Japanese Dating Sim posted:I have been a little befuddled about the VMWare to Nutanix thing honestly, given that I had (like I said) never even heard of Nutanix, but I'm not in a position to judge. It's a municipal org so it might be one of those things where they have to go out for bids and the VAR pitching Nutanix won? I don't know how those things work for IT infrastructure purchases. Not my problem, I guess! Nutanix has been aggressive bidding on RFPs in my area for state/local/education organizations. I’m guessing the RFP processes don’t usually involve people who have worked with or talked to people who have used Nutanix, or it’s somehow way cheaper. It usually winds up getting ripped out the next RFP cycle.
|
|
#
?
Dec 30, 2020 00:52
|
|
|
|
| # ? Jun 6, 2024 08:23 |
|
|
klosterdev posted:Still hasn't changed a whole bunch, save for maybe MS Going to disagree with the MS comment. That is still very much their culture. I could give a bunch of examples, but go try to open a ticket about anything in Azure or O365 and see how support acts. I don't think I've ever gotten through a ticket successfully, and always passed between 4-5 departments. Hell, they can't even pass tickets between some of them. They ask you to open a new ticket.
|
|
#
?
Dec 30, 2020 01:08
|
|
|
Internet Explorer posted:Hell, they can't even pass tickets between some of them. They ask you to open a new ticket. There's an SVP somewhere doing this to fudge Time To Close numbers .
|
|
#
?
Dec 30, 2020 01:25
|
|
|
Microsoft outsources all the front line people and it’s a cold day in hell when you can actually get a non-presales MS engineer on the phone. This is with Unified Support and ime even yelling at account managers gets you nothing. God help you if you open a critical ticket though - they bombard you with calls like there’s no tomorrow and you have to talk to the crit sit manager, a random ticket Sherpa, and then five other layers before they connect you with another rando outsourced ticket watcher who can very slowly interact with the actual MS teams. Edit: and if their dev teams for various products actually talk I’d be amazed, there is so much overlap and duplicated effort especially with their SaaS security stuff
|
|
|
#
?
Dec 30, 2020 01:39
|
|
|
What kind of Microsoft security tech stack duplication are you seeing?
|
|
#
?
Dec 30, 2020 01:42
|
|
|
12 rats tied together posted:If you're going to approach this purely with python, I like/have used concurrent.futures. It has support for both threading and forking, so you can try both and see which you like best. This is way late, but I finally got a chance to look at this this weekend and it looks great. I did some proof of concept stuff that seems to work and I'm gonna try to put it into practice soon for something useful. Thank you!
|
|
#
?
Dec 30, 2020 01:47
|
|
Gabriel S. posted:What kind of Microsoft security tech stack duplication are you seeing? For starters, you have MCAS and Azure Identity Protection producing literally the same alerts. I don't know if they're trying to close the gap, but MCAS will produce it's own alerts reading out of the same logs as Azure. So you can easily have alerts in both places, and if your teams aren't huge one or the other is redundant. I prefer MCAS since it seems like they're trying to consolidate on it, but MCAS also does the reverse proxy stuff and is like... a CASB as well? And then you can send firewall logs to MCAS for cloud app monitoring (assuming you used the right kind of firewall that allows MCAS to actually do DNS and doesn't just identify everything as AWS), but you also might be sending the same to Sentinel. Obviously Sentinel the SIEM vs MCAS the... whatever it is aren't literally the same but there is overlap that can be irritating imo. You have Azure ATP (I think this was rebranded as Defender for Identity) that can do some cool stuff in terms of monitoring domain controllers for a range of things and lateral movement, but if you're using a SIEM (particularly Sentinel) they're reading the same logs and are capable of reporting on the same exact things. But Azure ATP has the honeypot stuff bolted on and a few other features that might mean you want to use it. Azure ATP will also send alerts to Security in the Incidents section, while also maintaining it's own ever-growing list of LDAP recon done by innocuous vendor service accounts (usually for security products in my experience). Now that Security and Compliance have been torn apart, you also wind up having alerts living in both places or elsewhere for the same exact things (edit: and in MCAS!). Security also has advanced hunting now using the same exact KQL queries that you'd use for Sentinel/Log Analytics (which can also produce alerts) while Compliance doesn't because... I dunno I guess someone thought splitting them in two and developing them separately worked better? So let's just say, for example, someone coughed their credentials up and Conditional Access/MFA failed. They create some forwarding rules and start spamming the poo poo out of people. Web only. Where are you going to see alerts? Compliance - alert that an email forwarding rule created Security - alert that an email forwarding rule was created, and an alert about spam activity (and it will likely have dumped poo poo into both action center and incidents for the same event) MCAS - alert that a forwarding rule was created, atypical travel alerts, spam alerts Azure Identity - Atypical travel All of these are closed separately. And the best part is, none of them will have alerted you that Exchange blocked outbound mail from that account. Also - I realize Sentinel and Azure AD stuff isn't part of their SaaS stack but I consider the Security/Compliance/MCAS/Defender overlap to be not great personally. Defender-branded products have their own databases/logs, but MCAS/Security/Compliance are reading data out of the same places for the most part so I'm not really sure why they spray the alerting all over your tenant the way they do. i am a moron fucked around with this message at 02:08 on Dec 30, 2020 |
|
|
#
?
Dec 30, 2020 02:06
|
|
|
Got an entire company staff email from a coworker announcing a mobile phone security policy change being enacted on Thursday of this week (I am off all week as is a lot of the company) that would likely lock out quite a few users. Got a follow up email about an hour later stating that this is being delayed until the 6th. Good stuff!
|
|
#
?
Dec 30, 2020 04:44
|
|
|
Bonzo posted:There's an SVP somewhere doing this to fudge Time To Close numbers . Oh, absolutely. And all the front line techs have all figured out that they can hide behind that bureaucracy .
|
|
#
?
Dec 30, 2020 07:02
|
|
|
Outsourcing is awesome, i've never found one that doesn't constantly pass poo poo off.
|
|
#
?
Dec 30, 2020 07:33
|
|
|
Internet Explorer posted:Going to disagree with the MS comment. That is still very much their culture. I could give a bunch of examples, but go try to open a ticket about anything in Azure or O365 and see how support acts. I don't think I've ever gotten through a ticket successfully, and always passed between 4-5 departments. Hell, they can't even pass tickets between some of them. They ask you to open a new ticket. I've had a ticket open for 3 weeks bounced between "Azure Team", "Intune Team", "Windows Team", back to Intune Team, Back to Azure Team, still no one has figured it out, no one wants to own the ticket. It's like the Christmas fruitcake of tickets. Asked VMware if they could figure it out, they half-solved it then stopped the person who knew everything and made her tell me to ask our account rep to sell me Onboarding Engagement Ticket Credits, been waiting days for a quote. i am a moron posted:Nutanix Deez Nutanix Zero VGS fucked around with this message at 07:54 on Dec 30, 2020 |
|
#
?
Dec 30, 2020 07:47
|
|
|
Opening an InTune/Endpoint/Autopilot ticket is a total loving waste of time. They ask for logs that I'm pretty sure they could get to themselves, and follow a script that is wholly unhelpful. The one time they've came through is when they connected me to an SME for comanaged deployments (it involved changing a piece or two their own published instructions for deploying the Config Manager agent via Endpoint), but outside of that I wind up sidestepping whatever MS is recommending to get something to work. The "Windows Team" is an endless pile of poo poo, I watched a VDI architect go partially insane as he tried to piece together why his non persistent environment couldn't do the normal same sign on auth as regular desktops/laptops. He wound up disabling WAM to get it all working while Microsoft insisted it was an unsupported fix. They have, months later, failed to diagnose the issue or provide a fix. This poor guy has been up the account managers rear end about it and they still have produced nothing of value. I dunno when ADAL will wind up not working and I'm not looking forward to the inevitable back and forth with MS over it. edit: another thing that pisses me off about Microsoft is that I work for a Gold partner, we have advanced specializations, yadda yadda and part of the whole deal is deployment assistance. That deployment assistance is literally someone reading off of their own Technet articles. When you encounter something that is broken, or the documentation is wrong, or whatever, their deployment people tell you to pound sand and open a support ticket. Well they give partners poo poo-tier support, so the deployment folks wind up literally just saying "uhhh follow the documentation or open a ticket". Which is loving hilarious when you're working with a client that Microsoft's sales team referred to you and their products are broken beyond belief. I could probably write a 40 page manifesto of why you should never use Microsoft products at this point, but I'm cursed to be at a .net/Azure shop so it would be screaming into the void. i am a moron fucked around with this message at 08:09 on Dec 30, 2020 |
|
|
#
?
Dec 30, 2020 08:00
|
|
|
i am a moron posted:Right! That person will have to do more than Nutanix now and/or in the future. Cover the expertise with some capex in the short term and help the person you suggested apply to the position grow into what they need to do. Yeah, or ask for professional services from Nutanix to help. Surely the skills gap and were factored in as part of the deal with Nutanix and therefore known about before tapping Japanese Dating Sim up for the job. Bad management. Good luck with your future endeavors Japanese Dating Sim, sounds like you may have dodged a bullet. Nutanix, lol.
|
|
#
?
Dec 30, 2020 10:29
|
|
|
Japanese Dating Sim posted:Thinking the start of a new year would be a good time to sprint for a few certs while casually looking for a new role. I like where I am so I'm not in a rush, but I'm definitely going to be on the lookout, whereas I was content to wait before. Sounds like they’re likely to pick an external hire over you any time unless they can’t find some better. You ain’t being appreciated and this is unlikely to change. Start putting your resume out to vacancies you really like and are ideally a little bit out of your comfort zone. Picking up a cert will help to show you’re motivated to improve your situation and take initiative to do so.
|
|
#
?
Dec 30, 2020 10:55
|
|
|
What kind of crazy tickets are you guys working on that so complicated? I'll admit I do absolutely think Microsoft is doing a great job with Windows, Azure and Office 365. Even outside of that, Microsoft isn't the bullying company that in was in the 1990s. That said, I do think it is incredibly how certain parts of these things are just terrible. That's from Azure Backup to little things Microsoft taking decades to replace the Snipping Tool with something more modern like Snip and Sketch. Apple has done little neat little features like this back in freaking 2007.
|
|
#
?
Dec 30, 2020 11:27
|
|
|
Snip and sketch is garbage though
|
|
#
?
Dec 30, 2020 16:19
|
|
|
Gabriel S. posted:What kind of crazy tickets are you guys working on that so complicated? Any time two separate Microsoft services conflict with each other that would require separate MS support teams to cooperate with each other to find a solution its a nightmare. As an example, in Teams I created a channel ChannelA when there was already a folder in the relevant SharePoint called ChannelA, and a folder ChannelA1 was created and I couldn't find a way to get ChannelA to affiliate with a SharePoint folder ChannelA. Made a ticket trying to understand why, and the SharePoint team and the Teams team were passing me back and forth between each other, each saying its the other's responsibility, because they have no knowledge of the other product, and had no method to be able to communicate with each other's respective teams. I had a similar issue when I was trying to get PowerAutomate to write to a SharePoint and the PowerAutomate team and SharePoint team kept bouncing me back and forth between each other because this is clearly the other team's problem and again, had no established method to communicate with each others teams.
|
|
#
?
Dec 30, 2020 16:38
|
|
|
That's when I pay an MSP to be the middle man and fix it.
|
|
#
?
Dec 30, 2020 16:41
|
|
|
GreenNight posted:That's when I pay an MSP to be the middle man and fix it. I literally laughed out loud at this.
|
|
#
?
Dec 30, 2020 16:42
|
|
|
Internet Explorer posted:I literally laughed out loud at this. One throat to choke tho lol.
|
|
#
?
Dec 30, 2020 16:43
|
|
Gabriel S. posted:What kind of crazy tickets are you guys working on that so complicated? O365 has no competition in terms of features, visibility, etc.. Gsuite is a distant fifth, sixth option and there is nothing in any of the spots in between. My observations/annoyance with feature overlap notwithstanding. Azure is good. It's cool! But when you get into a situation where, say, a bunch of devs deployed function apps using the super-duper, gotta have it flexibility of non-reserved instances only to find out MS has really tight limits and don't publish them at all and it takes their support team three days to figure out why you're getting transient errors with zero explanation for this piece of poo poo your app team developed you really wonder why you'd ever pay them for support.
|
|
|
#
?
Dec 30, 2020 17:16
|
|
|
Internet Explorer posted:I literally laughed out loud at this. Maybe I'm spoiled but I have some pretty excellent MS partners who know their stuff.
|
|
#
?
Dec 30, 2020 17:31
|
|
|
klosterdev posted:Any time two separate Microsoft services conflict with each other that would require separate MS support teams to cooperate with each other to find a solution its a nightmare. As an example, in Teams I created a channel ChannelA when there was already a folder in the relevant SharePoint called ChannelA, and a folder ChannelA1 was created and I couldn't find a way to get ChannelA to affiliate with a SharePoint folder ChannelA. Made a ticket trying to understand why, and the SharePoint team and the Teams team were passing me back and forth between each other, each saying its the other's responsibility, because they have no knowledge of the other product, and had no method to be able to communicate with each other's respective teams. I had a similar issue when I was trying to get PowerAutomate to write to a SharePoint and the PowerAutomate team and SharePoint team kept bouncing me back and forth between each other because this is clearly the other team's problem and again, had no established method to communicate with each others teams. Copy your TAM on everything, and go to them when you get pushback from the technical teams.
|
|
#
?
Dec 30, 2020 17:40
|
|
|
My org pays MS enough money that we have dedicated support reps, we never have an issue bouncing between people because these reps direct the whole case the whole way. It's nice but hoooo boy must be expensive.
|
|
#
?
Dec 30, 2020 17:56
|
|
Huh. I just looked up the Unified Support and had no idea there were tiers. Supposedly for "Performance" you get something called a technical advocate (rather than just a service delivery manager) and it's 2% more expensive than "Advanced" support. I resent that I know this now. edit: my math sucks, it's 2% more of your total spend but would be 20% more expensive than Advanced.AreWeDrunkYet posted:Copy your TAM on everything, and go to them when you get pushback from the technical teams. This is not always as helpful as you'd want it to be. I'm sure all of this depends on org size and spend too.
|
|
|
#
?
Dec 30, 2020 18:25
|
|
|
I'm starting to think that MS artificially lowered their M365 support estimated reply time by creating a layer of unskilled "Support Ambassadors" who inevitably have to route administrators to people with actual technical knowledge. Lately MS tickets have been "transferring you to a specialist" from whoever initially takes the case, and it takes significantly longer than the initial estimate for a callback.
|
|
#
?
Dec 30, 2020 18:32
|
|
|
Oh absolutely. They'll claim they met their SLA since someone looked at the ticket, marked it in progress, and called you (when you requested email) to let you know someone else will be assigned to it. I just checked - the only ticket I've had routed directly to someone that didn't have something like (Accenture Private Solutions Ltd) (or any number of other outsourced names) was when I opened a ticket because the Teams Admin center was denotating every time I ran an extremely simple PowerShell script to get the names of everyone on all the Teams into a CSV. It would fail halfway through and then the GUI would show them as inaccessible. It was weird, I figured maybe somehow it was an API limit or something. Someone from Microsoft responded that they were escalating it, then the actual MS engineer got back to me a day later. They decided my issue was baloney and transient, so they closed it. I ran the same script thinking they knew something had happened and it was over, and the Teams admin center blew up again. Reopened for the same song and dance. Fun times.
|
|
|
#
?
Dec 30, 2020 18:47
|
|
|
Don't know anything about Azure but gently caress Accenture. Had a ticket open with them for over a month. It took that long of them closing my tickets and reopening again before some senior engineer who wasn't a piece of poo poo called me. He said he was the only one in his team who hadn't looked at it yet and the outcome was 'i think it's a bug, hopefully it'll go away soon'. I had to be a total piece of poo poo and phone the service desk more or less every other day because they kept saying they'd call me back the next day and never would.
uhhhhahhhhohahhh fucked around with this message at 19:09 on Dec 30, 2020 |
|
#
?
Dec 30, 2020 19:06
|
|
|
They get referred to as Accidenture over here in  for a reason.They're poo poo.
|
|
#
?
Dec 30, 2020 19:26
|
|
|
What would cause a laptop to keep getting kicked from the domain ("a trust relationship could not be established")? I checked for a duplicate name but that doesn't seem to be it. My fix is log in with local admin account, remove from domain, reboot, join domain, reboot. Works fine for a couple days and then it doesn't. It's annoying af because it's at a remote shop 30 minutes away. Any ideas on things to check would be appreciated!
|
|
#
?
Dec 30, 2020 19:33
|
|
|
Any issues with the clock?
|
|
#
?
Dec 30, 2020 19:35
|
|
|
Next time rename the device with an "a" after or something and see if it happens again.
|
|
#
?
Dec 30, 2020 19:36
|
|
|
regulargonzalez posted:What would cause a laptop to keep getting kicked from the domain ("a trust relationship could not be established")? I checked for a duplicate name but that doesn't seem to be it. My fix is log in with local admin account, remove from domain, reboot, join domain, reboot. Works fine for a couple days and then it doesn't. Have never seen anything like that but you might look at %windir%\debug\NETSETUP.log and see if there's any useful info there. That file logs some of the domain join process - I'm not sure if it would show any errors causing it to leave the domain, though. I'm assuming since the problem started you deleted the object out of AD?
|
|
#
?
Dec 30, 2020 19:36
|
|
|
Dick Trauma posted:Any issues with the clock? GreenNight posted:Next time rename the device with an "a" after or something and see if it happens again. Japanese Dating Sim posted:Have never seen anything like that but you might look at %windir%\debug\NETSETUP.log and see if there's any useful info there. That file logs some of the domain join process - I'm not sure if it would show any errors causing it to leave the domain, though. Ah I didn't before but I'll delete it from ad this time, before I rejoin. Thanks.
|
|
#
?
Dec 30, 2020 19:43
|
|
|
By the way - you don't have to reboot after leaving the domain and then again when rejoining - you can skip the first one. Assuming you're doing this through the System Properties window - leave the domain, close the prompt to reboot, go back in to the "Computer Name/Domain Changes" window, rejoin the domain, and then reboot.
|
|
#
?
Dec 30, 2020 19:47
|
|
|
Oh nice. I'm just usually wary about doing too much during a domain join. You can also change name and join in one step and it works but it always throws an error (something like, "something went wrong. Please try again later.") and with everything having an ssd and rebooting in 15 seconds I don't like to push my luck.
|
|
#
?
Dec 30, 2020 20:13
|
|
|
regulargonzalez posted:Good thought! Usually I catch that when I'm imaging/deploying but that might be it. Check for 5 minutes or greater clock skew. That will break trust relationships.
|
|
#
?
Dec 30, 2020 20:16
|
|
|
I've never had an issue with MSFT support, but we've always been paying a ton for Premier support, so if we do run into trouble our TAM gets involved. Low priority tickets can take a while to get to someone who knows what they're doing, but they've always resolved the issue. Someone really sold the company I'm leaving on Nutanix hard core at the upper management level. They're wasting millions of dollars trying to make Nutanix work with our environment. Hyper Converged has its place, and in certain situations I'm sure it works great, but it doesn't work that great in a lot of situations, or it's just insanely expensive. 5 year licensing and support on some of our nodes was being quoted at like 180 or 190K, plus the 60K for the physical HP DL380, and we were looking at 250K a node. Not to mention a ton of exceptions we had to carve out. Still have a VMWare cluster for RedHat. Nutanix Files was going to be way too expensive, so thats all going on traditional scale out NAS. I came from an acquired company where our Cisco UCS + NetApp solution was amazing and rock solid. Just a mess, but the politics involved... forget it.
|
|
#
?
Dec 30, 2020 20:41
|
|
|
Good posts y'all, While I've opened dozens of cases with Microsoft, I've never actually looked at the bills for support. If it's a bug, does Microsoft eat the cost or does that still eat out of the support contract?
|
|
#
?
Dec 30, 2020 20:47
|
|
|
|
| # ? Jun 6, 2024 08:23 |
|
|
regulargonzalez posted:What would cause a laptop to keep getting kicked from the domain ("a trust relationship could not be established")? I checked for a duplicate name but that doesn't seem to be it. My fix is log in with local admin account, remove from domain, reboot, join domain, reboot. Works fine for a couple days and then it doesn't. In my experience, a better nuclear option is you remove from the domain, don't reboot, readd to the domain, reboot. If you don't do that, a second computer object with the same name gets generated and poo poo stops working because they confused. Check AD for a second computer object with the same name. If you already in that boat, remove the system from the domain, reboot, delete all the duplicate computer objects. Then either wait several sync cycles until the change is propagated literally everywhere, or change the off-domain system to a slightly different name. Re-add to the domain, reboot, move the new object where it belongs, gpupdate, and if you renamed earlier, change the computer name back to what it was, reboot, log in again, gpupdate, and the name should update eventually in AD. klosterdev fucked around with this message at 00:26 on Dec 31, 2020 |
|
#
?
Dec 30, 2020 21:09
|
|