TheFourthFerret
August 16th, 2007, 09:22 AM
I am trying to work out some kinks after rolling out NOD32 Enterprise on a school network. I set up the NOD32 infrastructure with 4 administration servers to support close to 800 clients (one main server on the main campus and the three others at remote buildings, which report back to the main one, but pull updates for their mirrors directly from Eset).
I am focusing on the main server, which handles 3 buildings (each building and remote site has it's own subnet).
I have set up config files and installation packages for each server, so the clients pull updates and are administered by their local server. Initially, I had exported installers for script installs for local machines from the local server, but recently exported all packages from the main server for simplicity (yes, all installs now pull down from the main server, but the bandwidth consumed by this isn't a problem and a central install server, and therefore point of troubleshooting, at least until the problem is solved, is handy).
We have created images for all workstations (Norton's Ghost) so all machines with of the same generation and hardware setups have the exact same software environments, and in general, software, installed applications, etc. are the same across the board with all images. We are also running a Novell network.
The problem:
With some workstations, we noticed after running the script, NOD simply didn't install. After ensuring communication with the server, and running the installer directly from the network location (browsing out to the little .exe file and manually running it apart from the script) an hour glass would show for a split second and still nothing would happen. Checking logs, we found the following:
from the "nod32installer.log" file:
{QUOTE-> [2007-07-23 11:39:46.953] Status 1: NOD32 Installer (1.0.15) is preparing to install.
[2007-07-23 11:39:46.953] Status 1: NOD32 Installer is trying to connect to 'WESTVIEW-UDS1:2224'.
[2007-07-23 11:39:46.968] Status 1: NOD32 Installer is sending installation request.
[2007-07-23 11:39:46.968] Status 20000: NOD32 Installer was told to quit by the server 'WESTVIEW-UDS1:2224'. <-QUOTE}
And the "nod32ra.log" file:
{QUOTE->
[2007-08-02 15:14:50.697] V5 [46b22d2adbf7] <SESSION_INFO> Kernel connection from 172.16.1.199:42500 accepted
[2007-08-02 15:14:50.697] V5 [46b22d2adbf7] <SESSION_INFO> Kernel connection from 172.16.1.199:42500 closed (took 0ms, name 'WHS31101V', version '2433 (20070802)') <-QUOTE}
Note: These entries are not from the exact same install, or lack there of, but each time we have had the problem, we'd see the "0ms" entry in the ra log.
We haven't found any pattern in terms of which workstations are affected. Some install perfectly fine, and one right next to those with the same hardware and software image, won't pull it down.
Also, we have found that "pushing" the install from the server, using the exact same package works every time on the problematic workstations. We have been using this as a work around, but for the long run, we need to get the script install working.
Extra info:
The main server is also running WSUS, but I haven't found any conflicts between that and NOD32. Could this be it? Wouldn't all installs fail if there was a conflict?
Command switches for the install packages are as follows:
{QUOTE-> /SILENTMODE /INSTMFC /REBOOT /SHOWRESTART <-QUOTE}
I have not run into any errors or found any conflicts with port 2224 on the server either.
Any ideas?
Thanks in advance!
I am focusing on the main server, which handles 3 buildings (each building and remote site has it's own subnet).
I have set up config files and installation packages for each server, so the clients pull updates and are administered by their local server. Initially, I had exported installers for script installs for local machines from the local server, but recently exported all packages from the main server for simplicity (yes, all installs now pull down from the main server, but the bandwidth consumed by this isn't a problem and a central install server, and therefore point of troubleshooting, at least until the problem is solved, is handy).
We have created images for all workstations (Norton's Ghost) so all machines with of the same generation and hardware setups have the exact same software environments, and in general, software, installed applications, etc. are the same across the board with all images. We are also running a Novell network.
The problem:
With some workstations, we noticed after running the script, NOD simply didn't install. After ensuring communication with the server, and running the installer directly from the network location (browsing out to the little .exe file and manually running it apart from the script) an hour glass would show for a split second and still nothing would happen. Checking logs, we found the following:
from the "nod32installer.log" file:
{QUOTE-> [2007-07-23 11:39:46.953] Status 1: NOD32 Installer (1.0.15) is preparing to install.
[2007-07-23 11:39:46.953] Status 1: NOD32 Installer is trying to connect to 'WESTVIEW-UDS1:2224'.
[2007-07-23 11:39:46.968] Status 1: NOD32 Installer is sending installation request.
[2007-07-23 11:39:46.968] Status 20000: NOD32 Installer was told to quit by the server 'WESTVIEW-UDS1:2224'. <-QUOTE}
And the "nod32ra.log" file:
{QUOTE->
[2007-08-02 15:14:50.697] V5 [46b22d2adbf7] <SESSION_INFO> Kernel connection from 172.16.1.199:42500 accepted
[2007-08-02 15:14:50.697] V5 [46b22d2adbf7] <SESSION_INFO> Kernel connection from 172.16.1.199:42500 closed (took 0ms, name 'WHS31101V', version '2433 (20070802)') <-QUOTE}
Note: These entries are not from the exact same install, or lack there of, but each time we have had the problem, we'd see the "0ms" entry in the ra log.
We haven't found any pattern in terms of which workstations are affected. Some install perfectly fine, and one right next to those with the same hardware and software image, won't pull it down.
Also, we have found that "pushing" the install from the server, using the exact same package works every time on the problematic workstations. We have been using this as a work around, but for the long run, we need to get the script install working.
Extra info:
The main server is also running WSUS, but I haven't found any conflicts between that and NOD32. Could this be it? Wouldn't all installs fail if there was a conflict?
Command switches for the install packages are as follows:
{QUOTE-> /SILENTMODE /INSTMFC /REBOOT /SHOWRESTART <-QUOTE}
I have not run into any errors or found any conflicts with port 2224 on the server either.
Any ideas?
Thanks in advance!