This message was deleted.
# _general
s
This message was deleted.
k
I usually configure GPOs that would install the updates from a WSUS server and then do a staggered reboot, or reboot the serveres on different days in the week
l
Radical process Arthur, I tend to recommend you don't. I go thought he pain to maintain the sccm task sequence with the latest versions of everything and use osdbuilser to apply windows patches offline to a wim. I then rebuild the image on a regular basis and UAT it. It gives you much more stability and time to address fixes with all the latest versions of software.
a
Hi Leee, Are you sure for such process for Infrastructure Servers? DDC, SF, PVS, Director? I didn’t ask for update proces for golden images. Cheers Artur
l
Aaahhh okay. I see. Apologies Arthur. For a major version upgrade I also rebuild these, if possible from a task sequence also. Otherwise as @Kasper Johansen but staggered at different times so they reboot one after the other. If it's been designed in a H/A fashion then this should work. Always worth checking. Happy to dig into this with you.
a
Thanks!
👍 1
n
We use SCCM to patch our infrastructure servers. Knock on wood, it has been stable.
j
Pick your poison. Just make sure you test. Also doing rolling patching/rebooting.
t
Rebuild VMs and reprovision is what we always did, since we had no persistent desktops. Did this roughly monthly unless there was a serous secrity issue that needed faster attention. Only think pushed to existing VMs were anti-virus signature updates.
👍 2
u
For my guys it's all to the customer's tools for Managed Services Contracts. Where there's no SCCM for servers (yeah, it is the case) they use a tool called BatchPatch and use two lists of servers for week1 and week2 patching. The tool is super great but still, it requires manual intervention. They're migrating now towards changing that to an a patching feature that exists in the standard monitoring tool they use. That would automate would trigger alerts in case servers don't recover after reboots. So automated patching and active monitoring is what you should aim for.
n
WSUS with manual install...patch DR site and test one weekend, do prod the following weekend. Never had an issue with a Windows Update breaking a Delivery Controller or StoreFront server (yet) knocks on wood
j
+1 for BatchPatch. Super simple tool and cheap
c
We also used BatchPatch, great tool for patching and other basic automation tasks.
b
If you are using SCVMM because it is a hyper-v environment, you can technically use it to patch the citrix infrastructure as well. I have SCVMM scripts on my github for maintaining patch baselines for infrastructure servers.
l
I shudder when people tell me they use SCVMM its IMO the worst product Microsoft have.
⤴️ 3
🤣 1
b
Yea, I make a lot of money from it 🙂
Due to it's complexity, training people on it takes a lot of billable time.
c
I am using a combination of an update script that patches the master then runs patch my pc along with some of @Eric evergreen apps.
Just looked at BatchPatch seems stupid easy. Might try it.
c
BatchPatch is very good for how simple and cheap it is
a
Are you using any tools to verify if all is up and fully working after patching
m
BatchPatch and ABCUpdate