SAV 10.1.6 Corporate nightmare...
...last Thursday, a senior security analyst (not my term for him!) came in remotely and upgraded our SAV server remotely from v.8.1 to v.10.1.6, an upgrade we ultimately needed. Problem was this was done with no testing whatsoever in our environment (400 desktops, 150 laptops -- mostly remote, and 30 servers) and with no apparent backout plan. Furthermore, this so-called senior analyst had no idea of the current policies in our environment. This is a desktop environment of mixed XP Pro and W2K units, mostly the latter.
Right off the bat we had problems. The upgrade took place during business hours, late in the afternoon. The next morning, users started reporting problems their AV wasn't updating. It soon became obvious they were still on 8.1 and that a reboot was necessary to complete the upgrade to 10.1.6. When users rebooted to complete the upgrade, then the fun really started. About 60% of the users rebooted and the new version ran OK, reporting to the SAV server. The other 40%, well, I could probably write a book on all the problems we've had.
We've lost close to 20 W2K units right off the bat to some kind of video driver corruption. This driver corruption manifested itself in 2 ways: one, booting to a black screen with some speckled colored lines, and two, a "rolling reboot" (continual reboot). In both cases, we are able to boot into safe mode. The first corruption was easily fixed by uninstalling the video drivers (one an Intel 82815, the other an SMS driver), then reinstalling the driver only (not in the setup mode) from our software server. The rolling reboots are another story. We've tried removing drivers and reinstalling, then they might boot up fine, then they go back into the rolling reboot next time the unit's rebooted. Other times, it doesn't matter if we reinstall the vid driver. I've tried the chipset driver on a couple of units to no avail. I've tried numerous SAV uninstalls (tomorrow I'm going to try uninstalling with a forklift...I'll let you know how that one goes!) and reinstalls. Today, we had a W2K unit that had been rebooted successfully and happily ran SAV 10.1.6 for several days suddenly slip into the rolling reboot mode. I re-imaged the machine after a data and profile backup, it ran fine, went thru reinstalling printers (we've got dozens!), made the guy a local admin so he could do some stuff, only to have it slide into the rolling reboot when we restarted the PC. This is bizarre. This is taking place in all different dept's, with ass't app loads. Other W2K units run fine with the same app loads. We've had no hardware problems from these machines previously.
On top of this mystery W2K issue, the 10.1.6 push just didn't fully take on about 40% of working units. SAV services set to "Automatic" don't start on bootup, but if I go into console, I can start them with no problem for the user. Other units are still running 8.1, but that service may or may not be started.
In the meantime, day-to-day support has gone to hell in a handbasket, with users having no idea what we're going thru as staff and expecting normal levels of support.
Anybody here been thru or seen anything like this? This is about the most bizarre network/computer problem I've ever seen.
Ugh, life in Hell starts up again in 3 hours.
**sigh**
Do not use NoNav or add/remove programs to uninstall v10.1.6
I recently migrated my SAV CE from v10.0.2 to v10.1.6. When I performed the migration on my servers everything looked ok at first but then I noticed that on my primary server SAV kept kicking off installs one after another. At one point I saw at least six installation processes running at the same time. It was like a Chinese fire drill trying to cancel them all. This would happen any time I tried to open the SAV GUI or touch anything else that involved SAV. I called Symantec support and the engineer had me put the disk into the cd drive on the server and try to complete the install, it reported that the install was already completed. Yet SAV continued to try to install itself all by itself over and over again. The engineer then told me that I would need to uninstall and reinstall the product because something must have been corrupt with the initial install. First I had to move all my connected clients from the primary server to the secondary server and then uninstall the primary server. When I tried to uninstall through add/remove programs I got an error message telling me that there was an incomplete install and that I needed to let that finish before I could uninstall. Reboots didn't help this either. The engineer then sent me the NoNav utiity to run on the server to remove the program. This is when the real horror started. In subsequent phone conversations that I've had with Symantec engineers and an escalation manager it appears there is a fault in 10.1.6 that, when installed, associates itself with a lot of things that it has no business doing. As a result the NoNav utility wiped out most of my MS Exchange files and executables along with an as yet unknown number of other things. My NSI Double Take replication software no longer works, I have screwed up print queues and a whole range of problems. The consensus of opinion from the Symantec engineers and myself is that the server has to be rebuilt from scratch. They told me that it was not NoNav's fault because add/remove programs would have done the same thing.
To sum this up, according to Symantec, there is a known fault with the installer. They say it only effects the server installs. If you uninstall the product v10.1.6 using either the NoNav utility or add/remove programs it will remove a great many files that it shouldn't such as program executables belonging to other software and the operating system.
If you need to uninstall this product you'll need to do it manually with registry edits and manual file removal otherwise you're likely to have huge problems as I am having now.
