Bug 622799

hi,

I have been unable to reproduce this issue locally.

could you please attach a tarball of /etc/rc.d/, the list of packages installed
on the system (rpm -q -a will do), and if you still have the chkconfig sequence
to enable the different init scripts?

Thanks

This issue has been proposed when we are only considering blocker
issues in the current Red Hat Enterprise Linux release.

** If you would still like this issue considered for the current
release, ask your support representative to file as a blocker on
your behalf. Otherwise ask that it be considered for the next
Red Hat Enterprise Linux release. **

Created attachment 437881 [details]
rpm -qa --last output

rpm -qa --last output

Created attachment 437883 [details]
tar zcvf /tmp/rhel601_etc_rc_d.tgz etc/rc.d/

gtar ztvf rhel601_etc_rc_d.tgz  to unpack.

(In reply to comment #4)
> Created an attachment (id=437881) [details]
> rpm -qa --last output
> 
> rpm -qa --last output    

Hi Everett,

thanks for the quick response but I believe the attachment doesn´t contain the correct information.

Thanks
Fabio

Created attachment 437890 [details]
rpm -qa --last output

rpm -qa --last output

Hope this works......using mozilla 3.6.4 on FC 13.
When I try to select the file to attach a different file is attached.
Had to rename file.

Also, unable to enter the path to the file in the 'browse' field.

Here is the attachment again.

[intec@uxeverett tmp]$ head rhel601rpmqa.txt 
kernel-headers-2.6.32-44.2.el6                Tue 03 Aug 2010 03:45:03 PM EDT
perf-2.6.32-44.2.el6                          Tue 03 Aug 2010 03:45:01 PM EDT
kernel-devel-2.6.32-44.2.el6                  Tue 03 Aug 2010 03:44:20 PM EDT
kernel-2.6.32-44.2.el6                        Tue 03 Aug 2010 03:43:38 PM EDT
kernel-firmware-2.6.32-44.2.el6               Tue 03 Aug 2010 03:43:31 PM EDT
yum-plugin-versionlock-1.1.26-10.el6          Wed 28 Jul 2010 02:06:07 PM EDT
yum-plugin-aliases-1.1.26-10.el6              Wed 28 Jul 2010 02:06:07 PM EDT
yum-plugin-verify-1.1.26-10.el6               Wed 28 Jul 2010 02:06:06 PM EDT
yum-plugin-downloadonly-1.1.26-10.el6         Wed 28 Jul 2010 02:06:06 PM EDT
yum-presto-0.6.2-1.el6                        Wed 28 Jul 2010 02:06:03 PM EDT

(In reply to comment #8)
> Hope this works..

Yes thanks a lot for the information.

Fabio

NM cannot run at S10 because it requires dbus (22), which itself requires rsyslog (12).  NM might also implicitly depend on udev-post (26) though Idon't know what udev-post actually does.  Historically NM depended on HAL (26) but we havent' used HAL since F11, so the lowest # dep we need is 22 (dbus).

Besides; in the grand scheme of things, requiring some sort of "networking" is not what you want.  You want to require the resources that you *actually need*, not just "some network connection is up".  For example, if the service is bound to a specific IP address and not to 0.0.0.0, it really shouldn't be running until some interface has that IP address.  That's clearly not RHEL6 material, but that's the way most services should be working.  Unfortunately they don't, and thus they require a huge sledhammer where they only need a nail...

We can attempt to move NM to S23 (right after dbus), since it doesn't look like udev-post is anything interesting to NM.

Hi Everett,

we have been able to identify the different issues, and they will be addressed soon.

If you need a temporary workaround, please let me know and I'll gladly write down a quick set of steps to get everything back to work.

Sure...you can provide the temp fix.  Is there a file somewhere that maps the link levels, etc?

Hi Everett,

(In reply to comment #23)
> Sure...you can provide the temp fix.  Is there a file somewhere that maps the
> link levels, etc?    

as a temporary workaround, please remove NetworkManager package (rpm -e NetworkManager), chkconfig cman off, chkconfig --del cman, chkconfig --add cman, chkconfig cman on. This should re-establish /etc/rc.d/rc3.d/S21cman.

Once all affected packages will be updated (lvm2-cluster, chkconfig) the problem will disappear permanently and NetworkManager can be reinstalled (but disabled since corosync/cman cannot operate with NM active).

There are also other options but this is the cleanest I can suggest till fixed packages will be availble.

One of the other options is to change /etc/rc.d/init.d/clvmd script and add cman to Required-Starts and Required-Stops: entries to look like:

Required-Starts: $localfs cman
Required-Stops: $localfs cman

chkconfig --del clvmd
chkconfig --add clvmd
chkconfig clvmd on

that should establish clvmd at S25clvmd in /etc/rc.d/rc3.d right after cman.

With this latest solution you will get a notice that init.d/clvmd will be saved when upgrading the package that contains the same change.

Fabio

Where are the S init level links for cman/clvmd? I don't see them in any of the /etc/rc.[0123456].d dirs, I only see the K levels.

lrwxrwxrwx. 1 root root 14 Aug 25 16:47 S21cman -> ../init.d/cman
lrwxrwxrwx. 1 root root 18 Aug 25 16:59 S22cmirrord -> ../init.d/cmirrord
lrwxrwxrwx. 1 root root 15 Aug 25 16:45 S24clvmd -> ../init.d/clvmd

lrwxrwxrwx. 1 root root 15 Aug 25 16:57 K76clvmd -> ../init.d/clvmd
lrwxrwxrwx. 1 root root 18 Aug 18 11:14 K78cmirrord -> ../init.d/cmirrord
lrwxrwxrwx. 1 root root 14 Aug 25 16:57 K79cman -> ../init.d/cman


Fix verified in the latest rpms.
2.6.32-59.1.el6.x86_64

lvm2-2.02.72-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010
lvm2-libs-2.02.72-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010
lvm2-cluster-2.02.72-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010
udev-147-2.22.el6    BUILT: Fri Jul 23 07:21:33 CDT 2010
device-mapper-1.02.53-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010
device-mapper-libs-1.02.53-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010
device-mapper-event-1.02.53-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010
device-mapper-event-libs-1.02.53-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010
cmirror-2.02.72-8.el6    BUILT: Wed Aug 18 10:41:52 CDT 2010

Red Hat Enterprise Linux 6.0 is now available and should resolve
the problem described in this bug report. This report is therefore being closed
with a resolution of CURRENTRELEASE. You may reopen this bug report if the
solution does not work for you.