Best posts made by stacksofplates

Kelsey Hightower mentioned Zeit last year and I've been meaning to look into it. It's a pretty awesome serverless platform.

I didn't use their now client, I just added a repo to Zeit and it manages the build and deployment for me.

Here's a link to the repo: https://gitlab.com/hooksie1/zeit-test

This was written in Go. Using Go you just need an exported Function and take the http.responseWriter and the http.Request. Then do you work on the payload and return the info.

Here's the URL for the endpoint: https://goniq.hooks.technology/api/test

You can test it by giving it a list of strings and it will return only the unique strings. For example:

curl -X POST https://goniq.hooks.technology/api/test -d '{"list": ["testing","testing2","testing3","testing2", "mangolassi", "mangolassi"]}'

It will return this data for you:

["testing","testing2","testing3","mangolassi"]

I do a few different things. I use FreeOTP and Yubikeys to store legacy TOTP codes. I also use Yubikeys for u2f as well. If the service has a 2fa push option like Zoho does, then I'll also sign up for that. This way I cover my bases if I happen to lose something.

Plus you also have recovery codes for that as well.

@JaredBusch said in Zerotier on Windows firewall rule question ...:

I never have to make any rules to enable communication on a ZeroTier interface.

Same here

@scottalanmiller said in Nodeweaver:

@mroth911 said in Nodeweaver:

I honestly don't know the answer to this question yet. I installed it. was very straight forward, Mixes and matches hardware, auto load balancing hot swap drives. So your telling me I can build promos with lizard fs

Right, you could build all this without Nodeweaver. Not to knock on NW, I'm sure that they are fine. But like FreeNAS, it's unclear what they are adding to the existing capabilities we already choose not to use normally.

Not really. Libvirt doesn't have the ability natively to automatically fail over between systems. You'll have to set up corosync/pacemaker or keepalived, something along those lines. Out of the box you can transfer VMs as long as they share storage, but it doesn't auto fail over. It also doesn't have a feature complete web UI yet. You can't spin up certain systems in a specific order (which was mentioned) and is important for some legacy applications. Libvirt definitely does not auto balance like mentioned before either, you would have to write that yourself. You also don't have a shared network between hosts either. You would have to manually set that up as well. Then you have to maintain all of that on top of maintaining the guests.

@JaredBusch said in Keep my ssh config file synchronized between two systems:

@stacksofplates said in Keep my ssh config file synchronized between two systems:

I use git for this type of stuff. I have all of my dotfiles stored in a git repo and synced between systems.

I thought about that, but then it gets into git triggers and scheduled jobs.

I just do it when I open my terminal. It auto downloads when a new window is opened.

@warren-stanley said in Microsoft Teams:

@nadnerB said in Microsoft Teams:

@scottalanmiller said in Microsoft Teams:

@nadnerB OMG, is MS Teams like SnapChat?

Looks like it's a bolt on.
https://collab365.community/how-to-use-snapchat-filters-in-a-microsoft-teams-meeting/
https://www.onmsft.com/how-to/heres-how-to-use-snapchat-camera-on-windows-10-to-spice-up-your-microsoft-teams-calls

its a free-for-all in Teams with default settings..... users can do almost anything

That works with anything, not just Teams. It just creates a default webcam.

@JaredBusch said in Any recommendation for markdown support in Wordpress:

Since I spun up a WP instance for myselff again, I was looking at syntax stuff.

Markdown is really my preference anymore, but a fast search of WordPress plugins does not return anything obviously worth it.

So before trying various things out, I figured I would ask for a recommendation.

I'm not too worried about code highlighting or syntax recognition. But those are a nice bonus.

I used the generic WordPress.com and they had markdown built into the wysiwyg editor (if I remember right) but it kind of sucked.

I gave up on it and switched to what I'm doing now. I never found anything good, but again it was their hosted not something I was running.

@JaredBusch said in Any recommendation for markdown support in Wordpress:

@Obsolesce Yeah, no.

Join the dark side and switch to Hugo and GitLab/GutHub pages

The speed and ease of everything will get you addicted

@JaredBusch said in Any good getting started with Hugo resources:

@stacksofplates said in Any good getting started with Hugo resources:

3. cp -R themes/theme/examplesite/* .

Why do this? Why can't I just create my own things? Why bring everything from the example up?

I don't mind this, just trying to understand the logic.

Each theme has it's own specific settings. The example gives you all of those settings without having to read through all of their theme and find the settings for yourself. You can delete the files in content/ that you don't need. It's just a way to show you how to use the theme and give you a default config.toml.

@sadeghpm said in Anyone using Jitsi behind Nginx:

Hi guys.
Same problem. I have jitsi docker container behind nginx.
Where is /etc/jitsi/videobridge/sip-communicator.properties file to change values? in docker configuration files?
thanks

Don't edit files in the container. Make the sip-communicator.properties file on your host and mount it under /etc/jitsi/videobridge/sip-communicator.properties

@sadeghpm said in Anyone using Jitsi behind Nginx:

@stacksofplates
thanks guys.
bind /etc/jitsi/videobridge/sip-communicator.properties with local file and container config file updated, but problem exists! very strange!

Are you using their docker-compose file? They give you a .env with all of the environment variables you need to set.

I hadn't looked at their repo. You should set all of those options as env vars not in the config itself.

Once you have the env vars set up just run docker-compose up -d

@travisdh1 said in ESXi cluster, advice needed:

@pmoncho said in ESXi cluster, advice needed:

@travisdh1 said in ESXi cluster, advice needed:

@pmoncho said in ESXi cluster, advice needed:

@scottalanmiller said in ESXi cluster, advice needed:

I'll echo Travis. For such a small workload, that alone would normally rule out ESXi. ESXi is really only beneficial at massive scale (hundreds or thousands of compute nodes) where you are specifically leveraging unique tech and VMware as your support partner to handle, more or less, everything. Anything smaller than that, I'd recommend ruling them out completely and never entertaining the product. Just doesn't make sense otherwise - high cost, high complexity, high risk, no rewards.

Question - If a site is only using a single server, I cannot see what benefits other Hypervisors have over a $500 ESXi essentials license other than saving $500? (My knowledge is limited to ESXi and very limited Hyper-V).

I do see, based on much reading here over the years, the benefits of XCP-NG, Hyper-V, and others in a multi-server situation because of the free extra's. Not so much in a single server situation.

You need to stop looking at it from a standpoint of what you already know. Let me ask it another way.

Why are you going to spend that $500 for fewer features than the free alternatives?

In a single server, It seems the free alternatives lose their feature power and it all comes down to $500 (actually $576).

With that being said, ESXi vs Hyper-V in a large single server situation, It seems setup, stability, small footprint, install on USB/SD Cards are benefits of ESXi. I feel I can save up to $200 during ESXi setup alone. Plus, $300 for vCenter is an easy bonus for further configuration and management.

In the words of Dennis Miller, "That is my opinion, I could be wrong."

And I can spin up a new KVM based hypervisor in less than 5 minutes of my time. Plus I save time on the back end by not having to enable basic features for each VM (CPU and RAM dynamic allocation, blanking on what ESXi calls it.) It even includes a web based management platform (Cockpit).

I agree that Hyper-V is such a pain to setup and manage that paying for ESXi could make sense if they were the only two solutions available.

even includes a web based management platform (Cockpit).

This is not acceptable for 80 VMs. You can't clone through it. You're going to manually install 80 systems without a base template? Virt-Manager is still the best way to manage.

However you don't have a REST API for KVM (or Hyper-V). That's a big advantage that vSphere/vCenter has. That opens up a multitude of tools to automate everything you're doing.

Terraform does work with libvirt (through a community provider), but because of the way libvirt works your templates and disks are downloaded from where your terraform instance is running (see here: https://github.com/dmacvicar/terraform-provider-libvirt/issues/299). So you're now limited to only running over localhost unless you want to stream your disks to the remote end every time you build a system.

I still love KVM but I'm very sad that it's this many years later and they don't have anything to compete with this.

A shortcode is probably the easiest way to do what you want.

I stole this from here because I'm too lazy to write it myself. Just put this in layouts/shortcodes/image-resize.html

{{ $imagename := (.Get 0) }}
{{ $options := .Get 1 }}
{{ with .Site.GetPage "section" "media" }}
  {{ $original := .Resources.GetByPrefix  $imagename }}
  {{ with ($original.Resize $options) }}
  <img src="{{ .RelPermalink }}" width="{{ .Width }}" height="{{ .Height }}">
  {{ end }} 
{{ end }}

Then just add the image with:

{{< image-resize your-picture-name "150x150" >}}

@JaredBusch said in How can I scale an image in a Hugo page:

@stacksofplates But this begs the question, how do I get a picture name?

where the fuck do I put image files? and do I then need to define them someplace else?

I know I canput things in static, but my understanding is to avoid that unless I really mean to add some static thing that is not programmatically built.

That or I am completely not understanding Hugo

That specific shortcode is looking for them in a media directory in your content folder. The name is the file name minus the extension.

@Obsolesce said in How can I scale an image in a Hugo page:

@JaredBusch said in How can I scale an image in a Hugo page:

Currently, markdown does not natively have a way to scale images.
https://daringfireball.net/projects/markdown/syntax#img

This seems to be an issues for me with using Hugo, I have an existing site that I am converting and the images are scaled with the normal <img> tag.

Searching the net, it seems a lot of tools have added functionality to markdown to support image scaling, but it seems Hugo's implementation of Markdown has not.

I tried to just use the HTML element in the Hugo page, but it was not rendered.

Most Hugo themes have a way to add custom CSS properly, so that's my preferred method to changing visuals. And is a hell of a lot easier to maintain and write than trying to code all that other B.S. as in the solution. That's a huge turn off for me with Hugo as well as so many of the other annoyances. If it wasn't for the theme I found I'd be going back to WP or HTML template.

The problem with CSS for this is it it only works with one image or all of them. You'd have to continually add CSS properties for every image you create. The above solution is just a tag you write in your post. You just write the shortcode one time and then use it when you want to.

My personal opinion is that I think this is much easier to maintain. I don't have to manage a server, database, patches, plugins, worry about security of my system/CMS, etc or pay someone to do that for me. Every so often I'll have to write something like the above solution but it's not common and I don't have to manage anything. I can also redeploy the same site on literally anything because there's no backend to deal with.

Not saying I'm right and you're wrong, just that's why I really like this tool.

@scottalanmiller said in Termius cross platform sync:

@JaredBusch said in Termius cross platform sync:

@stacksofplates said in Termius cross platform sync:

@gjacobse said in Termius cross platform sync:

@stacksofplates said in Termius cross platform sync:

@IRJ said in Termius cross platform sync:

Remina is great on Linux platforms, but the question for me is why is this a need?

This seems like everyone could and should manage this independently. All you need is DNS name or IP to initiate a remote connection. In my opinion, it's better for IT team to know exactly where they are trying to go instead of clicking the wrong button or sending the wrong command

Yeah I agree. I'm assuming it's for syncing credentials across devices. Which means you'd have to trust their cloud service with your system credentials.

While yes, it would be nice to sync the entire session - connection and UserID / password. I'm more concerned with the connection itself. Yes, I can keep track of the addresses - but it gets to be a pain.. UserId / Passwords are different. I could care less - I mainly want the address; IP address or dns name..

I mean, honestly what's the difference between a word document/text file and the syncing at that point?

Right. An besides, even Windows has native SSH now. So why use anything else anyway?

Right, I've not used PuTTY in quite some time. Not that it isn't good, I just don't see the point of installing third party software that doesn't do anything any better than the built in tool that is always there and ready to go. And quite frankly, I find PowerShell's terminal to work far better for me.

I can't stand PuTTY. I'm not sure why, I've just always hated it.

@JaredBusch said in ZeroTier Own network controller // connection:

I've never bothered to setup my own ZT network controller, so I have no idea.

I think @dafyre did a long time ago, Maybe he can help.

He did. Back when it was only like 5-10 free devices (I forget what the number was). But since you get 100 free now there's not much point unless you really need a lot.

@Magrossm said in ZeroTier Own network controller // connection:

@stacksofplates actually I did manually, building the planets file in the working directory myself. Its a bit hacky but works. Problem is the nodes need the same planets file. I thought maybe there is an easier way.

Ah ok they must have finally enabled that. I never looked into it after a couple of years ago. If you have any automation/configuration management set up you should be able to drop that on all of the systems pretty easily.

@biggen said in Gluster and RAID question:

@scottalanmiller said in Gluster and RAID question:

@biggen said in Gluster and RAID question:

@scottalanmiller No problem. So I'm guessing if one really wanted to use the "distributed" type than RAID would really need to be required if you wanted redundancy. I think I'm wrapping my head around this now.

I think you are thinking about this all wrong.

First, you never use RAIN and RAID together. So anything that's making you think of using RAID with Gluster means you are thinking about it fundamentally wrong. It's not that it's physically impossible, but that it makes no sense.

Second, you never choose distributed if you want redundancy. So never would there be a case where you'd have the distributed type AND want redundancy. You'd choose the redundancy option instead.

So this takes me all the way back to my OP:

Are Distributed Gluster deployments typically in Production?

I guess if one didn't care about redundancy that would be the only use case for that specific architecture. Because the only way to provide it would be with RAID, and you say that running RAID under RAIN isn't the way to ever run RAIN to begin with. So using the "distributed" type of Gluster with RAID to provide redundancy would be a poor choice to ever use with like I was thinking.

You can do distributed and replicated for a volume. It's not just one or the other.

@biggen said in Gluster and RAID question:

@JaredBusch Once the volume is up and running how the heck does one share it out? That what I'm trying to do. I have a successful two node system running:

joe@glusternode1:/mnt$ sudo gluster volume info

Volume Name: gv0
Type: Replicate
Volume ID: ab19d123-eb34-4186-8a03-316a3fc790e3
Status: Started
Snapshot Count: 0
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: glusternode1:/data/xvdb1/brick
Brick2: glusternode2:/data/xvdb1/brick
Options Reconfigured:
transport.address-family: inet
nfs.disable: on
performance.client-io-threads: off

That volume must now be mounted "somewhere" to access it. How do I mount it so Windows clients can access it? Do I simply mount the share in one of the nodes under /mnt/big_ole_gluster_space and then share out that mount point via Samba from that same Gluster node?

The preferred way is to use the GlusterFS FUSE client. Last I knew it's the only one that automatically handles failover and HA.