Bug 107999
Summary: | NFS filesystems don't mount on boot | ||||||
---|---|---|---|---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | Joshua Weage <weage98> | ||||
Component: | initscripts | Assignee: | Bill Nottingham <notting> | ||||
Status: | CLOSED DUPLICATE | QA Contact: | Brock Organ <borgan> | ||||
Severity: | medium | Docs Contact: | |||||
Priority: | medium | ||||||
Version: | 9 | CC: | danny.hallwood, eric.eisenhart, jplans, rvokal | ||||
Target Milestone: | --- | ||||||
Target Release: | --- | ||||||
Hardware: | i686 | ||||||
OS: | Linux | ||||||
Whiteboard: | |||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||
Doc Text: | Story Points: | --- | |||||
Clone Of: | Environment: | ||||||
Last Closed: | 2006-02-21 18:59:24 UTC | Type: | --- | ||||
Regression: | --- | Mount Type: | --- | ||||
Documentation: | --- | CRM: | |||||
Verified Versions: | Category: | --- | |||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
Cloudforms Team: | --- | Target Upstream Version: | |||||
Embargoed: | |||||||
Attachments: |
|
Description
Joshua Weage
2003-10-25 23:45:08 UTC
Created attachment 96453 [details]
patch to /etc/init.d/netfs (initscripts) that waits for NFS servers to be available
We've encountered a similar problem on RedHat 7.3 servers with
initscripts-6.67-1. netfs fails to mount during bootup, but will succeed
later. We don't have any NIS.
The problem appears to be that the network driver modules reset the ethernet
cards during the "network" phase of bootup and the network doesn't fully
function until some time afterward. The time from "service network start"
(module loading) until full operation may be longer due to these machines
havine 3 ethernet interfaces.
I haven't checked, but I'm assuming that when the network card isn't fully
"on", the mount attempts get a "network unreachable" error instead of a "host
unreachable" and therefore don't retry as they normally would if the host was
merely unreachable.
Our fix has been to add a piece to the top of /etc/init.d/netfs that tries to
ping its NFS servers for a while before mounting. It's written such that
successful pinging will go past that step in a second or two and unsuccessful
pinging will wait up to 5 minutes before proceeding. The long wait is
acceptable in our case, since the servers can't accomplish their intended tasks
without access to the NFS server. A superior fix might be to have
/etc/init.d/network try a similar pinging routine on its own IPs for up to 30
seconds per interface.
I'm attaching the modified network script. The ping loops are to avoid any
possible problems with ping binding to the wrong network interface due to the
correct interface not being available yet.
*** Bug 110253 has been marked as a duplicate of this bug. *** *** This bug has been marked as a duplicate of 97610 *** Changed to 'CLOSED' state since 'RESOLVED' has been deprecated. |