Thinking Out Loud

May 14, 2017

DBFS and XAG for Goldengate P4

Filed under: 12c,GoldenGate — mdinh @ 3:13 pm

I encountered a situation where I was not able to start dbfs_mount to save my life.

2 people spent over 5 hours and was not able to resolve the issue.

1 person suggested to to restart clusterware because it looks like fuse group was added to oracle after oracle clusterware was up.

During configuration:
kernel-devel was missing, /etc/rc.modules did not exists, oracle was not part of fuse group

# rpm -qa |egrep 'fuse|kernel-devel'
# cat /etc/rc.modules
cat: /etc/rc.modules: No such file or directory
# grep fuse /etc/group
fuse:x:993:

Corrections made:

# cat /etc/rc.modules
/sbin/modprobe fuse

# grep fuse /etc/group
fuse:x:993:oracle
$ crsctl start res dbfs_mount
CRS-2672: Attempting to start 'dbfs_mount' on 'hawk1'
CRS-2672: Attempting to start 'dbfs_mount' on 'hawk2'
CRS-2674: Start of 'dbfs_mount' on 'hawk2' failed
CRS-2679: Attempting to clean 'dbfs_mount' on 'hawk2'
CRS-2674: Start of 'dbfs_mount' on 'hawk1' failed
CRS-2679: Attempting to clean 'dbfs_mount' on 'hawk1'
CRS-2681: Clean of 'dbfs_mount' on 'hawk1' succeeded
CRS-2681: Clean of 'dbfs_mount' on 'hawk2' succeeded
CRS-4000: Command Start failed, or completed with errors.

$ crsctl stat res dbfs_mount
NAME=dbfs_mount
TYPE=local_resource
TARGET=ONLINE , ONLINE
STATE=OFFLINE, OFFLINE

After restarting clusterware:

Note: this was started by root which I don't think is a good idea.
# crsctl start resource dbfs_mount
CRS-2672: Attempting to start 'dbfs_mount' on 'hawk1'
CRS-2676: Start of 'dbfs_mount' on 'hawk1' succeeded

Should have been started from grid user
$ crsctl start res dbfs_mount
CRS-2672: Attempting to start 'dbfs_mount' on 'hawk2'
CRS-2672: Attempting to start 'dbfs_mount' on 'hawk1'
CRS-2676: Start of 'dbfs_mount' on 'hawk1' succeeded
CRS-2676: Start of 'dbfs_mount' on 'hawk2' succeeded

$ crsctl stat res dbfs_mount
NAME=dbfs_mount
TYPE=local_resource
TARGET=ONLINE             , ONLINE
STATE=ONLINE on hawk1, ONLINE on hawk2

It really bothered me how this information was missed and where was it documented.

After hours of research, the information is not consistently documented.
Not Found:
White Paper: How To Setup 12c DBFS FileSystem. (Doc ID 1938421.1)

Found:
Configuring DBFS on Oracle Exadata Database Machine (Doc ID 1054431.1)

To pick up the additional group (fuse) membership for the oracle user on Linux or the workaround above on Solaris, Clusterware must be restarted.
For example, to restart Clusterware on all nodes at the same time (non-rolling), you can use the following commands as root:

Hopefully, you won’t have a much fun as I did.

Advertisements

Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: