Posts made by PSX_Defector

@storageninja said in Looking for virtualization advice:

Something I learned from being a group admin for Spiceworks for years is that enterprise types lurk heavily in SMB forums.

Hey, shush your mouth. I don't lurk, I troll.

Real simple:

start /B /W wmic.exe /interactive:off ComputerSystem Where "Name='%computername%'" Call UnJoinDomainOrWorkgroup FUnjoinOptions=0

You can either do an Enter-PSSession or use psexec.exe to execute. Put a reboot command in a batch and call it a day.

Make sure to create a new glassbreak account local to the PC, so that every machine has a specific account someone can perform work with.

https://ark.intel.com/products/88179/Intel-Pentium-Processor-G4400-3M-Cache-3_30-GHz

This is a Pentium, not an i3. Slightly different, but perfectly capable of running Windows and most of Office without much fuss.

"Office" use is subjective. Are they putting around on a website all day or crunching crazy piviot tables run from a local Access database?

@scottalanmiller said in Best ADSL Wireless Router-Home Usage:

OMG @PSX_Defector is alive!

Two weeks in the northwoods and no topics that really cater to my knowledge before that, yeah, I'm kinda hard to find.

@irj said in Best ADSL Wireless Router-Home Usage:

From my experience only consumer grade devices are an all in one. Even the low end professional routers will not include a modem.

http://www.cisco.com/c/en/us/products/collateral/routers/4000-series-integrated-services-routers-isr/datasheet-c78-733645.html

http://portal.adtran.com/web/appmanager/portal/Adtran?_nfpb=true&partNumber=1202890E1&productGroupId=4492&selectTab=modules&_pageLabel=waverunner_page_products_productdetail

Despite folks thinking otherwise, DSL is still alive and kicking. All of the RBOC still have large ADSL/ADSL2+ installs, with people like CenturyLink and AT&T using VDSL to supply service as well. Not to mention the SDSL circuits being pushed everywhere for "T1" service.

For OP, I would recommend Zyxel equipment, if serious for an all in one solution:

http://www.zyxel.com/us/en/products_services/802-11n-Wireless-ADSL2--Gateway-P-660HN-51/

The P-660 series is rock solid. I've got a few old school P-660s floating around from when I had older ADSL circuits. Very good at getting synch without having too much noise introduced into the line. Where the AT&T CPE would only get barely 3Mbps, this one would synch at the full profile.

@DustinB3403 said in Tape backup advice ?:

Hand carrying it poses risks as well, like risk from a natural disaster. The goal is to have a geographic separation beyond an hours drive.

That's what you pay Iron Mountain for. They have vaults that can withstand just about anything you throw at it. Short of a nuke going off in the city, then you have other problems to worry about than where your backups are.

Sneakernet throughput is insane compared to most affordable pipes. It takes about 2 hours to go through a tape straight stream. LTO7 holds 6TB/12TB. Even factoring in the transport time of ~4 hours after pickup and 1 hour to put it in a box, you are looking at ~550Mbps with full 2:1 compression. I can sustain that speed from two of my DCs, but hell if I want to send it across to London or Singapore.

Let's not forget about the fact that you are stuffing the pipe with all that traffic. I doubt most SMBs have even the ability to get a 10Gbps pipe I got. My 10Gbps pipe is not dedicated to our DR product, but all traffic. So even if I was replicating 6TB every 6 hours, I have plenty of room to allow others to do things. With most SMBs, they could hope for a 20Mbps up, but most are running on commodity turd pipes with shittastic peering which slows them down.

Of course, this all predicates on what exactly you are backing up. I have 4PB in one DC alone. Tape makes sense. Most SMBs, once their digital packrat tendencies has been culled, don't have even enough to fill up an LTO4 tape.

Cloud backup for small datasets, tape for large.

@DustinB3403 said in Tape backup advice ?:

Why Tape, it's slow, difficult to transport and requires human interaction.

Well, two out of three ain't bad.

Tape is fast as hell, plenty faster than anything you got for a pipe:
https://www-03.ibm.com/systems/storage/tape/oem/lto6/full-high/specifications.html

Difficult to transport? I can throw an LTO cartridge across the room and it's ready to go. It fits in my pocket, and really don't need much effort to put it in a box and have someone pick them up.

Human interaction is a real concern, but that's what you pay smarthands for. I would have a guy come by the DC once a week, pull out an entire tray of tapes from the library, replace them with fresh tapes, and be out the door in minutes.

@Dashrender said in Checking on patch levels with multiple clients. ninite pro?:

@PSX_Defector said in Checking on patch levels with multiple clients. ninite pro?:

n talk to any of my hosts in any of my datacenters. Just have to configure the client side to talk to your WSUS server, which isn'

My Win 10 1703 clients won't talk to my WSUS servers anymore - there appears to be a problem, but looking this morning, no solution yet.

Run get-windowsupdatelog, if it's not talking, it's gonna tell you why.

Most of the time when I look, I find that the machine has been flagged for not using WSUS and it's going to Microsoft, which in turn is blocked because we don't provide RNAT by default. Make sure your path is set correctly in HKLM\Software\Policies\Microsoft\Windows\WindowsUpdate and that UseWUServer is set to 1 in AU.

@Mike-Davis said in Checking on patch levels with multiple clients. ninite pro?:

@black3dynamite said in Checking on patch levels with multiple clients. ninite pro?:

Maybe a combination of WSUS and PDQ Inventory would help.

I'm not really considering WSUS because I would have to log in to every server to check on every client. Some of them don't even have a server. I'm looking for a single pane of glass.

Use a hosted WSUS instance and you will be fine, just open some ports to allow it through. I have thousands of VMs, they can talk to any of my hosts in any of my datacenters. Just have to configure the client side to talk to your WSUS server, which isn't that bad.

Remember though that WSUS is a pull system, not a push. You can lead a machine to water, but you can't make them drink. So configure it on the client side to force it as best as you can.

@Harry-Lui said in Need advise, restoring domain controller and email server.:

This morning, I noticed the exchange server have only 5% of free space left, I tried to free up some space by deleting old administrators users inbox, but exchange will not let me saying I don't have the permission to do so. Following https://community.spiceworks.com/topic/386021-cannot-delete-ad-user-insufficent-rights-or-protected-from-accidental-deletion

Strange, because you later say:

I'm in the process of restoring Exchange to the yesterday's 7pm backup now, which will take several more hours before it finish.

Which usually fixes the disk space issue. I'm guessing you are using Veeam or some other VM level backup rather than the proper application level backup which truncates the logs on the Exchange EDB.

On the plus side, assuming your restore works correctly and most of the stuff is in place, you might not need to do much more than that. But knowing the Exchange part of this, I'm guessing you also didn't backup your AD correctly either for an authoritative restore.

You are in for a rough night my friend. First and foremost, if the user objects have not been deleted, don't start now. Exchange ties into the GUID, so not having it will cause your restore to pretty much be useless. If you already deleted them, you are looking at some really messed up stuff. If you have good backups, you might have to straight up restore to a previous point in time completely, that means nuking everything. If you had a block level backup, you might be almost OK then, assuming you don't have too much of a delta between the machines.

Contact Microsoft, even at the worst paying $500 will help you out immensely when you need to do some more advanced AD stuff to restore.

@Shuey said in Free SharePoint?:

The two major road-blocks to getting this accomplished are:

1. Upper management doesn't want to spend what it would cost to either get us out to the cloud (via O365)
2. We don't have anyone on our team with the necessary skills to do the migration, so we'd have to pay someone to do it for us (and it would have to be migrated to the same version to keep the cost as low as possible).

IOW, they see no value in the Sharepoint instance and are fucking morons who think everyone should work for free.

This isn't a debate, they WILL pay someone somewhere to do this. No one, outside of maybe me, has any experience with WSS 3.0 nowadays. Real Sharepoint developers never used it, admins would just not bother with it. And no one worth their salt is gonna install a 10 year old depreciated service anyways. Your only option is to migrate the data to 365 on the cheap end. Full Sharepoint 2016 instances get pricey fast, not to mention the need for a SQL install. And if you want it to have better reliability, two in the farm plus a redundant SQL instance.

If you don't know what's on it, find out. Because it can be just as easy as copy/paste files over from the old site to 365. If it's got templates that no longer work in 2013, then you will need to know that before hand and fix them up post migration.

@NerdyDad said in Free SharePoint?:

I missed the point of the DC. I would just decom the DC, then P2V the rest of the host. If you need a secondary DC then just spun up another VM and dcpromo it.

Which is why it is verboten to use SQL Server in an Active Directory Domain Controller. You demote it, you break everything.

@Shuey said in Free SharePoint?:

I'd prefer to go with option 1, but I've never virtualized a DC

Do it offline, not hot. That's about the only major thing to know about P2V of DCs.

Really though, you might want to rethink the plans here. First off, it's trivially easy to deploy a new DC, so why keep around an ancient install when you can clean up your environment. Second, you are running "critical" services on the same box as your DC, not to mention you have SQL Server installed on it as well. That is strictly verboten by Microsoft. Dumb is dumb. Now is the time to fix it all up.

What are you running in Sharepoint? Flat repository? Full workflow? Customized templates? This is what makes the determination of where you need to go with it. Most of the people I've seen with this kind of setup are using it as a glorified file system with a minor amount of workflow in it. Would take me about an hour to migrate it to even the newest version or 365.

@matteo-nunziati said in the missing VoIP, the ERP and the solution in search of a problem:

Is this going to kill my dog?

Are you a shithead, because in my experience only shitheads kill your dog.

is there a way to let our PBX silently pass external calls to the new VoIP PBX and let this manage the internal lines dedicated to customer care?

Depends. What you got? Newer PRI based PBX? Old ass key system? One of those Avaya IP Office boxes? Definity? Aspect?

Easiest would be an IAX trunk. Next would be a T1 bridge across. Bad would be piping 20/30 analog lines over to a Rhino breakout box.

@Texkonc said in ODBC password:

@Grey said in ODBC password:

http://i.imgur.com/tWvH7c2.png

I've done that and then the ODBC tool doesn't update after changing it.

Did you reboot 3 times?

No, that's run iisreset then reboot three times.

Shit, an old ass Honeywell mini. That's something you don't see every day.

If you were closer, I would take that one off your hands for $10. Shipping on that monster would cost a fortune, not worth it to come back to Texas. With the number of dumb terminals, there might be some other mini/mainframe stuff in there. But that's very niche and not worth a lot.

Most of the other stuff is cleanable and sellable. The TI-99/4a, TRS-80 and VIC20 are easy money. The PS/2 stuff, not so much.

@BBigford said in Matching tempdb files to physical cores:

Though, the client and a DBA both said that they had issues if it wasn't set that way. I had asked why they would need to be manually tuned but I may as well have been insulting them. Who knows I guess.

As you should have been.

Let's put it this way. When I worked for the big red V, I had a physical SQL server for a car financial company. 64 processors, 512GB of RAM, running SQL Server 2008 R2. It ran one TempDB. Dedicated to a drive, standard procedure for us deploying. We never had performance issues, and I can guarantee you that there is no way anyone outside of another bank runs SQL Server that hard. If it is good enough for them, it's plenty good enough for the 200GB DB with 200MB of transactions a day.

People who claim bullshit need to be called out on their bullshit. This falls under the "never virtualize SQL/AD" bullshit.

@mroth911 said in Used Scale HC3 Equipment for Sale:

Can I get SIX

You breaking my balls. Five dorrar fifty.

@mroth911 said in Used Scale HC3 Equipment for Sale:

Model Number is HC1150 3.48tb raw 1.74 TBU- 64GB Ram.
It cost me 29,997.00 Open for offers.

Five dorrar!

@Shuey said in SQL Administration Woes:

I've heard legendary tales of SysAdmins who have done SQL backups all for free, but I'm still way out of my league on this topic, so I'd like some advice if possible...

ANy good SQL DBA can setup maintenance plans to handle indexing, cleanup, backups, etc. all from within SQL server.

Any good SQL DBA CAN do all that, but I'm not a SQL DBA, lol. So I either gotta pay somebody to take care of it, or I gotta hope I can learn and implement it while I'm in the interim (and hope it doesn't all go south during that time) :-S.

https://msdn.microsoft.com/en-us/library/ms187510.aspx

USE <<DATABASE>>;
GO
BACKUP DATABASE <<DATABASE>>
TO DISK = '<<PATH.BAK>>'
WITH FORMAT,
MEDIANAME = 'Full_<<DATABASE>>',
NAME = 'Full Backup';
GO

Run as SQL job, done.

That will be $500 for DBA work.