Mounting NFS Filesystems at startup, always fails..

stellarmarine1 · 08-24-2004, 02:33 PM

So here's my situation... I have roughly 42 Oracle Servers running Red Hat AS 2.1 that need to access an NFS server.

Here's my Problem. When booting it goes through the process of mounting the NFS and fails everytime. It fails with the error: No route to host. As soon as the system is booted up I can run a mount -a just fine, no errors. I would like to avoid having to run this command on all 42 servers when we need to reboot.

I have removed a few details for security reason...

Here is an example of my /etc/fstab file

Code:

LABEL=/                 /                       ext3    defaults        1 1
LABEL=/boot             /boot                   ext3    defaults        1 2
none                    /dev/pts                devpts  gid=5,mode=620  0 0
LABEL=/home             /home                   ext3    defaults        1 2
none                    /proc                   proc    defaults        0 0
none                    /dev/shm                tmpfs   defaults        0 0
LABEL=/tmp              /tmp                    ext3    defaults        1 2
LABEL=/usr              /usr                    ext3    defaults        1 2
LABEL=/var              /var                    ext3    defaults        1 2
/dev/sda7               swap                    swap    defaults        0 0
/dev/cdrom              /mnt/cdrom              iso9660 noauto,owner,kudzu,ro 0 0
/dev/fd0                /mnt/floppy             auto    noauto,owner,kudzu 0 0
myserver1:/export       /mnt/myserver           nfs     defaults

and... my /etc/hosts file

Code:

# Do not remove the following line, or various programs
# that require network functionality will fail.
127.0.0.1               localhost.localdomain localhost mylocalserver.mydomain.mil
555.555.555.555         myserver1.mydomain.mil myserver1
555.555.555.555         myserver2.mydomain.mil  myserver2

Thanks for the help in advance...

acid_kewpie · 08-24-2004, 02:36 PM

sounds like it is somehow loading network file systems before the network is suitably loaded. Does redhat had netmounts as a seperate service? if so, try nudgind the priortity of it down to after any other network related services.

stellarmarine1 · 08-24-2004, 06:28 PM

thanks, I have actually thought about making it mount the NFS last, but I am not sure how to do that. I peeked around in my rc.d files, but wasn't really sure what to add or edit. Maybe I should start a new thread asking that.

acid_kewpie · 08-25-2004, 01:56 AM

here's fine... so is there a service for the network mounts? it's more than likely that there is... if you do find it, it'll have a runlevel symlink for it, e.g. /etc/rc5.d/S40netmount, just change that number to somethign higher, e.g. 90 and then it'll get processed later in the bootup.

mind you the default settings should be fine in principle, does the network interface itself visibly load before then? have you changed anythign else about this?

stellarmarine1 · 08-25-2004, 06:19 AM

nope, it loads the ethernet interfaces before it loads the mounts. I am going to try to bump up the symlink number and see if that helps. I'll let you know, soon.

stellarmarine1 · 08-25-2004, 07:56 AM

I went ahead and renamed /etc/rc.d/rc3.d/S25netfs to S85netfs and everything is starting correctly now. Thanks a lot for all the help..

cmon_stoke · 08-25-2004, 10:04 AM

I am having the same problem with redhat 7.2, I can't alter the init value, so I have to debug the problem.

This is how far I have currently got

network loads eth0 and eth1 with an init value of 10

portmap loads with an int value of 13

I have modified netfs ( int val of 25) and just after it checks portmap
I have added the following

Code:

action $"ping dataserver: " /bin/ping -c10 server.ip
action $" check rpc info for dataserver : " /usr/sbin/rpcinfo -p dataserver.ip

the ping sends ten packets and receives no replies, while the rpcinfo fails to connect to the dataserver
Then the mount SUCEEDS...

If I replace the the ping and rpcinfo actions with

Code:

action $" Sleep for a couple of mins : " sleep 2m

the mount always fails. This has lead me to believe that the network modules are loaded but the physical connection is still not up until I prod it with the network operations.

EdR · 02-03-2005, 12:40 PM

Quote:

here's fine... so is there a service for the network mounts? it's more than likely that there is... if you do find it, it'll have a runlevel symlink for it, e.g. /etc/rc5.d/S40netmount, just change that number to somethign higher, e.g. 90 and then it'll get processed later in the bootup.

This is what helped me the most. I did figure out that you have to change the value in the rc"x".d directory for your particular runlevel ex- if you are running in run level 5 you modify the file in the rc5.d directory.