Red Hat Bugzilla – Bug 110253
nfs does not mount on bot as netfs parses /etc/fstab before nic has completed auto-neg,
Last modified: 2014-03-16 22:40:28 EDT
Description of problem:
netfs attempts to mount nfs shares whilst nic is still
autonegotiating port speed.
Version-Release number of selected component (if applicable):
redhat es 3
Steps to Reproduce:
1. boot server
no nfs mounts, error is rpc no route to host
nfs mounts should be mounted
a sleep in netfs of a couple of seconds fixes this. The netfs init
script should test the network interface has completed autoneg before
trying to mount nfs shares, or indeed perform any network dependant
This happens on all our ES boxes built on standard HP DL360/DL380
servers connected to a cisco switch infrastructure.
The machine has two static ip address's, 1 bound to each Nic. The
problem appears to be:
ifconfig plumbs the Nics, this then exits, our portmap is started and
netfs attempts to mount the nfs share. Unfortunately the nics whilst
up are still negotiating the connection media, 100 full 100 half 10
half etc, by the time they have negotiated a media the nfs shares
have failed to mount.
When the box has finished booting the nics have negotioated a medium.
meaning I can login and manually mount the share using fstab. ie
I believe this is the same basic problem as in 107999; I attached a
*** This bug has been marked as a duplicate of 107999 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.