192.168.6.42 port:7000
Mar 18 15:48:22 [main] check_host_env(377) WARN: Allowed open files 1024 too small, suggested 1024000
Mar 18 15:48:22 [main] check_host_env(386) Allowed core file size 0, suggested unlimited
Mar 18 15:48:22 [main] main(678) sheepdog daemon (version 0.5.5_188_g8b3d993) started
Mar 18 15:51:54 [main] jrnl_recover(230) opening the directory /mnt/sheepdog/journal/
Mar 18 15:51:54 [main] jrnl_recover(235) starting journal recovery
Mar 18 15:51:54 [main] jrnl_recover(291) journal recovery complete
Mar 18 15:51:54 [main] send_join_request(1060) IPv4 ip:192.168.6.42 port:7000
Mar 18 15:51:54 [main] check_host_env(377) WARN: Allowed open files 1024 too small, suggested 1024000
Mar 18 15:51:54 [main] check_host_env(386) Allowed core file size 0, suggested unlimited
Mar 18 15:51:54 [main] main(678) sheepdog daemon (version 0.5.5_188_g8b3d993) started