Re: Cannot restart node after server crash (possible bug)

[Darren Landrum]

Okay, all four NDB nodes are up and running! YAY! A victory, to be sure.

I still have two API nodes that are down, one in each nodegroup. It looks like this:

-----

ndb_mgm> show
Cluster Configuration
---------------------
[ndbd(NDB)] 4 node(s)
id=3 @xxx.xxx.xxx.58 (Version: 4.1.11, Nodegroup: 0)
id=4 (Version: 4.1.11, Nodegroup: 0, Master)
id=5 (Version: 4.1.11, Nodegroup: 1)
id=6 @xxx.xxx.xxx.61 (Version: 4.1.11, Nodegroup: 1)

[ndb_mgmd(MGM)] 1 node(s)
id=1 @xxx.xxx.xxx.57 (Version: 4.1.11)

[mysqld(API)] 4 node(s)
id=9 (not connected, accepting connect from xxx.xxx.xxx.58)
id=10 @xxx.xxx.xxx.59 (Version: 4.1.11)
id=11 @xxx.xxx.xxx.60 (Version: 4.1.11)
id=12 @xxx.xxx.xxx.61 (Version: 4.1.11)

-----

I was told that doing /etc/rc.d/rc.mysql.mysqld restart fixes this problem, and indeed it did for node 12, but not for 9, where I get this error:

-----

root@mail-c1:/etc/rc.d# ./rc.mysql.mysqld restart
mysqld: no process killed
root@mail-c1:/etc/rc.d# Configuration error: Could not alloc node id at xxx.xxx.xxx.57 port 1186: No free node id found for mysqld(API).
050706 15:43:39 [ERROR] Can't init databases
050706 15:43:39 [ERROR] Aborting

050706 15:43:39 [Note] /usr/local/mysql/libexec/mysqld: Shutdown complete


root@mail-c1:/etc/rc.d#

-----

Any idea?

Again, many many thanks for your help.