|
Any help would be greatly appreciated. I have been reading all the docs I
can get my hands on and I can't seem to figure this out.
I am trying to get UltraMonkey running under the "Simple Load Balanced
Service" topology as a single virtual service. Here are the steps I have
taken:
- installed the OS clean (RedHat 6.2)
- removed piranha
- added the drivers for the second NIC - now working
- installed the VA enh. RH 2.2.14 UltraMonkey source code
- compiled and installed the 2.2.14-VA.2.1.um.3 kernel
- installed the RPMs for UltraMonkey and followed the installation notes
- fixed the problem that developed with the NIC (kernel update overwrote the
driver)
So far, it appears that I have everything in place and operating as it
should. The problem I am having is that it is not answering the VIP. I
can't seem to get this to pass the traffic from the external address to the
internal addresses of the web servers. That is odd since I can actually
load web pages from the two "Real Servers" from the UltraMonkey server. I
don't understand why I can't get 10.10.10.248:8069 to answer. I am noticing
that nothing is 'listening' on 8069 when I do a netstat. What gives???
The setup is like this:
UltraMonkey Server:
NIC #1 (eth0) - 3COM 3c905c - 10.10.10.248 <- this is on our LAN.
NIC #2 (eth1) - Linksys LNE100TX - 192.168.0.1
Web Server #1 (WinNT 4.0 with Apache) - 192.168.0.3:8069
Web Server #2 (WinNT 4.0 with Apache) - 192.168.0.4:8069
Here's my "ldirector.cf" file:
--------------------------------------------------------------------------
# Global Directives
checktimeout=3
checkinterval=1
# Virtaul Server for HTTP
virtual=10.10.10.248:8069
fallback=127.0.0.1:8069
real=192.168.0.3:8069 masq
real=192.168.0.4:8069 masq
service=http
request="images/dist.html"
receive="Runlist"
scheduler=rr
persistent=600
protocol=tcp
--------------------------------------------------------------------------
Here's the output of "ipvsadm -L -n":
--------------------------------------------------------------------------
IP Virtual Server version 0.9.16 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 10.10.10.248:8069 rr persistent 600
-> 192.168.0.4:8069 Masq 1 0 0
-> 192.168.0.3:8069 Masq 1 0 0
--------------------------------------------------------------------------
The ipchains and the ldirectord services are running and start up when the
server starts. I did notice that when ldirectord starts it initially
reports as [FAILED] and then switches to [ OK ] as the system boots.
Stopping and starting the service after the system is running does not
exhibit the same behavior.
Any ideas? I don't know what I am missing here, but it must be pretty
small.
Jeff
Jeffery Smith - Sr. System Administrator - jsmith@xxxxxxxxxxxx
Fact City - www.factcity.com
245 Winter Street, Waltham, MA 02451
Voice (781) 684-0101 - Fax (781) 684-8401
-----Original Message-----
From: Gregory J Pryzby [mailto:gpryzby@xxxxxxxxxxx]On Behalf Of gregory
j pryzby
Sent: Tuesday, April 03, 2001 1:59 PM
To: Smith, Jeffery
Subject: Re: [Ultramonkey-devel] VA enhanced RedHat 6.2.3 with
2.2.18pre11 -va2.0 kernel?
Don't give up!
I would try to use 6.2.3 as it is tested and works well.
I do not know of the problem w/ 6.2.4, but if you sen the
error messages I can try to help. Chances are good that you
don't have to install the kernel or initscripts w/ 6.2.4 as
it is very new (less than 2 weeks old I think) and I believe
it has the patches for UM already.
Ask horms@xxxxxxxxxxx if you need the kernel and initscripts
for 6.2.4. Best to look at the versions of the packages and
if the ones on VALE are more recent, use thse first.
WTR the NIC. If you installed the VA Kernel, the NIC should
be recognized on boot. Was it not? Did you update LILO and
reboot (as the readme said)?
Problem is that there are a NUMBER of tulip drivers as the
card kept the same name, but the specs changed leading to
alot of problems w/ getting the tulip to work with the tulip
dirver as somethimes you use a different driver for teh
tulip card! Confusing!
Anyway, I hope that the info above will help.
On Tue, Apr 03, 2001 at 01:55:37PM -0400, Smith, Jeffery wrote:
> Oh, I did originally use the VA distribution from ftp.valinux.com (version
> 6.2.4). It would not allow me to install the initscripts rpm, claiming it
> was incompatible with the kernel. I couldn't install the kernel either,
due
> to incompatibilities with lots of other stuff. I tried to compile the
> source for the kernel and then just switch to it, but that didn't go very
> well...the compile didn't work.
>
> So now I am trying the setup with a regular RedHat 6.2 install. The
kernel
> and all the other RPMs are installed fine. Now I can't get the second NIC
> to work. It is really irritating. The first NIC (eth0) is a 3COM 3c906c
> and it works fine. The second is a Linksys LNE100TX (v 4.0) and I can't
get
> it to work. It uses the 'tulip' driver and comes with the source, etc,
but
> after 'make'ing it I still can't seem to get it to work. Oddly enough,
the
> cards both work fine under the VA enhanced version I had previously
> installed.
>
> So my dilemma's are:
> 1. VA enhanced RedHat 6.2.4 from Web - incompatible with UM
> Initscripts
> 2. RedHat 6.2.4 with UM kernel - incompatible with Linksys LNE100TX
>
> I can't win here...
>
> Jeff
>
> -----Original Message-----
> From: Gregory J Pryzby [mailto:gpryzby@xxxxxxxxxxx]On Behalf Of gregory
> j pryzby
> Sent: Tuesday, April 03, 2001 12:07 PM
> To: Smith, Jeffery
> Subject: Re: [Ultramonkey-devel] VA enhanced RedHat 6.2.3 with
> 2.2.18pre11 -va2.0 kernel?
>
>
> I am confused. I see you sa that you used the VA load then
> the RH load. WHich did you use?
>
> If you got the VA load from ftp.valinux.com, you should be
> fine. The 6.2.4 is the best, but an earlier one should work.
>
> I would d/l all the rpms into their own directory
>
ftp://ultramonkey.sourceforge.net/pub/ultramonkey/ultramonkey-1.0.2beta0/RPM
> S/
>
>
> and then follow the directions on
>
http://ultramonkey.sourceforge.net/ultramonkey-1.0.2beta0/installation/index
> .html
>
> Please send the exact errors if that fails.
>
>
> On Tue, Apr 03, 2001 at 12:01:05PM -0400, Smith, Jeffery wrote:
> > Hi Greg,
> >
> > Thanks for your reply. Yes, I did remove piranha prior to installing
the
> > RPMs for UM, however I am noticing that no matter what I try, I can't
> > install (or force install) the UM 2.2.14 kernel or the initscripts. Is
> this
> > because I installed the VA enhanced RedHat from the VALinux.org site? I
> am
> > going to try adding UM to a plain RedHat 6.2 install.
> >
> > Jeff
> >
> > -----Original Message-----
> > From: Gregory J Pryzby [mailto:gpryzby@xxxxxxxxxxx]On Behalf Of gregory
> > j pryzby
> > Sent: Monday, April 02, 2001 5:43 PM
> > To: Smith, Jeffery
> > Subject: Re: [Ultramonkey-devel] VA enhanced RedHat 6.2.3 with
> > 2.2.18pre11-va2.0 kernel?
> >
> >
> > Did you remove all the piranha stuff first and then install
> > all the rpms for UM?
> >
> >
> > On Mon, Apr 02, 2001 at 05:40:22PM -0400, Smith, Jeffery wrote:
> > > Hi,
> > >
> > > I actually have a couple of questions to anyone who can answer.
First,
> > has
> > > anyone tried or does anyone know if UltraMonkey will run on VA
enhanced
> > > Redhat 6.2.3 with the 2.2.18pre11-va2.0 kernel? I downloaded this
> version
> > > of linux from www.valinux.org and installed it with no problems.
> > > UltraMonkey appears to work fine with it, however I can't seem to get
my
> > VIP
> > > to answer.
> > >
> > > I am trying to get this software to work for the first time. I have
> > looked
> > > everywhere I can think of, but I can't seem to find an answer to this
> > > question. Must the real servers be running the same VA enhanced
RedHat
> > > Linux OS? I have specific software for my web servers that currently
> only
> > > runs under Windows NT. I'd like to have a Linux UltraMonkey setup
that
> > > distributes the load over two NT boxes. I don't know if this is
> possible.
> > >
> > > Below is some information about the configuration I am using. I have
a
> PC
> > > with VA enhanced RedHat and UltraMonkey with two NICS. Eth0 is
> configured
> > > on our LAN under address 10.10.10.248 and Eth1 is on the UltraMonkey
> > segment
> > > using address 192.168.0.1. The two apache servers (WinNT) are at
> > > 192.168.0.3 and 192.168.0.4. Both NT web servers are configured to
> > > specifically answer port 8069 and I can load web pages from each of
them
> > > from the load balancer.
> > >
> > > I followed the installation directions and everything is (as far as I
> can
> > > see) configured according to the docs. It even appears that the
service
> > is
> > > working and running, but I can't get the load balancer to answer at
> > > 10.10.10.248:8069.
> > >
> > > Any insights and assistance would be greatly appreciated!
> > >
> > > Jeff
> > >
> > > Jeffery Smith - Sr. System Administrator - jsmith@xxxxxxxxxxxx
> > > Fact City - www.factcity.com
> > > 245 Winter Street, Waltham, MA 02451
> > > Voice (781) 684-0101 - Fax (781) 684-8401
> > >
> > >
> > >
> >
>
----------------------------------------------------------------------------
> > > -------------------------------
> > > Output of 'ifconfig':
> > >
> > > eth0 Link encap:Ethernet HWaddr 00:03:6D:1E:E8:B7
> > > inet addr:10.10.10.248 Bcast:10.10.10.255
Mask:255.255.255.0
> > > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> > > RX packets:39709 errors:0 dropped:0 overruns:0 frame:0
> > > TX packets:4218 errors:3 dropped:0 overruns:0 carrier:6
> > > collisions:0 txqueuelen:100
> > > Interrupt:10 Base address:0x400
> > >
> > > eth1 Link encap:Ethernet HWaddr 00:C0:4F:01:B2:7E
> > > inet addr:192.168.0.1 Bcast:192.168.0.255
Mask:255.255.255.0
> > > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> > > RX packets:94609 errors:0 dropped:0 overruns:0 frame:0
> > > TX packets:134314 errors:0 dropped:0 overruns:0 carrier:4
> > > collisions:79 txqueuelen:100
> > > Interrupt:11 Base address:0xdc00
> > >
> > > lo Link encap:Local Loopback
> > > inet addr:127.0.0.1 Mask:255.0.0.0
> > > UP LOOPBACK RUNNING MTU:3924 Metric:1
> > > RX packets:258 errors:0 dropped:0 overruns:0 frame:0
> > > TX packets:258 errors:0 dropped:0 overruns:0 carrier:0
> > > collisions:0 txqueuelen:0
> > >
> >
>
----------------------------------------------------------------------------
> > > -------------------------------
> > > Contents of '/etc/ha.d/conf/ldirectord.cf':
> > >
> > > # Global Directives
> > > checktimeout=3
> > > checkinterval=1
> > >
> > > # Virtaul Server for HTTP
> > > virtual=10.10.10.248:8069
> > > # fallback=10.10.15.5:8069
> > > real=192.168.0.3:8069 masq
> > > real=192.168.0.4:8069 masq
> > > service=http
> > > request="images/dist.html"
> > > receive="Runlist"
> > > scheduler=rr
> > > persistent=600
> > > protocol=tcp
> > >
> >
>
----------------------------------------------------------------------------
> > > -------------------------------
> > > Status of 'ipchains':
> > >
> > > Chain input (policy ACCEPT):
> > > Chain forward (policy DENY):
> > > target prot opt source destination
> ports
> > > MASQ all ------ 192.168.0.0/24 0.0.0.0/0 n/a
> > > - all ----l- 0.0.0.0/0 0.0.0.0/0 n/a
> > > Chain output (policy ACCEPT):
> > >
> >
>
----------------------------------------------------------------------------
> > > -------------------------------
> > > Output of 'ipvsadm -L -n':
> > >
> > > IP Virtual Server version 1.0.2 (size=4096)
> > > Prot LocalAddress:Port Scheduler Flags
> > > -> RemoteAddress:Port Forward Weight ActiveConn InActConn
> > > TCP 10.10.10.248:8069 rr persistent 600
> > > -> 192.168.0.4:8069 Masq 1 0 0
> > > -> 192.168.0.3:8069 Masq 1 0 0
> > >
> >
>
----------------------------------------------------------------------------
> > > -------------------------------
> > >
> > >
> >
>
----------------------------------------------------------------------------
> > > -------------------------------
> > >
> > > _______________________________________________
> > > Ultramonkey-devel mailing list
> > > Ultramonkey-devel@xxxxxxxxxxxxxxxxxxxxx
> > > http://lists.sourceforge.net/lists/listinfo/ultramonkey-devel
> >
> > --
> > gregory j pryzby < gjp at valinux dot com >
> > Herbivores ate well because their food didn't ever run!
>
> --
> gregory j pryzby < gjp at valinux dot com >
> Herbivores ate well because their food didn't ever run!
--
gregory j pryzby < gjp at valinux dot com >
Herbivores ate well because their food didn't ever run!
|
