6 messages in com.mysql.lists.clusterRe: Gentoo Linux and NDBD doestn work| From | Sent On | Attachments |
|---|---|---|
| Norbert Schmidt | 09 Sep 2007 06:05 | |
| Matthew Montgomery | 09 Sep 2007 08:17 | |
| Norbert Schmidt | 09 Sep 2007 08:50 | |
| Matthew Montgomery | 09 Sep 2007 09:23 | |
| Norbert Schmidt | 09 Sep 2007 09:51 | |
| Norbert Schmidt | 09 Sep 2007 09:51 |
| Subject: | Re: Gentoo Linux and NDBD doestn work |
|---|---|
| From: | Norbert Schmidt (clus...@rocknob.de) |
| Date: | 09/09/2007 09:51:27 AM |
| List: | com.mysql.lists.cluster |
Which start phase are these nodes blocking at?
Use the "ALL STATUS" command in ndb_mgm to find out.
both Nodes in phase 1
Does the same error still appear in the ndb_1_cluster.log ?
2007-09-09 14:46:09 [MgmSrvr] INFO -- Node 1: Initial start, waiting for 0000000000000004 to connect, nodes [ all: 0000000000000006
If it does then it is likely a still in Phase 1. The problem being that the two ndbd processes are able to communicate with the ndb_mgmd process but not with each other. This is caused by the fact that one ndbd process is connected via 127.0.0.1 (lo) and the other is from 192.168.0.1 (eth0). You should set your ndb_connectstring=192.168.1.109 rather than 127.0.0.1 so that the ndbd traffic occurs over the same interface.
thats right. now i connect two times from the second server and that works.
in the configuration file i have configured hostnames, and from both hosts the hostname doesnt lookup to 127.0.0.1
now i have configerd it without hostnames and with ip-adresses and it runs perfekt :)
ndb_mgm -> show
that shows me the same but it runs only if i have configured ip adresses, im a little bit confused. the nameresolution works perfekt