This past weekend my media NAS server was intolerably slow. When I investigated, I found out that one of the RAID-1 partitions is experiencing read errors and is timing out. I decided to risk a reboot and to my surprise the RAID-1 partition did not recover with one fail drive, but mdstat recorded with an inactive status, something like this:
md2 : inactive sdc1[0](S)
After some Google search, I found that I had to do the following to resurrect the md2 device.
This reactivated the md2 partition. I replaced the failed drive and re-added the new drive to the md2 device. The RAID-1 partition is now rebuilding.
The inactive state is a new experience for me, so this was a bit of a surprise.
During this exercise I also found out that the SATA connectors on my SATA add-on card were loose causing intermittent connections. I will have to find a way to address this in the future.
This morning on the way to work after picking up breakfast from McDonald’s (the one on 9989 BAYVIEW), I decided that heading East bound on Major Mackenzie from Bayview to Leslie was too busy, so I decided to turn right onto Boake Trail from Major Mackenzie.
After my right turn I was stopped immediately by a police officer. He told me that I was not allowed to turn right.
Since I had a dashcam installed on my car, I replayed the footage and found that there was simply no sign indicating that a right turn is restricted. The matter was even more confusing when there is a dedicated right turn lane. See for yourself below.
Dashcam Video
Also on Google Maps, the no right turn sign is also non-existent.
I decided to contest this matter. I am providing this quick blog entry as near to a contemporaneous note as possible of the event this morning in case I need to refresh my memory three months from now in court.
Update: My sharp eye wife spotted the no-left turn sign beside the traffic lights. So even with the video and Google Map reference I missed it on multiple, post-event analysis. The dedicated right turn lane threw me off and baited me into the turn, an honest mistake.
I am in the process of building and configuring a media server for my parents. After my recent media server upgrade, I have extra gear lying around. By purchasing a power supply and a small case, I can cobble together another media server with my old processor and motherboard. I will call this my parent’s media server. The goal is to replace the current Raspberry PI unit that is currently running OSMC acting as their media server. Although the OSMC solution with Raspberry PI has been working really well, it is under powered to play any HEVC encoded video at full 1080p HD resolution.
I wanted to convert the majority of our video media to HEVC simply to save storage space. If I do this with my media library, I will not be able to share our media with them because of their under powered Raspberry PI.
To solve this issue, I installed Ubuntu 18.04 along with Kodi on my parent’s media server that I just created. I have been testing this solution for the past couple of weeks and both the hardware and media player works really well.
I also configured the box to auto mount USB disks, and installed SAMBA so that both videos and music files can be shared with other devices on the same network. The SAMBA is primarily used by my parents with their SONOS speakers.
With this media server at their location, I can also consider future upgrades such as replacing their WiFi network with a Ubiquiti solution, and even ponder on a site-to-site VPN solution with both of our networks.
Perhaps that is looking too far into the future. My immediate concern is how to remotely administer the box. With the Raspberry PI, I just had a simple SSH setup. However with the extra horse power, and a full blown Ubuntu distribution, I can now setup OpenVPN.
I followed these instructions on the DigitalOcean site, and it worked flawlessly. During the setup, I made a major error. I skipped the firewall (ufw) setup on the box, thinking that I don’t need a firewall because an external firewall already exists. However, OpenVPN will not route external traffic to the internal private network if IP masquerading (NAT) is not setup properly. Thanks to a coworker’s advice, I configured the firewall with IP forwarding NAT, but also change all default actions to ACCEPT so that the firewall only function as a NAT router. Lesson learned!
Since this VPN will only be used by me for remote management, I will not configure any HTTPS tunnelling or install and configure ObfsProxy. We will continue to use UDP and stick with the default 1194 port.
We will do some final testing before finally deploying it to my parent’s place.
I have been a huge fan of Apple’s fusion drives. They are an excellent compromise for affordable mass storage while still able to give you SSD performance. The concept is simple pair a fast but small SSD drive with a large but slow and much affordable, mechanical HDD. You get good performance and have lots of storage without breaking the bank.
I have falsely assumed that this capability only existed with Apple’s macOS operating system. This week I was pleasantly surprised to have discovered that LVM Cache can do more or less the same thing on Linux. This new found knowledge along with an excellent deal on a 500GB NVMe Samsung 970 Evo Plus M.2 drive gave me the itch to experiment this weekend with my NAS media server.
The hardware was easy enough to install, but I had to move one of the existing SATA connection because the M.2 slot on the motherboard shared a PCIe bus with a pair of SATA connections. Luckily I bothered to check the motherboard manual, otherwise I would have been scratching my head while the server fail to boot.
The software configurations were a bit more involved. Before I purchased the NVMe card, I did some experimentation with two external USB drives, one SSD and one HDD. I found this article to be super helpful in configuring LVM Cache with my test drives. However, these configurations were not fully restored after a reboot. After many hours of research on the Internet, I found this article indicating that my Ubuntu Linux distribution was missing the thin-provisioning-tools package. I also had experimented between the two different cache modes that were available, writethrough and writeback. I found out that the write back mode was a bit buggy and did not sync the cache and the storage drive. Yet another article to the rescue.
I preferred the write back mode due to its better write performance characteristics. Apparently to fix the issue, I have to increase the migration threshold to something larger than the default of 2048 because the chunk size was too large.
Here are the steps that I did to configure my existing logical volume (airvideovg2/airvideo) to be cached by the NVMe drive that I just purchased. I first have to partitioned the NVMe drive.
Model: Samsung SSD 970 EVO Plus 500GB (nvme) Disk /dev/nvme0n1: 500GB Sector size (logical/physical): 512B/512B Partition Table: gpt Disk Flags: Number Start End Size File system Name Flags 1 1049kB 500GB 500GB primary
Create an LVM physical volume with the NVMe partition that was created previously /dev/nvme0n1p1 and add it to the existing airvideovg2 volume group.
Now my original logical volume is cached and I have gained SSD performance economically on my 20TB RAID setup for less than $200. Below is my final volume listing.
Two and half years ago, I performed a CPU and motherboard upgrade to my media server. You can read the account here.
Although the AMD Athlon 5350 APU was energy efficient, it proved to be under power for on demand video encoding when Plex wanted to transcode video for a player on a device that is not compatible with the playing video. For example, when an Apple TV (not 4K) wants to play 4K material from Plex on my media server, the server will have to transcode the 4K material to a compatible 1080p format. Unfortunately, this is very CPU intensive and if more than one person in the house hold is trying to do the same thing, which is not unheard of, this causes stuttered playback issues.
Given the choice between saving a few dollars a year versus usability, I choose usability. Therefore I started to research what I need for the upgrade. My goal is upgrade the system so that transcoding will not be an issue and I can also use the system for future video encoding of security camera footages. We can also use the system for background video encoding of family videos as well.
I continue to prefer the AMD brand, and decided on the following combo:
Corsair Vengeance LPX 16GB (2x8GB) DDR4 DRAM 2666MHz (CMK16GX4M2A2666C16)
The above were all purchased through Amazon and cost me a grand total of $473.24. The AMD CPU was the most expensive part costing almost $190.
Taking out the old motherboard and CPU combo and replacing them with the new parts went smoothly. The side SATA connectors bucked against one of my HDD chassis so I opted not to use them, and decided to connect all of my RAID SATA connectors to the SATA accessory card that I purchased and discussed in this post.
Last time I did an upgrade like this, the Ubuntu operating system had no problems and booted without any issues. Unfortunately, this time is very different. After the machine posted, Ubuntu booted into a blank, black screen. After some research, I learned to reboot the Ubuntu kernel with the nomodeset option. I learned to press and hold the shift key so that I can select the desired kernel that I wanted via the GRUB menu, and I learned to press the ‘e’ key in the GRUB menu to modify the boot options. Finally pressing F10 to boot with the custom changes (effective for only one time).
The above trick got me a login prompt. After I gained access to the command prompt, I noticed that the kernel did not recognize any ethernet devices. I now have a machine that is not connected to the network. After some more Internet research I found out that the current 4.15 Linux Kernel that I have is insufficient to run on the Raven Ridge architecture, the AMD code name for the Zen CPU and Vega GPU combination on a single chip. I have to upgrade to the 4.18 Linux Kernel.
However I cannot upgrade through the Internet, because the machine is not on the Internet. I have to download the Debian packages on a USB stick with another machine and manually install them. At this point, I learned that you cannot simply download a single package for this. I had to decide whether to go with the Linux Mainline Kernel packages or go with the Ubuntu HWE (Hardware Enablement) packages. After reading through Ubuntu’s LTS Enablement Stack article, I decided to HWE packages. I found the linux-generic-hwe packages and their prerequisites on pkgs.org. This took several iterations as I did not get all the dependent packages on the first try.
Once all the packages were installed, the machine booted without the need for the nomodeset option. However, the internet interface device was still not there. I had to run the command netpath, to find out that new motherboard’s ethernet device’s logical name was em1. To register the new logical name, I had to edit /etc/network/interfaces file.
Finally, the machine booted with an active ethernet connection. As a sanity check, I executed:
Ensuring that my new media server has all the required kernel packages. We are still not done. The IP address of the server has changed, because we now have a different MAC address, so the DHCP server provisioned a different IP. I tried to change the Unifi Controller to provision a static IP address to this new server but I was unsuccessful. I suspect that the new server is also running the Unifi Controller may have something to do with it. Since the IP address has changed, I needed to update the following configurations:
Firewall rules
Unifi Controller name space configurations
Samba configurations because we only allow for local machines to share
All of this took from 4:30pm to 11:00pm last night, 6.5 hours worth of hardware assembly, research with Google, trial and error, and finally success. I cannot imagine if Google and the super helpful community forums did not existed. Fingers crossed that the new media server will run smoothly.
This past weekend I continued to upgrade our NAS server. Last weekend, I upgraded my raid array with an additional 8TB of mirrored storage. This yielded two old 4TB WD Blue HDD. I noticed that my case has a total of 9 internal storage bays. One was used by my 500GB SSD Boot Drive, and 6 were populated by HDD drives making up the current raid array. This means I have 2 more storage bays left. However these remaining bays were meant for 5.25″ storage devices like Optical Disc Players. For me to place my old 4TB WD Blue HDD into these bays, I will need a 5.25″ to 3.5″ bay converter. I had one, and purchased the other one on Amazon. I ended up buying the ORICO Aluminum 5.25 inch to 2.5 or 3.5 Inch Internal Hard Disk Drive Mounting Kit.
I also did not have enough SATA slots and purchased the IOCrest SI-PEX40071 SATA III 8 Port Controller Card. This card along with the 4 builtin SATA slots on the motherboard gave me enough SATA connections for my 9 drives.
Once I installed the old 4TB drives, I proceeded to create another md raid level 1 device and created a matching physical volume which I used to extend the current logical volume group. When the setup is completed, I ended up with a 20TB+ fully mirrored NAS server. I love LVM in combination of mdadm.
I figured while everything is fresh on my mind, I minus well proceed with the dreaded 16.04 to 18.04 Linux Ubuntu upgrade.
The upgrade was surprisingly very smooth. However the new version of OpenVPN caused some troubles. The new OpenVPN no longer works with my old PureVPN configuration files, because the certificate files that came from PureVPN used an outdated and deprecated hash algorithm. After getting the new configuration files from PureVPN, everything worked like a charm.
I also have to reinstall the Unifi Controller along with Let’s Encrypt certbot utility.
Super happy with the outcome and the upgrades should last another 2 to 3 years.
Our NAS has run out of space again. I saw a deal that the Seagate IronWolf 8TB NAS Hard Drive was on sale at newegg for $309 CAD. I jumped at the chance and purchased two.
I am now following the same step as I outlined in this post. Replacing two old 4TB drives with these two new 8TB drives.
So far so good. Hopefully when all is said and done, my NAS will have a total of 18TB in a RAID 1 configuration of 6 hard drives in total. Two 4TB, two 6TB, and the two new 8TB.
I noticed that I could fit two more drives in my chassis and may decide to re-add the two old 4 TB back in, but first I’ll have to check if my power supply can handle the demand.
I really like this mdadm and LVM setup.
Update: After 2 mdadm syncs, each of which was around 8 hours, and a pvresize that also took another 5 hours. I had to convert the filesystem from 32 bits to 64 bits using these very helpful instructions. Only after I converted to 64 bits can I then expand the existing filesystem to more than 16TB. It was a learning and yet rewarding experience. Next step is to reuse the 2 old 4TB drives in the same chassis and add them to the logical volume.
This week I came a cross a bike computer article and saw this picture. A Wahoo Elemnt Bolt with a gear selection display. Raw emotions took over and I immediately purchased one from MEC.
I longed for a bike computer that can show my current gear selection so that I can avoid looking down and see which gear I am at, or finding out whether I am on the small or the big ring. When I got the SRAM Red eTap two years ago, I totally forgot that it has the ability to communicate its status with a compatible head unit. I was totally okay to continue to use my Garmin Edge 500 until it dies.
The new Bolt unit arrived yesterday. After downloading the companion app on my iPhone it was a breeze to setup and I customized the data pages that I wanted during the ride. It was also super simple to pair the Bolt with my Heart Rate monitor, Power meter, and the SRAM Red eTap.
Garmin Edge 500
If you are a data head, then you will love the Bolt. It has more metrics than you know what to do with. I particularly like that fact that it sync’s with my iPhone and when my ride is done it automatically uploads my ride to Strava.
I also allowed the unit to upload health related stats to my iOS Health App. But to my disappointment, it seems that it doesn’t quite sync the calories count. Instead, the Health App got the calories reading from my Apple Watch which has a very skewed reading because I did not turn on the Workout App. I think for my next ride, I will use the Workout App so that the Health App will get a more accurate calories count, and I will turn off the Health App upload on the Bolt to avoid a duplicate recording. I may change my mind later as I am still experimenting which sync’ing technique is best. This is a first world problem, when you have too many health gadgets to manage at once.
As an added bonus, I don’t need to reconfigure my head unit whenever I decide to ride indoors. I can keep my old Garmin Edge 500 to be my indoor bike computer and permanently affix it to my Cervelo R5.
There are many types of cycling, and each type of cycling comes with their own types of bikes and enjoyment. This web page from CenturyCycles.com does an excellent break down. I love road cycling primarily because of its team work, speed, and relative simplicity in comparison to other forms of cycling. Of course different strokes for different folks. However on this post I am only going to discuss about road cycling since this is the basis of my own personal experience. Here are some fun times which I enjoyed with our local chain gangs.
Getting a bike
Before getting a bike, you will need to know about your own personal dimensions, such as your height, inseam, foot size, etc. These measurements are critical for determining the size of the bike that will work best for you. Different brands can have different sizing metrics. Here is a more in depth article at LiveStrong.com discussing fitting to a Cannondale bike. Below is a table selecting a Trek road bike frame size by your height.
A proper fit is important as you can potentially be riding your bike for hours at a time. An improper fit will result in discomfort and potentially repetitive, body injuries.
Once you know the type and size of the bike that you want to get, you will now have to decide whether to get something new or used. Stay away from buying bikes at major merchandising stores like Costco, Walmart, and Canadian Tires. Those bikes have inferior components and they will not work well and will break within a few rides.
Budget may be an issue here. Be prepare to spend $1500 to $3000 for either a high end Aluminum frame bike or a low end Carbon bike. As a sample, here is a sample link to a local bike store at D’Ornellas listing road bikes that are less than $2000 in price.
Alternatively, you can tap into the local used market. Here is an example from Facebook, a $1200 2015 Specialized Allez Comp. Kijiji is another popular site for locally used road bikes. I personally stay away from eBay, but you may find that more convenient.
I think both Aluminum or Carbon frames have their merits, and I would stay with at least a Shimano 105 group set or higher. The group set is the collection of components on the bike that relates to its transmission system, which is everything from brakes, chain rings, gear shifting, head sets, etc. Here is a quick guide to the different hierarchy of the Shimano road group set from Road.cc.
Pedals and shoes
Shimano Ultegra Pedal
For road cycling all of us use clipless pedals. There are two main types, the road racing (on left) and mountain biking style pedals. I personally use a pair of Shimano Ultegra Pedals. Pedal selection can be an exercise in finding religion. Instead of me going into details here, I will refer to this wonderful article from the New York City Bike Shop.
The only thing I would stress is that going clipless is a must if you want to keep up with a group. Your choice of the clipless system is really up to you.
Get kitted up
Cycling requires proper attire, primarily for safety, comfort and temperature control. Of course different weather will require different kits (the type of clothing you wear). Things to consider are:
Helmets
Cycling Gloves
Cycling Jersey
Cycling Bibs
Socks
If you decide to ride in the cold, then different linings, jackets, and long tights will also have to be considered.
When considering helmets, you want one that is locally certified for safety. Everyone rides with a helmet now and you will get yelled at on the road if you do not ride with a helmet. Other considerations when selecting helmet is fit, ventilation, visibility, weight, and the style you like. When it comes to bike attire MEC is a good reference for goods and information. Here is short article on helmets.
Personally I am definitely on the practical (or cheap) side when it comes to attire. I prefer fit and function over style and looks. If something works and is cheap then I am all in. I do not really care about brands and colours. I am not keen on any bling bling effects. Having said that, if you do ride in the dark it may be good to invest in some glow and bright clothing, so that you are more visible on the road. I like the sun and try to do all of my riding when it is light and dry.
Accessories
These are the things to have on your bike or on you when you go out for a ride. They fall into two general categories, items that keep your body fuelled and comfortable, and items that look after your bike in case of mechanical issues or road side mishaps and emergencies.
For myself, I usually bring some snacks along for the ride just in case I get into a hunger spell. For rides longer than 1.5 hours, I start bringing high carb content food such as a peanut butter and jam sandwich (cut into bite size units) or a banana. I also love these Endurance TAP maple syrup gels. I usually have one or two in my back pockets.
Aside from energy food, you will also need lots of water to remain hydrated. This means water bottles and bottle cages that fit on your bike to carry them. Everyone is different. You will have to experience how much water you will need. I typically carry a single bottle for rides between 1 to 1.5 hours, and 2 bottles for longer rides. For rides longer than 2.5 hours, I will need to seek an on route water source (eg. a convenience store) to refill the bottles. The amount of water you bring is also weather dependent. You will find that more water is required on more hot and humid days.
A Typical Multitool for Cycling
For the bike, a multitool to tighten, loosen, or fix odds and ends; one or preferably two spare inner tubes – when (and not if) you get a flat. For quick fills in seconds, a CO2 canister system is the way to go. However, you get infinite air if you use a hand mini pump. Make sure you buy one that can sustain high pressure (90+ psi). A patch kit is great when you run out of spare tubes. I have been on rides when I got more than 3 flats!
Last but not least is a fully charged cell phone so that you can call for help as a last resort or communicate with your riding buddies in case you get separated.
Garmin Edge 500
There are other items that people buy to measure their performance and accomplishments. Things like a cycling computer with GPS, a speedometer, and a power meter. I will not go into details here. I personally own a Garmin Edge 500 and it has served me well. All this hard work must be recognized somehow! It also helps to pace a ride.
Your local bike store, or Mountain Equipment Co-op is a good start to procure these accessories.
Join a group
Road cycling is exhilarating, exciting, and will certainly make you more healthy. It gives you more energy when you are not riding, and increase your sense of accomplishment. However, cycling on the road can also be dangerous. In the end, you cannot control who is sharing the road with you and their level of maturity and competence. It is always safer to ride with a group so that you have a bigger presence on the road and you look after each other. So once you are all geared up, join a local club or find some riding buddies in your neighbourhood.
Improve your riding skills by being attentive, tune your riding skill by learning from each other. Before you know it your group quickly turns into a coherent team. You will pleasantly discover a priceless sense of fulfillment and enjoyment once you are a part of an integral group. You will get there before you know it!
I recently watched a CBC Marketplace segment called, “How hackers take over your accounts using social engineering“. It was really alarming how bad people can effectively steal your mobile phone number. These bad actors use well known social engineering tricks to deceive a customer service representative of your phone company like Rogers or Bell. Once they convince your carrier that it is you, then they can proceed to associate their own SIM card with your existing number. They can also lock you out of your account by changing your PIN and password information related to your account.
This means any security that is tied to your mobile phone, which includes many two factor authentication schemes can be easily compromised using this technique. The weakest link is therefore your phone company’s authentication process and the competence of their employee who is acting on that process. This is a very thin shield against hackers who are well versed in social engineering tactics.
Unfortunately at this point there is no known defence that I am aware of. I hope that more and more companies will deploy two factor authentication that is not solely dependent on your mobile number, because as the CBC video shows, it can be easily hijacked. Companies should deploy a mobile app that requires authentication and use the app to facilitate two factor authentication.
Others have suggested to keep your mobile number secret, but I think this is largely impractical.
