PE2650/RHAS 2.1 Cluster/CA ARCserve Questions...

Burger, Matthew R maburger at indiana.edu
Wed Mar 19 11:54:00 CST 2003


Hello,

I am having several issues with getting a project going that I have
started.  I would appreciate any help that anyone on this list could
offer.  I have worked with Linux for several years now and feel pretty
comfortable with it, but I am having several major problems with getting
this setup going and stable.  We have two PE2650's tied to a PowerVault
220 disk array.  Also, attached to the first node of the cluster is a
PowerVault 132T for backup.  That is the setup.... and now my problems.

One question that I have is, when setting up services under the Cluster
Manager utility... If I am wanting to use this Samba setup as a PDC, do
I need to recreate all of the lines of code in each conf file that makes
it work as a PDC?  It appears that I have to create a separate service
or instance for each share that I am wanting to make, is that correct?
And, if so, to make this cluster work as a PDC... Do I have to specify
that configuration over and over?  Again, it may be that I am just not
understanding how to put this all together.

The second question that I have is about getting ARCserveIT for Linux
(7) to work correctly.  I actually had no problems getting it to
install, but I am having one heck of a time getting it to work with the
PV132T and also I cannot get the web interface to work.  I was wondering
if it had to be using a specific version of Java for it to display
correctly?  Everything that I have read mentions nothing of the sort.
The daemon starts up just fine and everything seems fine.  The web page
even loads the Java applet, but it is all scrambled and broken.  None of
the individual parts of the applet load correctly.  

We are also having several issues with the Broadcom NICs.  I have
updated everything, but still have issues there too.  We are using the
two onboard NICs in a team and also have one PCI Broadcom adapter.  I am
having very strange issues with them, though.  If I specify the bcm5700
module in modules.conf... each time eth2 (or the PCI NIC) fails to load.
If I use the tg3 module (update from RedHat) they all seem to work, but
teaming them together causes a lot of issues.  So, what I finally had to
do was to just use the two onboard NICs in the team and use the PCI NIC
as a heartbeat for the cluster.  It seems to work, but I have to use the
bcm5700 module for the two onboard NICS and then use the tg3 for the PCI
one.  I am just wondering if this will cause any problems?

Again, sorry for the long email, but I am having several problems.  Any
help or input would be greatly appreciated.  




More information about the Linux-PowerEdge mailing list