VS v7 - dat update fails - can't find valid repository

[cottonpants]

Hi - I have seen a similar post to this but it doesn't apply to my situation.(the other post was due to dual homed PC's)

I have some machines (both servers and workstations) that have VirusScan v7 installed and update their DATs ok for a while from our internal FTP server. But then _something_ happens (don't know what) and they refuse to update, giving a 'Can't find valid repository' error. Path hasn't changed and is valid, all the various req. files are present (sitestat.xml, catalog.z etc), and other machines are still able to upgrade OK from same location.

Drives me bonkers. I've had to resort to updating using superdats, but my list of affected machines is getting longer, and my days are not.

Anyone seen this before???
Even better, do you know how to fix???? (We would rather not use UNC)
I suspect something is screwy with their FTP engine in general (not just with Dual NICs).

Thanks in advance.

 

[grshpr]

Did you ever get a solution to this? I am having similar issues with 2 of my repositories. The first is on a windows2000 server, the second is on a windows 2003 server. I have 4 other repositories that work fine. 4.51 clients have no issues with getting updates from the same ftp site.

 

[nevets2001uk]

I have similar issues. We have about 80 workstations that use McAfee 7.0 and that can update from the Internet. However we tried creating a repositary as not all users should have Internet access and also we want to manage the updates more easily.

How do I go about creating a repoistary? I get the same message as in the first post. However I don't have any files such as sitestat.xml, catalog.z in the repoistary.

Have we missed something?

Thanks,

Steve

 

[grshpr]

What we have done is created a series of internal FTP sites, one at each of the WAN sites we have. Do you have an internal FTP server or UNC path that you are having your non-internet clients pull from? If not you will need to set one up first. If you are using an ePo server, you can then have it replicate the Master repository to the secondary, if not, you will have to pull the files in manually to the FTP site. The files you will need for McAfee 7 are found in the CommonUpdater forlder on McAfee's FTP site.

 

[nevets2001uk]

OK thanks. I have a FTP server which we should be able to use. I will get the files later and give it a go.

Cheers.

 

[grshpr]

I got a resolution fom Network Associates about the issue of McAfee Enterprise 7 clients not pulling from the internal FTP sites I have. It seems that ther is some sort of security issue on my server that is causing this. I found that kind of hard to believe since 4.51 clients have no problem connecting to the same FTP site and pulling the updates, and that I have 3 other repositories at WAN sites that have no issues either. But anyway, here is the work around.

1. In ePo, edit the configuration for the e-policy agent.
2. On the Proxy tab, check the "Use client IE Settings"
3. Under the FTP proxy settings, fill in the username and password that you are using to deploy installs to your clients.

Now you can either wait for the next agent to server communication or send an egent wakeup call to get the policy changes. Since setting this up, I have had no problems with getting clients to access the internal FTP site.

Hope this helps anyone with problems they have!

 

[nevets2001uk]

I'm having trouble finding those files. What is the link for the FTP site?

 

[cottonpants]

There is something funky with the way v7's FTP engine works - it really doesn't like any sort of proxy to be in the way - I never had problems with previous versions FTP'ing, but v7 just pukes. Doesn't matter what proxy settings I configure. No joy. So we are avoiding using FTP and going with UNC wherever possible.

In regards to my original post, I have found that ever since I got my Source Repository successfully pulling the DAT's from McAfee (using HTTP), and replicating to my internal repository sites, the update problems have gone away. Can't find any actual explanation to what the problem was, but I'm just happy it's working.