Tag Archives: technology

I’m Getting Too Old For This

It’s been a day.  Like I had just posted, my computer’s main drive went south and I had to do a rebuild.  There was a time that I used to do this almost for fun.  like if your computer was running slow, you’d just wipe it and start over.  But that was before having terabytes of data and dozens of applications and hundreds of online accounts that all need to be handled.

I had gone to lunch and just when I was pulling back into the house, the Amazon driver was stopping in my front yard with my new drive.  Great timing.  While I wanted to do something else with my early afternoon – a casa de chostic worthy post of adding gutter guards – they didn’t work out so well, so I had the day to devote to system restore.

After blowing the dust out of the desktop and installing the new SSD drive, it was off to the races, on crutches.  I wanted to have as clean of an install as possible to avoid any bloated pieces, but unfortunately, that also leads to broken dependencies.  And I’ve been out of the system building game for too long to understand what’s what in Windows anymore.

So I removed everything I thought I could.  Applied all the patches that were available to me, then began the the dual process of installing applications and rebuilding my RAID mirror.  After getting KeePass and Outlook installed, I was able to breathe a little easier.  The bigger headache was setting up Plex again.  I had backed up my old database before wiping, but I don’t think it’s going to be any easier to restore it than it would be to just rescan everything.  Once I saw my new install created a duplicate server instance in my dashboard, I didn’t want to head down that path any further.

So after about 6 hrs., by RAID is about 50% built and I have most of my daily-functional apps installed: Outlook, Vivaldi, Money, KeePass.  And a couple other ones, Open Live Writer (present), 7Zip, Faststone, Image Resizer, CDWinEmu.  The others can wait until their needed, like Visual Studio.  Because that’s going to bring with it setting up Hyper-V for my TFS server instance.  Oh hmmm.  I might have blown away my local SQL server and who knows what in-progress database I might have had on it.  Oh well.  Maybe someday, I’ll need them and I’ll find some way to read a NVME SSD externally.

But back to the point of this post, I am exhausted, and I didn’t even do anything strenuous today other than climb on the roof for about 10 minutes.  Just the mental tediousness of restoring and configuring a system for the hundredth time in your life is draining.  But what’s the alternative?  I’ve never been trusting of system restores, only data restores.  It’s the path I’ve chosen to take.  And I guess it’s probably going to continue every 3 yrs or so.

Etched In Stone

It’s rebuilding day. I can’t recall the last time I had to reinstall my desktop system. I’d have to do a little research, but it was probably when I bought all new hardware and went with two large mirrored drives with virtual disks on it. Feels like that was some time ago. I looked it up. August 2019. So 2.5 years of running non-stop without significant downtime. I guess that’s not too bad.

It’s worked pretty well, until suddenly it didn’t. The machine would reboot overnight for no good reason. When i would look in the event logs, it happened around the time of a Windows update. There were a ton of errors about the Windows Search service not being able to start.

Then I was having trouble installing updates. My Visual Studio was on something like 15.3.3 and I wanted to update to 15.5. It would install the update, then wouldn’t launch. So I’d reboot and I was back on 15.3.3 again and it’s bugging me to upgrade. I planned on just uninstalling and downloading the latest, but then I noticed that my Vivaldi web browser was also bugging me for updates. I had just done that the other day. I installed the update, it wouldn’t launch. This seems familiar. Reboot, back on the old version.

This sounds to me like the shadow copy service is broken. It can’t create any new system checkpoints and keeps reverting to the last version. So I follow the steps to delete all system restore points and move onward with trying to fix the search service. Everything I try fails. System file checker, chkdsk, dism, all have some problem or another. So I guess it’s time to start over.

Ok, then the first thing to do is get a current backup. BUT, Windows File History relies on the Windows Search Service, which is dead. It says my last good backup was October, 2022. Whatever, it’s fine. All my real files are on the mirrored drives anyway. I leave the machine on overnight with plans to rebuild first thing in the morning.

Amazingly, it didn’t reboot overnight, so then I got busy building a USB drive to install from. Shut down the tower and took it outside to blow the dust out of it and set it back up for install. First attempt was booting off the UEFI partition on the USB drive. But then it wouldn’t let me install anything on my internal drive because it was MBR. Ok, reboot again into the MBR USB partition and try again.

I try to delete the partitions and they wouldn’t delete, for whatever reason. Fine. i formatted both partitions and did the install. On first reboot, Windows loaded up and prompted me to log in. Excuse me? This is a brand new install. I type in my pin and I’m back at my desktop.

I have no idea what’s up with this cursed drive, but it’s going in the fire as soon as I can get a replacement. Amazingly, Amazon can have me a new drive, twice as big delivered within a few hours for like only $65. I feel like I paid 3x that when I built this computer.

Until then, I guess I’ll just poke around on the drive that refuses to change, or die.

The Last Time Is The Charm

In an earlier post, I talked at length about this keyboard that came from a thrift shop and had a problem and how I tried to fix it over and over and eventually gave up and left it sitting in the garage, queued for the dump.  Well, sometimes, I just can’t give up. 

It’s been so cold the last few days that I didn’t want to touch the keyboard, seeing as it’s all metal (official weight: 51.4 lbs. – stupid heavy).  Today was warmer and I left the garage open so it could warm up a little.  After work, I went out and hauled the beast back in for yet another attempt at repair.

Did I have a plan?  Not really.  I was going to take the keyboard assembly out and just deal with the main board through MIDI.  Doing that, I could at least move something in and out of the closet that was probably 30 lbs. less every time I wanted to make another repair attempt.  And with that, I set the massive key bed out and hooked up the synth to another of my keyboards.

MIDI worked,  And enabling a sound on the master keyboard verified there was still the pitch problem.  I dug out the service manual and went through the reset and test modules.  Everything seemed ok from what I could tell.  A couple of the faders didn’t seem to register any movement and I had the pitch and mod wheels disconnected, so some blanks were expected.

Noodling around on the master keyboard, I happened to think to try the pitch wheel and when I moved it, I was very surprised to hear it didn’t work as it should.  The pitch would shift a little bit and snap back to the original incorrect pitch.  That would suggest that the contacts might be dirty, but this is a remote keyboard.  These are MIDI messages being sent and the synth is not honoring them.  Or something else locally is spamming the pitch controller signal.

I finally had a lead I could work with.  I started thinking schematically about what components could cause the pitch wheel to be triggered consistently even when disconnected.  I had the idea I should reconnect the pitch wheel and see if I could stabilize the pitch by holding it a certain amount.  To reconnect it, I had to remove a circuit board that had all the fader controls on it, like volume and four programmable sliders.

When I got the board off, I immediately noticed the connector for the pitch wheel had some trauma.  It was bent at an odd angle.  That seemed suspicious.  I inspected it closely and didn’t see anything broken.  But the faders were disgusting.  And as long as I had this board apart, I thought I’d try out my new chemicals, Deoxit and Deoxit Fader, the latter of which is specifically made for cleaning and lubing faders.

As I cleaned the faders, it was pretty clear they were just shot.  The cleaner was running down the board in a black oil.  But I cleaned them up as best I could.  With the pitch wheel reconnected, I powered the synth back up and tried out a few keys.  It was suddenly in tune.  No drift.  The pitch bend worked locally and remotely.  And that’s the end of this saga.

Reflecting on the "fix", when I very first got the keyboard, it had a pitch problem.  I was the one that disconnected the pitch wheel, so that was not the source of the problem.  One of the faders had to be the culprit – it must’ve had a short somewhere.  But I’m not going to be overly concerned about it from here out.  If the problem comes back, I will just buy a new fader board or maybe the faders themselves, since that is where the problem is centered.

The important thing is I have the keyboard I originally purchased in non-working order, now in working order.  And on a similar topic, I purchased another keyboard on a whim this week.  So I now have five keyboards and a stand that can hold two, so three have to be in storage.  I’m not sure where this is going at the moment, but we’ll see.  One thing for sure, I’m not getting rid of any more keyboards.

DIYDNF

In the world of Geocaching, DNF means Did Not Find.  In the world of home improvement, DIY means Do It Yourself.  Somewhere in my world, DIYDNF means Did It Myself; Did Not Fix.

For a few years I’ve had this keyboard that I picked up in a thrift shop for something like $100.  My evaluation of it once I got it home was that this keyboard had problems.  Its pitch would wander, sometimes higher, sometimes lower.  It got stuck in a closet while I determined what to do about it.

There is a repair shop about 90 minutes away that handles keyboards and after a long, long time, I finally dragged it out there on New Years day.  Sadly, they wouldn’t work on it because there were no parts from the manufacturer and no service manual.  I actually had a copy of the service manual, but whatever.  So I lost 3 hours and had a approximately 80 pound anchor on my hands.

I started doing some research online and after watching a repair video of the same model keyboard figured I could attempt the repair myself.  And to some degree, it wouldn’t be so expensive to try as a last resort.  Of course, expensive is something that comes with time and is usually not in the initial budget.

After watching many repair videos, the plan was to replace the main capacitors.  Kind of like when a CD player doesn’t work, you first replace the belts. That’s the way it is with a lot of older electronic gear.  The videos I watched, over and over, replacing the capacitors solved the majority of problems.  So that was my plan.  Now to buy everything I’d need to accomplish something like that.

What do I own now that I didn’t before?

  • Soldering station
  • Silicone soldering mat
  • 300+ capacitors
  • Two different solder wicks

Overall, maybe I spent $150 for this endeavor.  And because of the time it took to get all of these things, I had a lot of anticipation and excitement to get started.  The last piece arrived today and I immediately got to work.

Before I start that, I need to explain that in the time waiting for some of the pieces, I practiced on a dead circuit board – removing and replacing capacitors.  I thought I was doing really good.  It wasn’t difficult at all.  I was instructed the ideal way was to use the solder wick to remove the old solder, but I had zero success with that and chose a two-step process of removing the component, then using a solder sucker to clear the mounting holes.  My technique worked very well.

Doing it for real then.  I had to remove the existing three capacitors.  That went pretty smoothly.  Step two was to clear the mounting holes.  This did not go well, at all.  I ended up with some solder in the holes and it would not come out.  I tried my usual technique.  I tried using wick.  I tried other people’s tips like adding more solder to pull the solder in the hole out.  Nothing was working.  It was probably about 45 mins of fighting with greater and greater desperation, eventually resulting in me damaging the board.  But there was a hole.

Tired and disappointed, I mounted and soldered the new capacitors in place.  The first one went well.  The other two did not, but I did get them in place.  Now it’s time to reassemble and see the results, if any.  I mounted the main board back in the case, laid the keyboard back, and brought the control panel back for connection.  Wiring up power and audio cables and flipped the switch.

It powered up.  I pressed some keys.  No sound.  Oh yeah, volume.  I had sound.  It sounded pretty good and I was feeling pretty good.  I powered up another keyboard to get a pitch comparison.  Hmm.  Slightly detuned.  I did a factory reset on the panel and tried again.  Now the pitch was off by an entire semitone.  Worse.  I held some notes and I could actually hear the pitch slowly changing -  up and down.  So, experiment unsuccessful.  Any further troubleshooting is out of my league.  I’ve literally watched the pros do it and I don’t understand what they’re doing and how they arrive at their ideas.  I don’t have an electrical engineer background.

So, to the garage for the keyboard for now.  I’m pretty sure it’s going to the dump.  I considered maybe parting it out on eBay, but what a hassle.  I already have a bunch of stuff I need to list and I can’t find the motivation to do it.  Maybe it will come in the next few days.  But for right now, I need to pack up my new tools of my failed hobby.

Farewell To Tweets

Since this is an unprecedented event in my time, I figured I’d at least record my thoughts on it to remember exactly what it was like.  I am referring to the sudden, rapid implosion of Twitter.  Since I’ve been a wordy motherfucker for decades, I obviously have no interest in Twitter.  It never suited my purposes and I never "got" what it was trying to sell.  So, this is clearly an outsider’s opinion.

Let’s start with my issues with the man behind the destruction.  Elon "This isn’t even my final form" Musk has been insufferable for years now and this is just the latest deed.  Fortunately, this is the one that pulls the curtain back on his actual lack of ability.  A spoiled brat falling upwards until he now seems to have reached the ceiling.  The only thing you can give him credit for is bankrolling other people’s ideas, like EV and space transportation.  I don’t buy for a minute all the people who say, "he’s a genius.  I’ve heard him talk and he knows his stuff."  He only has a skill of regurgitating other people’s knowledge, which is also a skill of a huckster.  He also has the self-important aura that makes him appear superior to others.  It’s no wonder he is an authoritarian, it’s his trajectory.

One of the biggest, biggest things that pisses me off about the Twitter problem is that it didn’t have to be a problem.  Everything Musk complains about is from his own doing.  Losing $4m/day?  It wasn’t before you got there.  Overstaffed, unproductive workers, company costs too high?  Wasn’t before you got there.  If Musk had just been a slightly better person and not tried to do some obvious market manipulation, resulting in him being forced to make good on an offer that was only supposed to make him richer, Twitter might still be around.

Next in line for gripes is the complete foolishness of Musk’s "management" style.   It’s not really management, it’s just barking orders.  The whole idea of, "I am the single source of guidance and direction" is impossibly stupid in an organization.  And as much as I hate to bring this other asshole into the conversation, it’s just like Trump being president.  Businesses and governments are built on a hierarchy for a very good reason.  It frees the people at the top from having to worry about the details, but authoritarians have to control every little detail.  And it sucks for everyone involved because there is no consistency and the second in command remains as clueless as the commoner.  Why even have a hierarchy then?

All of this superiority complex leads to the next point of stupidity.  Walking in on day one and firing the people in charge, then firing half the staff before you even understand how the company operates, then threatening the remaining people with double the workload and no additional incentive – still before you understand how the company runs – then, once a large number of those remaining people have bowed out, finally asking to be clued in as to how things work.  Any intelligent businessperson would spend months analyzing the system from the inside before making any changes.  Musk is lucky any of the other companies he bought survived his leadership and managed to stay on their original track.

I feel like I could go on, but I want to address the now and future of Twitter the service.

So, pre-Musk (PM), Twitter had a real problem with the quality of its userbase.  It had lots of harassment, incitement, and general bad behavior.  But so does every other social media site out there.  In that way, I am anti-social media in total.  I don’t think it has proven to be a good mechanism for communication.  The strengths it touts, allowing you to send off a quick message, as well as quickly reply in kind, are actually the wrong things to be promoting.  Spur-of-the-moment, off-the-cuff, spontaneous messages, spoken without consideration, as well as knee-jerk, impulsive responses, are not a conversation.  They are not anything but thoughts, and they lead to people doubling down and digging in on things they never should have said and can’t bring themselves to apologize for.  So again, quick messages are not good.

However, when it comes to news and alerts, quick messages are great.  And now a lot of governments and officials are wondering how they’re going to get the same effects after Twitter dies.  And again, I’m going to say, Twitter is not good for this use case either.  The problem I am focused on is that a lot of "alerts" are not internationally important or relevant.  The ones that people are worried about: active shooter, natural disaster, policy changes – these are all regional.  It does me no good to hear about an active shooter in CA when I’m across the country.  As best it’s a distraction.  And that’s the term I want to apply to Twitter broadly, it’s a distraction.  It causes you to concern yourself with things that are not something you can do anything about and are not time sensitive.  This is the problem the 24hr news cycle started and Twitter just turbocharged it.  So, I feel that governments are going to go back to the way they used to issue alerts, which were more regional. Journalists that cover those regions will subscribe to those alerts and will amplify the message appropriately. 

And I think what’s going to close up this post is the observation from someone who was there before the internet and seen how things got better and worse.  While the internet has been invaluable for accessing information that is more of a static nature, it has been more of a detriment for more transient information.  There’s lots of news that doesn’t need to be consumed right at the moment.  Even big news, like the Queen is dead, could wait for the evening.  That news doesn’t change what I am going to be doing for the day.  Again, it’s a distraction.  And I think the number of distractions we’re facing in a day is causing some serious societal harm.  I feel like I’ve written about this before, where if you read about 10 rapes in the news in a day, they feel like they’re all in your neighborhood.  The whole idea of being an interconnected world is not so appealing when you have to also bear the weight of the entire world’s problems.

It’s almost like we need some sort of hierarchical structure for news.

The Mail Must Go On

At some point this week, my communication system reached a tipping point.  Google had decided to block my mail server.  This is something that had been brewing for some time and it finally got to where I had to act.

My first experience with the mail issues occurred a while ago when someone asked if I would do some consulting for them.  I was on the fence about doing any side work, but replied and said we could talk about it.  Their mail server bounced my message back to me because my mail server was on a blacklist.  I contact the blacklist registry and appealed the block and they say it got removed.  That was fine, because I then never responded to the consulting requests again and it got me out of that situation.

Occasionally, I would have instances where I was told emails sent to me were rejected because my mail server was not considered trustworthy and looked like a spam server.  And the primary reason for this was that it looked like I was running a mail server on a dynamic IP address, which anyone can do and the dynamic nature helps spammers avoid detection and consequences.

But I don’t have a dynamic IP.  I have a static IP, and I need that to receive email reliably.  The issue at hand was that the reverse DNS for my IP address did not have my domain name and instead looked like a generic ISP host name.  Very sketchy.  And I knew this was the problem and ignored it for some time because it never really was that big of a problem.  Until it was.

So I contact Frontier, who is my ISP, for assistance.  I get on their help chat and make my request.  As I completely expected, they had no idea what I was talking about.  I got transferred to an "expert" support person who eventually told me, "Your request is not a level of support we can provide."  Now there’s a few ways to take a statement like that.  The knee-jerk reaction is "LET ME TALK TO YOUR MANAGER!"  I read more into that statement and took a more diplomatic approach. "Can that level of support be offered via phone?"  And the response was simply, "I do not know that."  Depending on your mood, this exchange could be read as sincere or as dismissive.  I chose the former.  I asked for the phone number for business support and got it (the number was wrong by one digit, but I figured it out), then I made the phone call.

I spoke to a tech pretty quickly.  I need to point out that the support in all of these cases was uncharacteristically quick – not much wait time at all.  He asked how he could help and I explained: "I need a reverse DNS entry created in your system that points my static IP to my domain.  Does any of that make sense to you?"  No, it didn’t.  But to his credit, he did not give up and say he couldn’t help.  Like many IT workers at many businesses, he’s working remote and has to use chat to communicate with his team and get assistance.  That was a slow process and he was not getting much help.  While we waited, I explained that while this might be an unusual request, it really isn’t when dealing with businesses and static IP circuits.  I started my IT career in ISPs, so I know about requests like these.

We were getting closer to giving up and I was sort of pressing to find out who the next level of support I needed to call would be.  He explained that the higher levels of support were all done by ticket systems and there wasn’t any number to call.  And then, he got a response from a lead support tech that provided the answer.  What I needed to do was sent an email to hostmaster with a request for my PTR records and it would be done.  That’s it.  I can do that.  I was actually surprised Frontier even maintained a hostmaster email, since such a standardized name would be a total target for spam and whatnot.  I thanked my tech and got off the phone.

I created a simple email to the hostmaster with my account info, my IP address information, and the PTR record with its current hostname and the desired host name.  I got an automated email within a few minutes that a ticket had been opened and would be processed in order.  The next day, I got a plain, simple email saying the record was created.  and that was it.  That problem was solved.

But, times have changed and email servers have many more tests they need to pass to be considered trustable.  A few things I had never heard of like SPF and DMARC records needed to be added to my DNS server.  And with those done, I guess I have to wait a couple of days for all the DNS changes to propagate and see if it had any improvement.  So, I hope I’m on my way to being a trusted source of email on the Internet.  Security never sleeps.

BTC FOMO WTF, I Dunno

There’s plenty of talk recently about bitcoin.  It’s something I’ve never understood, believed in, or trusted.  However, I feel it’s finally come time for me to at least have a conversational knowledge of it.  I don’t fully understand it from a technical perspective, because I do know enough about that part to retain my stance that I don’t believe in it or trust it.  What I want to be able to do is explain the (or a) process of using bitcoin.  Because I expect at some point, someone if going to ask me about it and how to get into it and when I say I don’t know, they’re going to think I’m stupid, because I’m supposed to be the all-knowing geek.

I start my quest with general searches on buying bitcoin.  Obviously, you need a place to store your stupid, fake money.  You can choose to have it stored on someone else’s website.  Yeah right.  I’ve been on the internet for a very long time.  You don’t trust the fucking internet for anything.  If not on someone’s website, you can store it in software on your computer or on a dedicated storage device.  This has a parallel to password vaults.  You can store your passwords in a vault online, like LastPass or you can store them in a file on your computer, like KeePass.  I chose KeePass, and I will choose the same for my bitcoin wallet.  Step 1 sort of complete.

I choose to install one of the better known wallet apps called Electrum.  I run through the default wallet setup, storing all the security information in KeePass in a symbiotic relationship.  Ok.  I’m ready to make a purchase now.  Gonna buy some fake money.

More searches on where to buy bitcoin.  I go to the first recommended place and start the process.  I’m immediately hit with a request for ID.  I have to submit a picture of an ID, either drivers license or passport, plus a picture of me holding the document.  Are you fucking kidding me?  What did I just say, you don’t trust the fucking internet.  We’re dealing with an unregulated product here, there’s nothing ensuring any security of any kind and you want me to give you a copy of my ID?  You can fuck right off.

Further research suggests that bitcoin is getting a little more legitimacy at least in the idea that it can be taxed by the IRS.  I don’t know if that’s a good thing or not.  Mostly I think it’s not.  If eliminating the anonymous aspect of bitcoin is the price of legitimacy, I don’t know.  So I look deeper.  I find there is a way to purchase bitcoin for cash using a special ATM machine, one of which is in my city.  That seems anonymous enough (although of course any agency that wanted to, could track me down with little problem).  I’m not trying to do this in the shadiest way possible.  I’m just trying to learn more about this concept and I don’t want to expose a bunch of my personal info to untrusted websites if I’m not going to be a devotee to the cause.

I watch a video on how to use the ATM and one thing I need is a QR code for an address to send my fake money to.  Electrum has a lot of different values in it.  I wanted to send the money to my wallet, so I went to Wallet Information and generated a QR code for my wallet ID.  That afternoon, I drove to the ATM and tried to buy some bitcoin.  Unfortunately, when I scanned my QR code, the machine said I had to use a supported wallet.  Step 2 failed.

Later, back at home, I think I generated a QR code for the wrong thing.  I though your wallet ID was unique and I’m sure it really is, but your wallet holds multiple addresses in it and each of those addresses are what you send and receive the bitcoin with.  The default view in Electrum didn’t show those addresses, but when I found it, things made a little more sense.  I generated a new QR code for one bitcoin address and I will attempt to use that.

Until I get back to the ATM, I figure I will try to buy some bitcoin online anonymously.  How about PayPal?  They’ve been making noise about supporting "Crypto" (The slick marketing term for this, I guess).  I quickly find out that any bitcoin you buy in PayPal can’t be transferred to your wallet.  So essentially, you have an online wallet with them.  I love you, PayPal, but no thanks.

I find another website that supposedly lets you buy without ID.  I create an account and get to the point of purchase.  They need a credit card number.  Well, here comes that mistrust again.  Not only that, but if I give my CC number, they’re going to hit me with a cash advance fee and interest.  Fuck that, too.

After a lot of puzzling over this, I came up with a solution.  Unsurprisingly or not, it’s PayPal.  I have my PayPal linked to a savings account for cash purchases.  That account is always kept at a low balance, so in case of compromise, I don’t lose all my cash.  PayPal allows you to make a virtual CC number to access the funds in any linked account, called a PayPal Key.  There’s my solution.  Now I’m ready to go.  I return to the bitcoin exchange and place my order.  It’s about $37 for me to learn this new concept.  And when I submit the form, I’m immediately told… I have to verify my identity.  God damn it.

So after a lot more searching and a bunch of other website visits, it doesn’t seem that I’m going to get very far without IDing myself, unless I want to pay a hefty premium for person-to-person trading.  Speaking of premiums, that is something about bitcoin that annoys the shit out of me.  Everything you do has a transaction fee.  It’s like having an account with a bank that has no ATM network.  You just get dinged the more you use it.  I guess people into this stuff just accept it as a cost of business.

I tried out a couple more sites and got stopped at the "provide ID" step.  I guess the ATM method is going to be my go-to method.  Looking at the ATM provider’s website, most of their machines are in sketchy locations – gas stations, vape shops, etc.  But, they do have some in a couple hotels, which I find surprising.  One is not too far from me, along a route I’ve travelled plenty of times.  So that’s going to be my next attempt.  I’ve figured out how to create a read-only copy of my wallet on my phone, so I can generate QR codes for any of the bitcoin addresses I have in my wallet.

I arrive at the hotel and find the bitcoin ATM next to the regular ATM in their lobby.  Using my mobile wallet, I set up a buy, stuffed in $40 (because in my previous online attempts, $20 wasn’t enough for a minimum purchase), and completed my purchase.  I immediately got a text message with my purchase confirmation.  Step 2 complete, I guess.

I went to a nearby convenience store and bought some snacks.  When I got back to my car, I opened my wallet and saw I had a new transaction in my history.  It said the transaction wasn’t confirmed, but, hey, it was there!  Of course, all it says is that I owned a tiny fraction of a bitcoin.  I went online and did some quick math.  It looks like $5 of the $40 purchase went to fees.  Holy shit.  But bitcoin is nothing if not the most volatile investment out there, so tomorrow I could be up $5 or down another $10, who the hell knows.

I drove back home and opened the wallet on my PC.  The transaction was there as well and now it was displayed as confirmed – I guess 15 servers could see that transaction and that was considered good.  Now that I owned bitcoin, I had to learn how to give it away.

My whole drive home I was mildly stewing about the $5 fee I paid to get my fake money.  And it made me wonder how things worked when I went to give some away.  Who pays?  And like I said earlier, it’s a racket.  Everyone wants paid.  I came to lean that even if I’m giving fake money away, I’m still paying someone to give it.  Hw much am I giving away?  Funny enough, the answer is, it depends.  How soon do you want your payment to go through, if at all?  The people facilitating the transactions work on the ones with the biggest fees first.  If things are really busy and they don’t get to your cheap-ass fee transaction in time, well, your transaction is cancelled.  And if not cancelled, you’ll wait potentially for days and your recipient is going to be beating down your door saying, "I want my two dollars!"

There are some interesting features that exist to help this situation.  One of which involves initially setting a low fee, then allowing changes to the transaction that are all fees.  So you can be cheap at first, then increase the fee if there are no takers in a reasonable time.  Another way to use that feature is to set a low fee initially, then let the recipient change the transaction to add any additional fee if they want the money quicker.  I don’t expect I’m ever going to be doing anything like this, but it’s kind of neat to know this is an option.

I contact a friend and we go through the setup of a new wallet and I perform a "Send" of about half my balance.  I chose a moderately low fee, but since everything in bitcoin is in a totally different scale, all you can do is make some rough estimates as to how much you’re losing in the trade.  So the transaction was made and it showed up on the other side almost immediately, but it remained unconfirmed.  I left it go overnight and in the morning, the transaction showed as confirmed.  Step 3 complete.

And that’s about all the more I care to experiment with bitcoin.  I spent $40 and I have $15 left in my wallet.  I’ve seen the process of receiving and the process of sending.  I’ve seen how much you lose in fees in the process.  Bitcoin is in a decline right now, so I’m probably losing value as well.  but I can now say that I can pay and be paid in bitcoin now.  That’s pretty much all I wanted.

Bring Out Your Dead

I have a problem.  Not really, but some would say, yes.  It’s my CD players.  I’ve discussed it little before and maybe joked about it.  It’s still kind of a joke.  here’s the continuation of that joke and my rationale.

So I have, let’s see.  Let me count them… 8.  Eight CD players.  Since the last time I mentioned this, I’ve added a JVC XL-V141, which is a 90’s player, and yesterday, a Yamaha CDX-520 from 1989.  This is the Yamaha:

IMG_20210424_102515

My collection all meet certain criteria, partially for my promotional purposes if I get to the point of selling them.  All have physical power buttons (Off is OFF), all have headphone jacks, so the youngin’s can enjoy them without needing a full stereo.  Most have volume knobs for the headphones, which is a excellent touch.  You see, when things started getting cheaper, features got cut.  This is back when CD players were premium devices.  And yeah, these were about $300 when new.  Back in those dollars, which is probably $600 today.

So, I had purchased this neat new player for $25.  I was eager to try it out and clean it up when i got home.  Right away, I plugged it in and gave it a smoke test on a CD and headphones.  Tray opens and closes, that’s good, the disc TOC reads, good.  Ohhh..  It has a chronic skipping problem.  It just stutters all over the track like it’s on fast-forward.  Damn it.

Pop the case off and look around.  I don’t really know what I’m looking for, just something out of place.  It all looks good.  I push a bunch of buttons and I notice that the CD doesn’t skip on later tracks, more on earlier tracks and chronically on the first track.  This is a clue, but I don’t know what it means yet.  This is pretty much what I had in front of me:

IMG_20210424_101235

While I’m reading my CD player repair document and looking over the player’s schematics from its service manual (which mean nothing to me), I go back and forth between the computer and the player.  Stopping and starting the player over and over.  Oddly, now it seems to be skipping less.  Then it’s not skipping on the first track anymore.  Is that all it was?  The player was tired and had been sitting too long, maybe just needed to warm back up?  What a crazy solution to the problem.

I tried a few more CDs.  Nope.  Skipping was still there.  So, I considered the problem wasn’t electronic, maybe mechanical.  Maybe the gears and rails for the laser transport needed cleaned and lubed.  I disassembled it and lubed up the moving parts with silicone lube on a swab (not recommended around electronics, but I’m stubborn).  Not any better, maybe worse.  Well, I’m going to have to level up on my repair skills.

The repair manual discussed adjusting the lens tracking and focus using potentiometers on the circuit board.  The troubleshooting guide also said tracking issues would cause skipping.  Ok, let’s do this.  $25 already gone, right? 

I took a picture of the pots before I fiddled with them just to be absolutely sure.

IMG_20210423_174910

While a CD played, I slowly twisted the left pot and the skipping got worse.  Well, that’s a change in some direction, so I felt I was on the right track.  However, I couldn’t get things to get any better.  In fact, I experienced something the the guide warned about, with the CD spinning out of control at very high speed.  So, after powering the player down and resetting the pot to neutral, let’s try the other one.  In a couple small changes, suddenly the transport quieted down and didn’t seem to rattle anymore.  The skipping stopped!  This is the setting I ended up with:

IMG_20210424_101116

It’s not a major adjustment, but I imagine it’s not supposed to be.  I’ve been testing out the player on multiple CDs and no more skipping issue at all.

So, my little joke of a collecting problem has given me a new repair skill.  Formerly, my repair abilities were limited to changing belts, which honestly is pretty good and has brought a couple players back from the dead.  Now I have a new solution for a new problem under my belt.  I can fix even more players now.  A good hobby is one where you continually grow, right?

Welcome To My House

You are in my house right now.  What do you think?

Over the weekend, I migrated my blog from my hosted server with GoDaddy to a server sitting in my guest bedroom.  The process should have been easier than I made it out to be and that was just because of a technical detail that took a little bit to remember.

When you have a web server that is accessible from inside your network and outside your network, you think would want to make the connections to that server consistent, just to simplify everything.  So, you do the port forwarding on your router to direct web traffic to your web server and you point the DNS to the external IP of your router.  Everything goes through the router.

This is how it works for the outside world, but it doesn’t work the same for your internal network.  It’s not only inefficient, but practically non-functional for an internal computer to route to an external address only to come back around and get routed to an internal address.  It’s like going to the bathroom by going outside, then back in through the front door every time instead of just walking across the house.

To solve this, you need the concept of external DNS and internal DNS.  If you are inside the network, your computer resolves the addresses as internal IP addresses, and if you’re outside the network, you get external IP addresses.  Because I use a pi-hole as my DNS and DHCP server internally, this meant I had to modify the HOSTS file on the pi-hole to override any DNS requests for my web sites I hosted inside my network.

Now that my web server is up and running, I still have one more month to migrate my email server off the old server to my house.  Then my GoDaddy server will expire and I’ll be all on my own.

Frontier has been less than stellar during the migration, but the mistakes weren’t too egregious.  The first issue happened a week before my new install.  I woke up and my internet was off.  I called up tech support and tried to resolve the issue.  After a little while of no progress, I mentioned that I suspected this had to do with my new service order and the disconnection of the old service.

That led the conversion in a different path, which ended up in Accounts, who said everything is fine, and then was sent back to tech support who finally put the pieces together and determined that when my new circuit was provisioned, the old one was deactivated.  Although I couldn’t have predicted that exactly, I had the sales rep put notes in the order saying that the old circuit was to remain active until the new one was installed.  No one reads notes.  I should have known that.  After two hours on the phone, I had Internet again.

The day of the install, I woke up and I had no internet.  Of course.  This one I blamed myself for.  I had an opportunity to set my deactivation date for a day later than the new install date and I didn’t take it.  Of course the note saying not to turn off the old circuit until the new one was live doesn’t matter to an automated system that says, deactivation date is xx/xx/xxxx.  So I just went without Internet until the tech showed up.

The install went fairly smoothly.  At first, I was happy that I got to keep my old router, which meant I didn’t have to do any reconfiguration, but our initial tests kept showing my upload speed capped at 100mbps instead of the 500 I was supposed to get.  We eventually swapped the router out for a new one and that solved the problem.  I was surprised that the tech said upload speed is managed by the router.  Makes me think that could be hacked in some way.

And in the end, I did have to reconfigure the router and of course, at one point I had two DHCP servers running on the network, which screwed up my IP cameras, and my pi-hole was bypassed, so I got all the browser ads.  But eventually it all came back into order and things were normal again.

For what it’s worth, I honestly can’t tell the difference between 100 and 500mbps.  I haven’t really done a lot of downloading and I don’t do a lot of video streaming, and I’m only one person, so maybe I wouldn’t be able to tell.  But the speed is supposedly there.

Bringing It All In House

Last December, I made a decision to start becoming more self-reliant and not utilizing free online services as much.  To accomplish that, I moved my blog off of WordPress and onto my own hosted server.  This year, it looks like I’m going to go a step further and be completely on my own.  It’s a huge risk, but it comes with some benefits I just can’t afford any other way.

At one time, I had my email hosted through some web hosting provider.  It was ok, but I didn’t have a lot of the flexibility I wanted.  And at that time, I also had a simple web site hosted at the same provider.  I made the radical decision to change from a simple hosting plan to a virtual server.  The virtual server would let me install anything I wanted on it.  I installed a mail server.  I installed a web server.  And later, I installed WordPress and things have been going pretty smoothly. 

What were my risks back then?  Mostly hacking worries.  But, I’ve been pretty good.  I had one instance where the mail server got compromised due to my lack of cleanup of development accounts, but otherwise, no issues.

The consideration this year is to bring the entire server from the virtual to the physical and keep it not in a massive data center, but in my house.  By many accounts, this is a pretty bad idea.

To start, a data center has massive bandwidth and multiple, redundant internet connections.  The downtime is going to be minimal at best.  unnoticeable in reality.  Second, the server hardware is going to be highly redundant and isn’t going to go down either.  The server is virtual.  If the hardware fails, it just activates on new hardware.  And you don’t have to worry about it.  No hard drive failures (they’re part of a massive drive pool), no power supply failures, no UPS failures.  No worries about patches (they’re automatically applied).  Why would I give that up?

What am I sacrificing for this security and reliability?  Well, I’m locked into a specific server.  It has a fixed CPU, fixed RAM, and fixed hard drive size (and I just noticed today, fixed bandwidth).  Those are listed in increasing importance to me.  Right now, I have a project that I want to take public.  My current hosted server has 2GB of RAM and 60GB of drive space total.  That also includes the operating system.  The project I want to release has a data size of 1.5TB and is constantly growing.  I can’t even get a virtual server with that amount of space.  I would have to have a dedicated server, which would run over $500/mo.  And I would have to fully manage it – remotely.  Hard drive failure?  Call someone in CA to visit the data center and swap the drive.  It’s not reasonable.  So again, my plan is to bring the server into my house, where I can maintain it and upgrade it as needed and it can serve the world.

Today, I called Frontier and asked about their Business line of FIOS products.  After all, this is going to be a hosted server.  This is not a residential setup (although I could kind of get away with it using dynamic DNS, which is hokey AF).  I had some questions and I got some answers and the answers seem to indicate that I am going to be able to do this.

First question, do you have to be a business to get Business FIOS?  Yes.  Ok.  So I have to set up an LLC for myself.  I’ve been through this before.  I don’t exactly like it, but maybe it’s for the best.  Maybe I’ll start doing consulting again.

Last question, how much does it cost?  This is important, because Frontier’s website only shows the promotional prices.  $50/mo for 100mbps and $90/mo for 500mbps.  And the numbers for what they call month-to-month aren’t that bad.  I’m focusing on 500mbps and that’s going to run around $125/mo.

Is $125/mo a lot?  Considering some people pay that much so they can have all the cable channels with sports and movies, I don’t think so.  Is it a lot for me?  It would be, except…  I pay $75/mo for my 100mbps FIOS now.  I pay $480/yr for my virtual server.  Add all that up and do some math and that’s $115/mo I’m paying for my internet needs.  An extra $10/mo to get 500mbps and full control of a server where I can have TB’s of data online?  I think it’s a fair deal.

My hosting will expire 11/4, so I have a couple of months to get prepped for the change.  I need to buy another server and set it up.  I need to make some more improvements to my project.  I need to plan to change my DNS.  Migrate my mail, export and reimport my WordPress stuff.  It would be a busy week or so of work.

And once that’s done, I’ll be completely on my own.  And what’s the scariest part of that?  If my internet goes down, or I move, or I die (well, if I die, it’s my survivor’s problem), there’s no more email.  That is a critical service that I should think hard about.  But again, I can’t get the features I want without self-hosting it.  The old saying, hope for the best, plan for the worst means you have to always think about the worst.  That’s hard.