cluberti
Content Type
Profiles
Forums
Events
Posts posted by cluberti
-
-
If you've got an MSI already you can create an MST via ORCA, but that's somewhat difficult. There are other MSI-related tools out there of varying degrees of cost that can also assist in reading in an MSI and helping create an MST, although if you don't have an MSI and the .exe doesn't have "silent" install options, your only real option is repackaging the installer (MSI, NSIS, InnoSetup, etc).
0 -
Or something you've done to both of your machines?
For what it's worth, all of my Vista SP1 (2) and SP2 (3) lab machines installed the update fine today.0 -
I honestly don't know, but until we actually see WUSA log of a manual installation of a problem MSU failing, there's nothing more we can do to figure it out. Complaining it's not working without providing data means you're not as likely to get useful help
- hence why I've asked for the WUSA /log logs of a failing install.0 -
The old skin was IPB2 specific, and for various reasons it wasn't feasible to get it working with IPB3. The Orange and Blue theme we had for awhile was a bit jarring, but some people liked it. This current theme is a bit blockier, but it does return us back to the "old" MSFN color scheme which (in my opinion) is far easier on the eyes.
You'll get used to this one, just like the original MSFN theme (there were detractors to that theme too, for what it's worth, who said that theme "sucked" too). Change with a board upgrade as large as ours is unfortunate, but inevitable. Please don't take this as coming across as callous - I understand that there are issues with this skin compared to previous ones, but this one does actually work quite well once you get used to it. Give it a chance.
0 -
When I opened the forums today I noticed completely different theme. It's horrible. How do I get back the the previous one?
edit: The link to mark the forums read is gone. There is no theme I could select anywhere. The options are chaotic. Seriously, if it works, don't fix it.
While I won't comment on the quality of the skin at any point in this forum's history back to 2001, I will say that there are links at the bottom left for marking a forum read, etc.
0 -
Well, I cannot really answer that question - If you think you'll ever need a user from one campus domain to be able to access resources from another, then it makes sense. I'm still in favor of adding just a new site and subnet to the original domain (you could still rename that if necessary), but you could also simply create a new single-domain forest for the new campus, sure.
0 -
Why create a forest when a second DC and a second site should be all that's needed? You only need a new branch in a forest when you want administrative or security separation - if they're going to be on the same domain, if it's just a physical thing (DFS, local file / DC servers, etc), then just create a new site and subnet in AD, make sure to use the Default IP Site Link, and you should be good to go. I'm assuming you already have routes between the two sites so that the one site can already talk to the other - if not, you'll have to make sure that works as well (you'd have to do this one in a multi-domain forest scenario as well).
If you *need* to create new forests, what you will need to do (for ease of migration) is rename the "CAMPUS" domain to "DIST999" (if it's the only domain, then it should be the forest root domain already), create two new child domains (CAMPUSNORTH and CAMPUSSOUTH), and then migrate the user objects from DIST999 to the new CAMPUS* domain(s) that you need to move these to. Note you will need at least one DC for each domain (so if you have one now, you will need at least three to do this - if going with redundancy, you should currently have at least two, and you would need at least 6). Also note that with a multi-domain forest, you have DNS delegation or conditional forwarding consideration designs to address as well. To reiterate, doing this is *not* as easy as the wizard might portend when you're considering it, and you *must* plan this out properly if you don't want to be troubleshooting a nightmare afterwards. If you do decide to go the multi-domain forest route, consider strongly making a Disk2VHD (or other) VM copy of the current DC/domain, and build up a copy in your lab on an isolated VM network; this way you can try out anything you'd like and blow it away if you screw it up, etc. I would not be doing any changes to a production single-domain forest without doing a VM test of all proposed design changes, and I strongly suggest you follow this advice as well. Measure twice, cut once, so to speak.
You probably want to read these, at least, if you do decide to go the multi-domain forest route:
FSMO placement and optimization on Active Directory domain controllers
Active Directory Best practices
DNS design options in a multi-domain forest
You will also want to make sure your current domain is "healthy" before you do ANY changes, even adding a second site, using BPA.
0 -
What are the updates exiting with as an error code, and what do the logs say when you install manually with WUSA (wusa.exe <package>.msu /log:<log file path and name>.log)?
0 -
I would recommend 1.2, yes. I've seen debugdiag do this in the past on Win7 systems, but I've never figured out what was in common amongst the systems where it did this. Looks like you might need the 1.2 beta, indeed.
0 -
Are you perhaps running into this?
0 -
Would probably be worthwhile to download/install DebugDiag and use that for memory leak detection (it has a memory/handle leak "rule" that you can run and take manual dumps of the process over time, and then use DebugDiag to analyze those same dumps for leak patterns).
0 -
If it's picking up volume_KMS, then you're on a network with a KMS server (and have a gvlk key installed on your desktop OS that was activated). However, you can "unlicense" with slmgr -upk and get your .xrm-ms license files to activate instead.
0 -
No problem - good to hear you figured out which was causing it.
0 -
-
-
Error 1722 is "RPC_S_SERVER_UNAVAILABLE", so it's unable to make an RPC connection - however, it's probably a local RPC (LRPC) call given it's trying to run "RegisterForNotification" in winlogon itself. I wonder if your network card driver is at fault here, or if something 3rd party on the machine that would normally be running in winlogon is failing to register itself with callback events properly.
Does running the following 2 commands (and rebooting) change this at all?
netsh int tcp set global rss=disabled
netsh int tcp set global autotuning=disabledIf that doesn't help, your next best bet is autoruns, disable anything not absolutely critical for boot (including stuff in the winlogon tab) and see if that changes the behavior any.
0 -
.NET doesn't (by default) start or run anything by default, so if you get an exception I doubt it's the framework that's broken. In fact, that callstack looks like a .NET application is calling into native code (poorly) and crashing. I'd guess you had or have something .NET-related that's trying, albeit in vain, to run at system startup. Might be worth using autoruns to get rid of anything not absolutely system-critical from your startup (services and programs) and start putting things back a few at a time until you can reproduce the error (and thus know where it's coming from).
0 -
That's good
.0 -
Assuming this is the file you're downloading, it must be your link or some hop between you and the site. I'm getting ~2.3MB/sec downloading this.
0 -
XPerf isn't necessarily meant to give you the ultimate answer, it's meant to show you where to further troubleshoot (unless the problem is so obvious that it makes sense from the xperf data alone). In this case, 40 second delays in winlogon init usually means policy issues - enable winlogon logging to see what it is (or isn't) doing during that time, because that is a long time:
Fixing Group Policy problems by using log files
How to enable user environment debug logging in retail builds of Windows
How to read a userenv log part 1
0 -
Once installed but before you register it, what does the output of slmgr /dlv say?
0 -
Yeah, if you don't see any spikes right before it goes off the reservation, then it's not your instance that's causing it.
0 -
The time won't synchronize if your clock is farther than a set amount of seconds wrong compared to the time source (skew), so if the skew is more than (for example) 5 minutes, it may not synch and throw that error. If your time really is skewing that far over the course of just 2 days, it might be worth looking at your CMOS battery in the BIOS to see what time the BIOS says it is on reboot, as Windows uses this (partially) to determine and try to keep time.
0 -
I'd get a perfmon started at 1 second intervals, and watch things like server queues, HDD queues, etc. Assuming you don't see any difference in a working vs slow timeframe, you can be pretty sure the problem is with someone else's VPS chewing up the parent server's resources (especially if it's on what would appear to be a consistent and repeatable timeframe).
0
For what it's worth, all of my Vista SP1 (2) and SP2 (3) lab machines installed the update fine today.
- hence why I've asked for the WUSA /log logs of a failing install.
Windows updates failing
in Windows Vista
Posted
To run an .MSU and get a log, you run the following command:
wusa.exe <package>.msu /log:<log file path and name>.log