@scottalanmiller said in Persistent malware in Edge:
@BBigford said in Persistent malware in Edge:
@Dashrender said in Persistent malware in Edge:
UWP apps are like Android apps or iPhone apps. you uninstall the whole thing, then reinstall it.
What about organizations that allow Edge, but not the Windows Store? You wouldn't be able to reinstall Edge in that case.
Perfect
lmao, perfect trickery. Tell someone to uninstall, don't allow reinstall.
There was a certain term/concept that someone used some time ago about SQL transactions during a power outage. I'm trying to remember what that was. It was basically protection during a power outage, when you have transactions sitting in memory, waiting to be written.
If you have transactions sitting in memory waiting to be committed to the database, what is the protection that those won't be lost during a power outage, possibly leading to DB corruption?
...and then you have companies that cluster servers, with each server having RAID configured. Sacrificing some usable storage there.
I have no idea where to start with this one because it is something I've saw in many environments. Here's a couple scenarios:
*Organizer cancels an event, but attendees do not get the cancellation.
*Attendee sends a proposed time to organizer, but organizer does not get the notification.
The only partial work around I've found is to completely delete the event series both on the organizer's calendar and the attendees. Then send out new invites. This works about 80% of the time, but doesn't always stick. It's tedious and not a long term solution.
I've witnessed this on Exchange 2007, 2010, and 2013.
Any idea? I can't really just poke at it since it is something that the client CEO is upset about. So I have to have a few real solutions to bring to the table so I'm not just wasting time.
Just wondering if anyone has witnessed this, maybe a cause and solution?
@scottalanmiller said in Should We Ever Talk About JBODs:
JBOD is an abbreviation for "Just a Bunch Of Disks" and is meant to refer to, well, no one is exactly sure when to use it properly. People often use it to mean "not RAID" or not "hardware RAID" or "not yet RAIDed" but they forget that spanning, RAIN or whatever would make something no longer a JBOD. And some people mean it only to refer to hardware, but RAID is software, as is RAIN. Everything is JBOD at some point and basically nothing is by the time that it is in use.
I'm not clear when we can ever usefully use the term where it doesn't prompt a long and confusing explanation to clear up what is actually intended. I have a feeling that this is a term that simply should not exist as there is just no time that using it is useful.
Does anyone have any idea of a time when it would actually be important to have this term and, if so, how would we use it without further explanation?
Storage pools on Windows Server. You can add disks as needed, and over allocate with thin provisioning. I consider that a straight up JBOD because of over allocation and adding disks as needed. Not much to it other than grabbing disks out of primordial and assigning to a VHD. Also consider it JBOD because though you can choose parity, it is not inherent.
@fuznutz04 said in I know you're not crazy...:
We started testing MS Teams for our group chats. I'm not overly impressed yet, but I haven't really put it through the paces yet. One thing I really like about teams is the continuous chat log.
If you didn't know, Slack had actually consulted Microsoft on that project. It was a flurry on forums when it was coming around. Though a Slack dev/project manager, can't remember, threw a low blow in a side interview and said "we could give them an exact design of Slack and they would still release a poor product."
That's funny, was just talking about this quite a bit today with @scottalanmiller. It really comes down to licensing. I would rather license one VM and run containers if I have licensing issues, but if the licensing is no sweat, then I would rather scale out VMs as needed to separate applications. There would be the question, "if you're having licensing limitations then, why couldn't you just use XenServer and have everything Linux?"
Aside from licensing, it really is just preference. By isolating resources, both do the same thing. One could argue that containers are more light weight, but not enough to notice with today's technology. I prefer VMs over containers though, if licensing doesn't play a role.
Also adding, if I wanted to squeeze every last little bit of performance out of a machine, I would consider containers.
I've been so torn on my primary desktop between Fedora, Mint, Manjaro (random, I know), Korora, and -surprisingly- Zorin. Each has it's ups and downs with their package managers, amount of repositories (free and paid aside), and of course look and navigation. Solus was in there for a while but decided against it.
Thoughts on your own favorite daily driver out of box? What do you like and don't like?
I was working with NetScaler within Citrix (configured as a separate VM), and was thinking about VMware. It has been since a 5.5 upgrade to 6 that I have worked with Horizon View. I know it used to be that if you wanted a load balancer, you needed something like F5, NetScaler, etc, as a third party solution. Since VMware didn't have anything native, like Citrix.
Is that still the case? I know a separate appliance is the preferred method, and DNS round robin can be configured but not preferred. Unsure of what is native now and web searching isn't revealing much of anything.
I'm pressed to get this done today and I'm struggling with something very simple. The issue I'm running into, is running a .msi while bypassing UAC. I've read lots of Kaseya threads (what I'm pushing from), and still coming up short.
What I'm trying to do is a simple 3 step process. Make a directory, distribute the file, install the file. It's a very small endpoint protection agent. Everything up to installing works fine. If I try and run it, nothing happens. So I went locally and noticed that installing it from C:\CarbonBlack prompted for UAC. So I went through this, while also adding a two minute pause, but it still didn't work. The pause is pretty long compared to the install, which takes about 30 seconds.
Here's a screen shot of the procedure I currently have scripted. The end is cut off, but simply says "All Operating Systems", "Halt on Fail".
@stess said in Scripting install - help:
Not an expert here, but does the .msi installer has /q or /silent install parameter?
Personally, I use PDQ Deploy from AdminArsenal when install applications. With correct credential and parameter it bypass UAC.
I've tried with an without /q and /silent. Same issues.
PDQ Deploy won't work, because of how fragmented their network is. I can only reach it via Kaseya because their DNS/DHCP is partially handled by their ISP. Also some of the workstations are integrated, some aren't. It's a mess.
@MattSpeller said in Scripting install - help:
@BBigford said in Scripting install - help:
@stess said in Scripting install - help:
Not an expert here, but does the .msi installer has /q or /silent install parameter?
Personally, I use PDQ Deploy from AdminArsenal when install applications. With correct credential and parameter it bypass UAC.
DNS/DHCP is partially handled by their ISP
What the....
Yep, I know @MattSpeller. I had to ask someone why stuff looked super weird. When I asked if that's how it was, it was confirmed. That client has a really shitty setup but don't want to spend the money to fix it, even after being shown the value.
Good example of seeing the value... I can't work on their other tickets right now, because I'm working on this one, which is holding up the show due to inefficiencies in the setup.
@MattSpeller said in Scripting install - help:
@BBigford said in Scripting install - help:
@MattSpeller said in Scripting install - help:
@BBigford said in Scripting install - help:
@stess said in Scripting install - help:
Not an expert here, but does the .msi installer has /q or /silent install parameter?
Personally, I use PDQ Deploy from AdminArsenal when install applications. With correct credential and parameter it bypass UAC.
DNS/DHCP is partially handled by their ISP
What the....
Yep, I know @MattSpeller. I had to ask someone why stuff looked super weird. When I asked if that's how it was, it was confirmed. That client has a really shitty setup but don't want to spend the money to fix it, even after being shown the value.
Good example of seeing the value... I can't work on their other tickets right now, because I'm working on this one, which is holding up the show due to inefficiencies in the setup.
How does stuff like that even happen?
I'm going to go hug my nicely setup (if old and creaky) infrastructure now.
We took them on as a client. So whoever setup their stuff in the beginning looked like they were just lazy and careless.
Some stuff is free, some isn't. Just curious if anyone has tried protecting a physical workload with this agent here.
Couldn't find anything about cost, so wasn't sure if it is a free offering. We're currently protecting VMs with Enterprise Edition.
I heard one DBA talking about matching tempdb files to the physical core count of the VM. In essence, 4 tempdb files to 4 cores. Autogrowth being disabled it looks like, and matching the file sizes. When I asked about it, he simply said "because it performs better" but didn't have any explanation.
Anyone care to input while I'm searching online?
Anyone have a guess why journaling in Microsoft Exchange is a user mailbox whereas archiving can be its own database? Seems like journaling would perform better if it were a database and you were trying to return search results. I'm guessing there is a logical reason between it being a user mailbox and a mailbox database...
I've been trying to figure this out. I've deleted and re-created the User Profile Service Application. I've changed Use SharePoint Active Directory Import to Use SharePoint Profile Syncronization. If you don't have Profile Syncronization selected, you can't start User Profile Syncronization Service in System Services I noticed.I've got that all running, but when I go to Alert Me on a list, the user selected (in this case, SPAdmin) doesn't show an email address. If you go to the User Profiles, that user has an email address. If you go into User Properties, 'Work Email' has Replicable as selected. I've also manually run the timers and created a new Syncronization Connection. Here's some screen shots... Any ideas?
Not a new subject, and I know a common consensus is "this is not done well overall."
Ticketing systems seem to be a dime a dozen. I've been using AutoTask most recently. Spiceworks obviously can't handle client isolation, as it wasn't designed for that.
I haven't used ZenDesk, but saw they do have a MSP offering. Also haven't used osDesk but have heard good things and was told they do have client isolation.
If you are working at a MSP, what are you using and why? Just some of the high points.
Got it squared away. They have only one public IP right now, so what ended up happening is they are already going from a pair of Barracuda's in their datacenter, but now they are also going through SendGrid using 2 IPs in case one gets black listed.
They have a 3rd as well, that they'll use strictly for reporting. If they get both blacklisted, they could cut over to the 3rd if need be, while they swap out the other 2 for clean IPs.
I made the outbound deferred messages to email filtering more strict, so they can catch it sooner. Also working on some automation to temporarily ban any abusive accounts for 24 hours if they have too many deferred in a period of time.
As far as going through SendGrid, the recipient sees it comes from SendGrid, but the reverse lookup in the header does show the single public IP that they are using now. Things look a lot better now, but still a little work to do.