Got it. I was using the account id with an application key.

A more careful reading of the above linked page shows that:

0_1542076748186_e6061550-5167-4def-b9b5-5b89e859f408-image.png

So that was the disconnect.

Application Key requires Application ID. Account ID requires Master Application Key.

When the examples clearly stated ACCOUNT_ID and APPLICATION_KEY

0_1542076683650_e8a7c2f9-0a93-4c35-b429-0bf24767fe33-image.png

IPvFoo link for Firefox

@DustinB3403 said in Anyone looked at Yealink DaaS - Device as a Service:

One thing you haven't taken into consideration though - is the time value of money.

But I have. It takes moments to do the math that this is a bad idea. In any case we're discussing a business who is installing a new phone e system and phones.

So the business has the money, but would it be better to expense with however many thousands upfront? At least in my business they would opt to make the spend today, and then they'd write it off over the usable life of the equipment.

With that, yes it's still a capital expense. And hopefully that expense was planned for and already budgeted.

That had been the typical situation here until recently. Much to my utter amazement we've leased two pretty major pieces of gear - CT machine and our entire printer fleet. I know in the case of printers, we'll be saving money over what we were paying our previous provider, the purchase option might have saved us a small amount, but we would have to have taken out a loan, and the interest would have made the end spend nearly the same.

@aaronstuder said in Vultr Instance options growing:

How can you be sold out of Windows?

0_1541469819516_54438dc1-b95a-4406-a24b-7c883e6f623c-image.png

Almost certainly it is limited to certain hardware instances.

@matteo-nunziati said in How can I build this displaylink driver for fedora 29:

@JaredBusch said in How can I build this displaylink driver for fedora 29:

I have a standard Dell USB dock but want to use it with a Fedora 29 user.

https://github.com/displaylink-rpm/displaylink-rpm

That repo has a pull request in for the Fedora 29 updates, but I figured I could build it early for testing. Also I want to get more familiar with doing this on Linux in general.

I have no *devel bits installed on my system. Obviously, I can clone it down and run make, but I wonder if I need any other resources?

Inside the "ci" folder there is a script to let fedora downlad everything and compile via makefile

That did it.

Finally. Very nice.

@DustinB3403 said in How can I backup an Azure SQL database to Backblaze B2:

@matteo-nunziati That's pretty bad design if there is no automated way to backup to a remote blob provider or to push said backups to said outside provider. . .

That $399 is sounding incredibly reasonable now.

A main goal of any cloud service from major vendors is to lock in. Try to access a backup in GCP to export data: you can't.
I'm digging a bit to understand if the SQL backup in azure is accessible or managed by any infrastructure account which prevents users to access them directly...

BWT, 399 lifetime is good.

Very nice!

@Dashrender said in What makes people want IPSEC at line speed:

You're 10 miles apart, any chance for a site to site wireless link?

OK so making them split between some things local and some remote - why not move them 100% remote? Give the users a full RDS desktop, and have them completely stop using their local system?

1 Gb connection for CAD is still going to be an issue in my mind - I don't really see this solution being better, but who knows, you might get lucky.

What is your end goal for backups? If it's to continue taking tapes to the bank, why not just pick up two tapes/drives, whatever, one from each site and deliver them to the bank instead of copying over the WAN?

no wireless without large towers unfortunately, I looked into it before settling on what we did, but I didnt want to try and deal with renting space on someone else's tower. It was getting intimidating and that plan would have been probably more than I could have pulled off.

I am not sure a full RDS desktop would work under the CAD load, and I know it is not allowed under autodesk licencing without getting Citrix involved.

In theory, the 1gbps WAN should be similar to the 1gbps LAN, at least that was my thought. I realize now that latency may still be an issue, but it has only been in place for maybe 2 months. Time will tell if that is the long term solution.

For backups, I have been and currently am doing everything from my location, which is now the HQ. I am backing up roughly ~600GB onto a 1TB external SSD via usb3. I've got somewhere between 6-8TB of total data that I would like to backup, but I had neither the space nor the time to get that all onto a single device that I could take offsite. This forced me to have to choose what to backup, because of lack of anyone higher than me that could/would give me a solid business policy to follow. I don't like being responsible for deciding what does and what doesn't make it into these offsite backups. One problem I am running into is that the person giving me the requirement for offsite backups (the CEO) has no clue what there even is to backup in the first place, because no one here (with a few possible exceptions) can even understand this stuff. I had a conversation just yesterday with him about wanting some direction on how long he wanted to retain backups, and if he wanted that retention done onsite or offsite. He couldn't really give me an answer, he just wants the "drawings" to be backed up "forever". In the end, I basically talked him into officially telling me to do what I had planned on doing in the first place, just so that we had "officially" talked about it. That is probably off topic though.

Current: Like I said, we are currently backing up 600GB worth of files to a single usb SSD that I rotate out on a weekly basis. Before the IPsec was in place, it took ~150 hours to complete, which since they were weekly backups, took basically the entire week. Now they are completing in ~50 hours, but I am still pulling individual files across the WAN.

My plan at this point is to move everything over to a single new host at my HQ. This host will be running local SSD's, see https://mangolassi.it/topic/18201/large-or-small-raid-5-with-ssd. I've got two existing hosts (I picked one up along the way) that will be repurposed once the new host is in place. One will become a veeam host (it will be getting new storage), and the other will become an empty host used only for restores. All three hosts will be on a new 10G network, and the veeam host will be getting a tape drive (most likely, see https://mangolassi.it/topic/18209/adding-tape-drive). By using LTO-7 tapes, I can backup literally everything I have, and take those offsite. I am going to backing up to disk on the veeam host, and then copying those to tape. I am also going to be copying my backups across to my branch site. With the new setup, I should be able to do the offsite copy job in a matter of hours. So, I will have 4 copies of the data, 1 production, 2 onsite backups, and 1 offsite backup. I will also be able to run everything from veeam instead of trying to mix that with individual files.

I still need to decide on how much storage to give said veeam host, but it seems challenging to determine how much each backup requires in the way of storage space, especially since I am deduping mine now using windows server.

@NerdyDad Nope, by then I'm just starting my own 6 month upgrade cycle. Because in 3-4 months a new Fedora will come out, I'll wait 2-3 months again, and then upgrade. Like a clock 🙂

@JaredBusch said in ScreenConnect agent on Fedora fails rpm install:

rpm -U ConnectWiseControl.ClientSetup.rpm

ConnectWise Control version 19.0.23234.7027 resolved the failure to update. but they neglected to remove the old screenconnect version as the new install is named connectwisecontrol

bc755dec-f586-4439-a406-222af7737ed0-image.png

@JaredBusch Me either. I just read the article. I loved that Red Hat wasn't joined with anyone. I know that IBm has been moving toward open source for a long time but I just don't know about it.

@dave247 here is a mental exercise to try (don't do this for real, just do it in your head...)

Go back and read the discussion. Then picture yourself printing it out, and handing it to your CEO for a review of your thinking. Ask him to look over the discussion and give his opinion, not of it UTMs make sense or not, he has no way to know, but of your approach to thinking about them. Would you be happy with him reviewing your "logic" behind spending so much money, spending so much time, and how you are looking to secure your business in the future. Are you confident standing behind your decision making?

Take the reverse. I'm completely comfortable knowing that my statements are public, my boss can (and does) read them, as do my customers. I feel confident that, regardless of if a UTM is correct in any given situation, that my logic, points, consistency, etc. speak for themselves and I stand behind them. I have no worries that someone to whom I answer, even if it is just other peers in the community, can look at this and will agree, perhaps not with my conclusions, but with my statements and logic.

Thinking of it as a private conversation between two people makes it far easier to lose perspective. Thinking about your decisions of this nature as being continuously reviewed by someone to whom you answer can help you step away and look at your logic with an outsiders view and can help force a perspective.

Do you feel that your reasons would hold up if someone whose money was being spent, and whose data was needing to be protected, was the one who was reviewing your logic?

So it looks like RC MIGHT actually lock the phones, and you'd be forced to get them to unlock the phones before you can change the password firmware etc.

https://www.reddit.com/r/techsupport/comments/1ddnsj/ringcentral_locked_phones_and_i_dont_want_to_buy/

@mattbagan said in Install NodeBB 1.10.2 on Fedora 28 with Mongo:

@JaredBusch worked like a charm. Thank you!

Thanks, great to hear.

Nice write up, thanks!

@Reid-Cooper said in ManageEngine ServiceDesk no longer unlimited technicians:

That's a common tactic. Make free customers keep thinking about the vendor and that they are getting something for free. VMware does that, too.

This is really not a big thing to me either.

@dustinb3403 said in New pi-hole install has no defualt white list items:

You can add the lists rather quickly as well.

Obviously. But I think I will wait and see.

I'm impressed with their forthright business practices and openness with their platform right down to the DWGs. 🙂

I'm glad they've been publishing their stats. IIRC, they got a bit of grief from drive manufacturers when they started to publish their experiences?