Listen to the Podcast
28 Aug 2020 - Podcast #708 - (23:38)
It's Like NPR on the Web
If you find the information TechByter Worldwide provides useful or interesting, please consider a contribution.
If you find the information TechByter Worldwide provides useful or interesting, please consider a contribution.
This is a little story that's all about me, but it reveals a problem that might haunt you, too, even if you're careful. Despite caution when buying hardware, the result isn't always ideal. It's easy to misread reviews and specifications, but it's also possible to gather bad information from these sources.
My 25-month old Netgear R9000 router started misbehaving in July. I wasn't sure how long Netgear's warranty period is, but what I found when I visited the company's website wasn't encouraging. It showed my router's serial number and below the serial number were these words:
It did offer a service contract, though, and if I wanted to get it fixed, I would have to buy one for $230. That would also mean that I would have to send the router in for service and wait for its return, all the while being unable to connect more than one computer to the internet. Both my wife and I need internet access, and we need that access simultaneously, so sending the router in for service would never do, even if I wanted to spend $230 for a service contract, which I didn't.
Looking at my purchase records, I found that a Netgear router I had purchased in 2014 lasted a little more than four years before it failed, so I decided not to buy another Netgear router and instead identified a Linksys Max-Stream AC3000 Tri-Band Mesh Wi-Fi 5 Router for $258. I drove to the store, picked it up, brought it home, and installed it.
Click any small image for a full-size view. To dismiss the larger image, press ESC or tap outside the image.
Everything went well until I plugged in the external USB drive that I use for network attached storage. The router immediately told me that it recognized the drive (good), but that I couldn't use it (not so good). Why? Because, although Linksys installs a USB port, it doesn't support a hard disk. Maybe it supports a printer. I don't know and I didn't try to find out because the printer is attached to an Ethernet port. The router interface offered a link to the Linksys website where I learned that the MR9000X has no USB functions for disk drives.
Online spec sheets and other resources explicitly called out the USB port on this router, and there were even published reviews that described the router's robust USB3 port's use with an external drive to create network attached storage. Well, it's been about 20 years since I've owned any Linksys gear, and it will probably be another 20 years before I consider Linksys again.
To be fair to Linksys, I have to say that the installation process is the best I have ever seen for people who are unfamiliar with installing a router. It uses a phone-based app that walks the user through the process with a single question or action on each screen. The setup took about the same amount of time I would have needed to do it manually, and structured the task in a way that anyone would be able to do it — regardless of technical knowledge. If only Linksys had been equally forthcoming about the limited functionality of the USB3 port.
So I took the Linksys router back and paid an extra $54 to purchase a TP-Link Archer AX6000 router.
Setting up a router is something that's not at all complicated. I had even downloaded the router's manual to confirm that the USB3 port would support an external drive, so I expected the setup process to take 15 minutes. Maybe half an hour at most. Two hours later, I still didn't have a network connection. This wasn't entirely the fault of the router. Somehow the cable modem had become confused and was returning an IP address of 0.0.0.0. Once that was resolved, the rest of the installation proceeded normally. Until I got to the USB drive.
The router recognized the Western Digital drive and showed all the files and directories on the drive. It offered to share them all, and things were looking good. But pointing a file browser at the network address \\TP-Share revealed an empty directory. I could not add a file or directory to that location.
The router's setup page for the NAS drive showed an entry I couldn't change. The entry suggested that there was a folder called "G:" or maybe that it wanted to mount the drive as G. The problem with that is that I already have a drive G and several applications depend on it continuing to be drive G. But just to test, I temporarily remapped drive G to another letter to see if the network drive would show up as drive G. It didn't and that solved one mystery: "G:" is really just a directory called "G".
That suggested I might be able to connect to \\TP-Share\G and map it as drive Z:, which is the letter I have always used for network shares. Indeed I could do that, and it seemed that the problem was resolved. That would have been the case if I wanted only to read files from the directory. It was impossible to add a file, edit an existing file, or delete a file because the entire folder was read-only.
Even though I had performed the appropriate due diligence, including downloading and reading the relevant parts of the operations manual, the inability to use the NAS drive was a fatal error. The manual didn't say that the NAS drive would be read-only, but it also didn't claim that the drive could be written to. By then I had spent another hour on a chat session with a TP-Link support technician, who seemed to think that it should work, but was unable to explain how.
My question bubbled quickly up to a senior technician, who referred me to an explanatory page on the company's website. The control panel for the router should have displayed an "Access Authentication" button that, when turned on, would then have opened a function where user access could be enabled. That button, and one labeled "Enable Media Sharing", were not present in the router's control panel. Had they been, finishing the setup would have been quick and easy. The senior technician said my screen shot that showed the absence of those two controls was "really odd".
Replacing a router isn't supposed to be something that takes several days. Fortunately, I had completed several projects ahead of schedule and my wife had two days of vacation, so I could devote two days to solving a problem I didn't want to spend time on. This meant that I had to ignore the cat and she was quite put out by this. My wife and I could have continued to work normally because neither of us depends on the network drive for daily work, and the Wi-Fi part of the router was working well.
TP-Link did offer a online support, but the company's headquarters is in Shenzhen, Guangdong, China. Despite having a US presence in San Jose, the company's support seems to be offered only from China. The technician I would have worked with works from 9:00 to 18:00 in a time zone (UTC +8) that is 12 hours ahead of my Eastern US time zone, so he's available only from 9pm to 6am my time. In short, we never made direct connections, but did exchange a bunch of email messages and the technician confirmed that every unit he tested worked as expected..
The Netgear R9000 router did everything I needed it to do, so I bought a new one and thought that setting it up would be a quick and easy Saturday morning job. Two hours later, it was still no-go.
The setup process is handled by a phone based app. The app couldn't see the router and it instructed me to open the phone's control panel and manually connect to the router. When I did that, I found that the phone was already connected to the router. Well, if the app can't see the router, I could still connect to the router's IP address from the computer that's connected via an Ethernet cable. The computer couldn't reach 192.168.0.1 or 192.168.1.1, but 192.168.1.1 did launch a Netgear information page that told me I should connect to the internet to use RouterLogin.net.
Perhaps you already see the logic flaw there: If the router can't reach the internet, I can't connect to RouterLogin.net. So setting up the router seemed impossible, but I continued to tinker with it until late morning. Part of the tinkering involved trying to find a phone number for Netgear support, but all I could find was a website that I connected to using a phone. That worked about as well as the router did: After giving the website the router's serial number, the option to proceed was grayed out. If I wanted to proceed, I would need to register the router. That seemed like a good place to end the exercise.
Another defective router. Another return. This was becoming quite tiring.
I decided to try another TP-Link Archer AX6000 router because overall I liked the way it works and the senior technician in China seemed certain that the problem with read/write permissions on the USB drive was simply the result of a defective unit. I asked both TP-Link and MicroCenter if there was a way that the router's operation with a USB drive could be confirmed before I brought it home. My concern was that MicroCenter's stock might all be from the same batch and they might all have the same undiagnosed problem. Many users probably don't attach a USB drive, so the problem could go unnoticed until someone needs to attach one.
MicroCenter was willing to work with me, but I decided not to bother them with a problem that wasn't of their making.
Another week's worth of back and forth with a very helpful TP-Link technician named Aaron in China and a useless response from a Western Digital technician named Anna, who responded as if she hadn't read my explanation of the problem, convinced me to scrap the network attached storage drive idea. Not the drive; just the idea.
The USB drive is now attached directly to my computer as an external USB device. It mounts as drive M and I have shared it on the local network. That solution provides about 90% of the functionality I had with the network-attached drive. What's missing is the ability to access the drive when the primary computer is turned off. Another alternative, and one that I may switch to later, would be to connect all of the directories that I want on a network drive to Google drive. That would allow the files to be available on all computers at all times and also to ensure that changes made on another computer would be reported back to the primary computer.
Although the final solution is annoying and somewhat disappointing, it's better than spending another few days or weeks trying to solve a problem that has no straightforward solution. This causes me to remember The Gambler that Kenny Rogers sang about, and knowing when to hold 'em, knowing when to fold 'em, and knowing when to walk away. For me, it was time to walk away.
Whether you call them potentially unwanted programs (PUP) or potentially unwanted applications (PUA) they're annoying and some of them are dangerous. Some browsers attempt to block these annoyances, but Microsoft's latest Windows 10 update aims to put the protection at the operating system level.
Click any small image for a full-size view. To dismiss the larger image, press ESC or tap outside the image.
Microsoft's new Chromium-based Edge browser has had the capability since March. If you have the new version of the browser, type edge://settings/privacy in the address bar to see the settings.
The security section deals with tracking protection and other issues at the top. Scroll all the way to the bottom to find a section labeled "Block potentially unwanted apps." This is turned off by default, and I recommend turning it on.
An unwanted app can be safe but annoying, such as changing the browser's default search engine, but it can also be something that installs malware or spyware, or something that installs unwanted so-called "security" applications that repeatedly tell you there's a problem with your computer. The latter apps promise to fix the problem, but only after you buy the "professional" version.
They can also show ads, track users so that the developers can sell their data to advertisers, change browser settings, install security certificates, and even disable security controls.
Sometimes unwanted applications are easy to remove, but sometimes they can be next to impossible to uninstall. Having protection at the browser level is helpful if a website attempts to install something you don't want, but it's helpless to protect against an application that tricks users into installing unwanted code or hides the process altogether in a silent installation.
The new protections in Windows 10 really aren't new. Windows Defender has been able to protect against these rogue applications for year, but the function could be enabled only via Group Policies, a feature that's not available on the Home edition of Windows 10 and is nearly invisible in the Pro edition unless you're a system administrator and know where to look.
Starting with version 2004 of Windows that has been slowly rolling out to computers since May, protection against these applications is in Settings > Update & Security > Windows Security > App & browser control > Reputation-based protection settings.
If you can't find this option, it's probably because your computer hasn't yet been updated to version 2004. More about that in a moment. If it's present on your computer, you'll see that it's turned off by default. Microsoft recommends turning it on and enabling both option.
If Microsoft is going to provide this protection at the operating system level, then it should work at the operating system level, and not be used to give Microsoft's browser an unfair security advantage.
A surprising number of Windows computers are still running version 1909 and haven't been updated to version 2004 even though version 2009's release date is right around the corner.
Microsoft numbers releases with the last two digits of the year and the month the version was released, so 2004 means April 2020. Oh, and try not to mention that the April 2020 version wasn't really pushed out until May and possibly still hasn't been pushed out to your computer.
If you'd like to fix that, it's quick and easy. I have to insert the standard warning here: Anytime you install a new version of the operating system, manually or automatically, something can go wrong. Make sure your backup is current before starting.
That said, version 2004 has been out long enough and is installed on enough computers that the likelihood of something going wrong is reduced.
Start at Settings > System > About to see which version of Windows is installed. If it's not 2004, it will probably been 1909. If you see 2004, there's nothing more to do.
Click any small image for a full-size view. To dismiss the larger image, press ESC or tap outside the image.
The next stop should be Settings > Update and Security. You might see a "coming soon" message about version 2004 below the button to check for updates. If you see that and you're tired of waiting, here's how to proceed:
Go to Microsoft's version 2004 status page to check for known problems. If hardware installed in your computer is listed and the issue is not shown as resolved, you should stop and wait for the issue to be resolved before installing the update. All major issues and most minor issues have been resolved.
There are two choices for updating: The Windows 10 Update Assistant or the Media Creation Tool. The Media Creation Tool is intended primarily for system administrators who have multiple computers to update, so let's use the update assistant.
Go to the Download Windows 10 page and click on Update Now. This will download a file to your computer.
When the file finishes downloading, double-click it to launch the updater.
When the app opens, click the Update Now button and the updater will confirm that the computer's CPU, RAM, and disk space are adequate. Assuming they are, click the Next button.
The next screen will ask what you want to do. Select "Upgrade this PC now."
The updater will download some files and, when that process is complete, it will tell you that the computer must be restarted. You can do nothing and the computer will restart in half an hour, or you can save all open files, close the applications that are in use, and click Restart Now.
The computer will restart several times during the process. Depending on the speed of the computer, the process may consume just 20 minutes or so -- or it could take a lot longer. Patience is a virtue.