Topics created by Obsolesce

@Obsolesce said in Mouse pointer periodically freeze in place:

@dafyre said in Mouse pointer periodically freeze in place:

Open Task Manager and see if there are a lot of "System Interrupts"

I don't see where it says that

Sort everything by name in the details tab....

d83df99d-5ba8-484d-b0d4-96d60b679fc2-image.png

As long as System Interrupts is a low number, don't worry about it, but if it hovers around 3 or 4 % (or higher), turn off bluetooth and see if it drops.

@dafyre said in Designing a Potential Large SaltStack Infrastructure:

The biggest bottleneck I see would actually be bandwidth to shuffle stuff around to the remote clients...

Even with tens of thousands, not too likely.

I thought it would be good to give an update on how the headphones are holding up to my usage and expectations.

Just a reminder, I got the Sennheiser PXC 550. They are over/around-ear Bluetooth noise-cancelling headphones.

So I've been using them almost daily now for the past almost 4 months, about 45-60 minutes each way to/from work on the train. I don't always use the noise cancelling on the train, but when I do, it shuts up what I am trying to avoid hearing pretty well. 🙂 I also use them at home for watching Udemy courses, at work, music for R&R at home, and sometimes for a video game. I usually use them wirelessly, but sometimes I hook them up to my computer's USB while listening when they need a charge. The charge still lasts well over 20+ hours. I don't time it, but it lasts weeks before needing a charge. Definitely 2+ weeks, depending on usage, which I mentioned above, give or take.

I found out a co-worker has had the same pair for years, and they also still look and work like new.

To wrap up, they are definitely holding up well, and surpassed my expectations. I definitely recommend!

Thanks for the example. It makes so much sense now.

You had said that the /etc/passwd table returned too many results. I asked that if you didn't want a list of all of the users on the system, how would you define an appropriate list to return as you obviously had to have some shortly list in mind to make the statement. I wanted to know, how would you describe a better return... which subset of users do you want returned when you query the userlist (everyone I know of what's all, not some kept secret.)

You agree not all, but you didn't explain how you would determine which ones to display and which ones to hide. And if the Windows tools don't show all, how do you manage the secret accounts that those commands do not display? Are you saying that there is yet another set of PowerShell commands to manipulate the true local user list?

Your response doesn't follow the discussion. Those are commands, not users.

@DustinB3403 said in Comparing PowerShell to Linux User Manipulation:

@scottalanmiller said in Comparing PowerShell to Linux User Manipulation:

@flaxking said in Comparing PowerShell to Linux User Manipulation:

It interesting to think about, one complaint about Linux is that it has fragmented off into tons of different distributions, however it's has managed to keep a lot of the tools standardized across them all.

That's very true. It's even moreso than that. Many of those tools remains standard across not just operations systems, but OS familys, too. AIX, Solaris, BSD, Linux... all those families tend to share a lot of tooling.

OSX even has kept mostly uniform with the command set from UNIX. It's surprisingly nice. I'm appreciating OSX more because things are easier to repeat over and over and to do things remotely.

Yeah, even with how rough many things are in OSX, their shell and command sets are definitely one place where even Apple is keeping ahead of MS.

Worth noting....

Clojure is a true LISP, implemented on JVM.
F# is not LISP, but OCaml variant, implemented on .NET.

LISP and OCaml are both fully functional languages, but also very different. But Clojure and F# are the two prominent functional languages today, with Clojure being way out in the lead.

https://techsnips.io/

@biggen said in Microsoft is Replacing Edge with Chrome's Blink Engine:

This appears to be for ARM only from what I’ve read.

Just where they are starting. Article says the goal is Edge itself.

@s-hackleman said in Can anyone recommend a great GPS tracker for bicycles?:

I looked into these about a year ago before I got my new bike. At that time there were so many gotcha's. The small ones had bad battery life, the bigger ones were obvious, the cheap ones had horrible signal quality. In the end, it was best for me to invest in a quality lock with theft insurance. That way it is locked, and if it is taken, then I just get a new bike. Sorry I can't offer a better solution.

No, that's good advice so long as it's an option available where I'll be, and not one I had thought of.

@Obsolesce said in VPS Backups:

@scottalanmiller said in VPS Backups:

@Obsolesce said in VPS Backups:

Rather, you temporarily use a snapshot in order to create an image or backup, then the snapshot is deleted and merged afterwards.

That's called exporting a snapshot. There are cases where what you describe is the process. But it's important to understand that the resulting "image or backup" is called a snapshot still. And what you are describing is the process used to produce the snapshots that we are talking about.

Veeam, for example, takes snapshots, that's how it works. All non-file based backups are snapshot based backups.

File backups work from the filesystem level. Snapshots are the only way to backup from the block level.

Ya that makes complete sense. I don't know why I was thinking down that path.

THIS ^^^^^^

@Obsolesce said in Converting a cron job to a systemd timer:

First some info as it may have impact on whether or not to do this in the first place:

I have a SaltStack deployment of 600-ish minions. Currently, I have a highstate set to run every 45 minutes via cron:

*/45 * * * * root /usr/bin/salt -b 25\% '*' state.apply

I know one of the benefits of systemd over cron is the logging, and I think this would be great to have better logging of when this runs. More can potentially be done with it automatically, like look for issues and send emails just as one easy example.

But I don't know how to test this... maybe it doesn't produce the kind of log I want, and salt may already do this in it's own log even better. Or maybe this will produce a log that's too big and cause issues later. I can watch it in the beginning either way, so it's fine.

Can that cron job be converted into a reliable systemd timer? If so, how?

600ish minions. Wow. Curious about your setup. (Types of systems, types of states, how to manage and review output of all of that, etc.” Maybe an upcoming post? 🙂

@tonyshowoff said in Which Cloud Company Should Get The JEDI Contract?:

Yes it starts with a choice of good companies, but soon enough they'll leave after having to deal with technically incompetent committee members who don't understand what a scope is and the only ones left will be companies absolutely incapable of doing the job.

I can't upvote this enough. This is supposed to pay $10b. I will bet dollars to donuts that whoever gets the contract will A) lose money and B) have about 10 deadline extensions.

And after all that, it will be slapped together and almost unusable because of the red tape and ever moving goal posts of the scope of work.

I got a 2-year SSL cert from NameCheap for $15 (it's Comodo). Installed and working great. No need for a $200/year cert lol.

@scottalanmiller said in Discovering FakeRAID in the Real World:

@wrcombs said in Discovering FakeRAID in the Real World:

@scottalanmiller said in Discovering FakeRAID in the Real World:

@wrcombs said in Discovering FakeRAID in the Real World:

@scottalanmiller said in Discovering FakeRAID in the Real World:

@wrcombs said in Discovering FakeRAID in the Real World:

@dave247 said in Discovering FakeRAID in the Real World:

@scottalanmiller What about RAID 1 on the motherboard? Yeah it's "fake", but it's still effectively mirroring the drives. I use that on my home pc and it hasn't failed me.

Thats what we use .

Is it? I thought you just determined that you didn't have FakeRAID, but had to use Windows RAID instead (which is the recommended choice anyway.)

I just know that we mirror Drives, WE use RAID 1 : Using CLTR+Fduring the boot. Is it possible we are using FakeRAID? Completely! like we have all came to the conclusion of: This place is diseased with horrible practices. Why we dont use Windows RAID- but instead use FastTrak? I have no Idea.

Oh, if you are doing CTRL-F during boot to configure RAID, pretty likely it is FakeRAID. But the one in your other thread didn't support that. That's a unique one, though?

Unique for sure, Because we us FakeRAID: but we could set it up to use Software RAID..

Right, you have options, but chose the weird one, it would seem. And the one that can't be used uniformly.

Sadly this falls under : Being Plagued by horrible practices . . .

Updated the OP.

I tried it today...

It tries to do something, but crashes without error.

@pete-s said in Comparing 15k SAS and SSD:

@jaredbusch said in Comparing 15k SAS and SSD:

@pete-s said in Comparing 15k SAS and SSD:

@obsolesce said in Comparing 15k SAS and SSD:

@travisdh1 said in Comparing 15k SAS and SSD:

@obsolesce Nice to confirm what most of us have been saying already.

Yeah its nice to see the numbers.

There is also the reliability rate to consider. SSD drives have 3 to 4 times lower annualized failure rate.
And you're more likely to use more 15K drives and do RAID 10 instead of RAID 1 with SSD.

Regarding SAS/SATA SSD they will all disappear soon. NVME drives in the U.2 format have superior performance and a small price premium per GB compared to the same drive in SATA.

As far as I'm concerned SATA/SAS SSD should be bought for legacy use only.

That is totally not useful as the server manufacturers are not manufacturing servers to use those generally. Once that changes then I would recommend switching to those.

That might have been true in the past. But just looking at Dell Poweredge servers with one or two sockets, the following support hot-swap NVME drives : R6415, R7415, C4140, R440, R740xd, R640, R7425.

Yes, it is becoming standard, but it is not ubiquitous yet.