John Bowdre

By day, I manage a large virtualized server environment, with a focus on cloud and automation technol...  
Read more

John's Collections

2021

Sep 21, 2021
Sep 21, 2021
Patched a VMware environment
Installed critical security updates
VMware published a pretty large vCenter security advisory today, VMSA-2021-0020. This includes a number of nasty vulnerabilities, including one (CVE-2021-22005) bearing a CVSSv3 base score of 9.8: 

A malicious actor with network access to port 443 on vCenter Server may exploit this issue to execute code on vCenter Server by uploading a specially crafted file.

YIKES.

Anyway, I've started patching my vCenter environments. You probably should do yours too.
Sep 18, 2021
Sep 18, 2021
Competed in an autocross event
It was our fourth event of the season, and the forecast called for storms all day. As a result, I didn't bother with swapping on my race tires (Bridgestone RE71s) and I instead just went with my daily-driver all seasons tires (Continental DWS06s). 

I felt pretty good with that decision during my first run, as the course got hammered with pretty heavy rain:
https://www.youtube.com/watch?v=AzzdcW7SsJ0

Of course, things cleared up a bit as the day went on and the course was mostly dry by my last run, just a few puddles and slick spots to worry about:
https://www.youtube.com/watch?v=ZgmlZOr_xWM

I had forgotten how talkative my street tires are - the Bridgestones typically put up much less of a fuss!
Sep 11, 2021
Sep 11, 2021
Worked on a project car
Learned something
Worked on an air-cooled VW
+ 1
It's been a fairly productive Ghia week!

  • Remember how I don't really know what I'm doing? That was confirmed this week when I realized that I'd basically been doing everything wrong. For some reason, I had (incorrectly) assumed that the single notch on the rear portion of my crank pulley must mark Top Dead Center, when the piston is fully extended inside the cylinder; I then read something that suggested the notch instead indicated 7.5 degrees Before Top Dead Center. To check this, I removed the spark plug from cylinder 1, stuck a plastic straw inside as a feeler, and slowly turned the engine with a 21mm wrench applied to the alternator pulley. That let me feel when the piston stopped moving... and, sure enough, it wasn't when the notch was lined up with the crack on the case. In fact, this position matched with the hard-to-see dimple on the front portion of the pulley. Good to know! 
  • Of course, that also meant that my previous valve adjustments had been performed incorrectly; those are supposed to be done with each piston in its TDC position - not 7.5 degrees before. So I re-did those adjustments, and it only took about fifteen minutes this time! It didn't make a big difference, but there were a couple of cylinders that had just a wee bit (like 0.008" instead of 0.006") too much valve gap. 
  • And I think I've finally figured out what I'm doing with regards to timing the (aftermarket) 009 distributor which was installed in the car at some point. All the specs I found said to time it to ~10 degrees BTDC at idle. Since the 009 isn't equipped with vacuum advance (it just centrifugally advances the timing as it spins faster), it's actually way more important to set the peak advance to about 32 degrees at around 3500 rpm; that should drop to 5-10 degrees BTDC at idle, but the exact timing isn't as important here. I printed out a timing wheel and used that to mark the relevant points around the rim of the crank pulley (TDC, a range covering 5-10 degrees BTDC, and a hash to mark the 32 degree maximum advance).
  • And I discovered that the Innova 3340 tach/dwell (many other functions) meter I'm using seems to do a much better job of reading the RPMs when I use the inductive pickup on the cylinder 1 spark plug wire and put the meter in indistinctly-named "CON" mode rather than attaching the contact probe to the negative terminal on the coil and setting the meter to "RPM" mode. It had previous been pretty tough to read, and often indicated that the car's idle was set way too high. The measurements are much more consistent (and reasonable) when using that inductive pickup.
  • So I combined all of this knowledge to actually get the timing set correctly (finally!), with a maximum of about 30 degrees advance at 3500 rpm and a comfortable idle of 7ish degrees BTDC at ~850 rpm. 
  • And I used the improved RPM measurements to also tweak the carburetor settings a bit, per the Bentley manual: I used the big bypass screw to set the idle to about 850 rpm, turned the smaller volume adjustment screw until the engine ran at its fastest and kept turning clockwise until it dropped off ~30 rpm, and finally used the big bypass screw to reset the idle at about 850. So maybe the car is now tuned correctly?
  • I had intended to drive the Ghia to work on Thursday since the weather was going to be nice, but my pre-flight check that morning revealed that the brake lights weren't working anymore. That evening, I jiggled some wires and got them working again... hardly a proper fix, but enough to take the Ghia to work on Friday - which I did! It drove like a dream.
  • Today, my wife and I spent a few hours cleaning up the electrical connections for the front and rear signal lights. I was a bit surprised to find that each bulb only had a single wire running to it, relying on the light housing itself to provide the connection to ground and complete the circuit. We carefully cleaned the corrosion and grime off of the wiring connections and thoroughly cleaned the bulb sockets as well. Once everything was put back together, all the lights worked again (yay), and some of them even glowed more brightly than before. Hopefully this cleanup will help keep the current flowing better in the future. We plan to revisit the other end of those wires (at the fuse box behind the dashboard) soon.
I feel a bit silly for having been doing things fairly wrong so far, but I'm glad to have finally figured things out. I learned a lot this week, and the Ghia is running beautifully. I think we'll take the Ghia out for more joy rides and trips across town now that the weather is (maybe?) cooling off a bit and we're getting a bit more comfortable and confident with the car's transport abilities. 
Sep 05, 2021
Sep 05, 2021
Worked on a project car
Worked on an air-cooled VW
Now I've got a drain plug!

I noticed that the Ghia had been leaking a bit of oil in the past couple of weeks, probably because I hadn't been adequately prepared for that first tuneup and had reused the existing gaskets for the oil screen and the sump plate. I ordered several sets of gaskets, and also picked up an aftermarket sump plate with a built-in drain plug. That should make future oil changes much easier!

Plus it's shiny so that's +5hp.

I also did a bit of fiddling with the timing and carburetor settings. I realized that my car doesn't have the factory-correct single-vacuum double-advance (SVDA) distributor and instead has an aftermarket Bosch 009 unit with no vacuum advance (centrifugal only). The specs say that I should set the timing for that to ~10BTDC, but the motor really didn't seem to like that. No amount of adjustment on the carburetor would get it to idle correctly. I kicked the timing back to the ~5BTDC I was using previously and I was then able to adjust the idle with the bypass screw and the mixture with the volume screw. It seems happier now.

I'm not sure why the timing would need to be set so far off from what everything I can find suggests the 009 should be, but of course I also don't know what else might not be factory-stock about this motor. 

But hey, I'm learning stuff and enjoying the car along the way!
Sep 03, 2021
Sep 03, 2021
Wrote an article
Used vRealize Automation/Orchestrator
Automated a server deployment
Used PowerShell
Used PowerCLI
+ 3
Running in-guest scripts on VMs from vRealize Automation

Once I sorted out how to use vRA to deploy VMs with specified virtual hardware configurations, the next big step was to perform actions inside the guest OS to further customize the deployment. I experimented with a few different approaches of varying complexity, but have (for the most part) settled on using vRA's Action-Based eXtensibility (ABX) component. This lets me write PowerShell code to connect to a vCenter through PowerCLI and leverages the VMware Tools connection to gain access inside the guest. From there, I can pass in whatever commands I want.

I wrote a post to document how I'm using ABX to expand a server's C: volume to fill the underlying virtual disk, add specified domain accounts as administrators on the system, configure the firewall, and more: 

https://virtuallypotato.com/run-scripts-in-guest-os-with-vra-abx-actions
Aug 28, 2021
Aug 28, 2021
Worked on a project car
Worked on an air-cooled VW
Another lovely weekend day, another couple of hours of wrenching on our 1974 VW Karmann Ghia...

  • We rechecked the toe alignment on the front end, and found that I'm a dumb-dumb and my previous alignment efforts hadn't been accurate since the suspension wasn't loaded. Rather than using a laser pointer held against the hubs, this time I used a pair of yardsticks held up between the inner edges of the rims stretching from one side of the car to the other. I marked where the sticks overlapped with the measurement taken between the front edge of each wheel and again between the rear edge. This revealed the car was actually sitting with about 1/2" toe out, which likely explains why I was having to fight so much to keep it centered on the road. I eventually got it dialed in to about 1/8" toe in.
  • I had to raise the front end again in order to get at the lock nuts on the inner tie rod ends, and along the way I noticed that the passenger side wheel had a little bit of side-to-side play in it. I removed the cap on the center of the hub and tightened the wheel bearing adjustment nut to eliminate the slop. I did the same on the driver side for good measure. 
  • I then lifted up the rear of the car to see about adjusting the drum brakes, which I suspected weren't releasing all the way. It turns out I was right: even with both wheels off the ground, I couldn't turn the left rear wheel on my own. I used a flat-head screw driver to spin the shoe adjustment star wheel things and got the wheel to rotate more freely. I dialed them back down until the wheel started getting some resistance, and backed off just enough to spin freely again. I repeated this for the right drum as well.
  • While I was under the rear of the car, I also identified a disconnected wire that had been dangling beneath the car (and driving me nuts every time I saw it). It's the starter end of the stock ignition circuit - so there's another explanation for why I can't start it up by turning the key and must instead press the button behind the passenger seat. I had hoped that I could just reconnect it to the starter and get back in business (and even confirmed that the wire gets +12VDC when the key is turned to the "start" position) but it still wasn't doing the trick. I reconnected the bypass circuit between the battery, the button, and the starter, and at least used a wire tie to keep the stock wire from dangling in the future.
  • My wife also spent some time removing the black vinyl wrap a previous owner had used to cover up the chrome trim pieces on both sides of the car. The chrome underneath is a little scuffed, but we both think that the lines look a lot better (even if some of the lines aren't completely straight...).
We went for a few short drives along the way to check our progress. The steering feels MUCH better; no more weird oscillation when making minor inputs. Brakes also feel better; the car stops well and it doesn't feel like the brakes are still applied when I lift off the pedal. 

All told we put another 15 miles on the car today, and I'm really enjoying the way it drives now. I can't wait to drive it to work again - once it cools off a bit more.
Get your own page like this