darksky.net is a good weather site and no ads, videos, etc. Very clean interface.

@bigbear said in I am going to start an ISP:

@nerdydad said in I am going to start an ISP:

@emad-r said in I am going to start an ISP:

@nerdydad

Best of luck, if you pull this off, you will be your own boss.

Thanks. My only concern is I will be trading a much shorter commute for more working hours. Will I even see my kids during the week besides when they are already in bed?

This should be a given with any startup in its initial years. My older 2 kids are getting to the age that I would be teaching them to mount and sync CPE's and properly seal outdoor grommets When the first storm comes through and knocks all your antennas off spectrum you will have help.

Don't forget the drip loop...

I installed a few of the wall racks with filters on the bottom and fans on top for exhaust like the ones Jared mentioned. Don't remember the brand or price. The equipment stayed very clean and cool. Hard part is remembering to change/clean the filters regularly depending on the environment.

I use Vultr and have each customer set up as a different site. Works perfect.

I am running Nethserver for a local email server for internal-only messages. Curious to know if anyone else uses it.

Thanks guys. Learned some new things today. I set up 2 more NC instances in my lab. Trying now to decide the best route to move the data directory out of the default location. Seems everything I read takes a different approach to it. I want to build a new instance with a large amount of storage and migrate a ton of files. Even NC recommends ( I know you are all aware of this )moving the /data out of /www but finding it difficult settling on a method to do so.

2 machines that are Windows 10 Pro and signed in with a MS account - What is the proper way to access a shared folder on a local network? Going to start a different post with the main problem for another discussion as to where this is coming from. The credentials are causing the issue.

Updated 2 more desktops from F29 to F30 and building another F30 KVM host.

I did a good one today - Backed up my PC to an internal 1Tb drive so I could reinstall W10 (on the same SSD). Realized I formatted the 1Tb drive when going through the install wizard. No files to move back to my fresh PC. Oops. Luckily, I have another backup on an external drive.

I've got it up and running again in my lab. Have 1 Fedora server VM running as a test. So far, so good.

Hyper-V has been working flawlessly on many hosts for me, for at least the last 7 years. 2012 R2, 2016, and 2019. I just think its time to discuss alternatives.

None of that really affects things one way or another. That Hyper-V is all but abandoned and isn't as good as KVM, matters.

That is another reason for a change. I don't want to deal with Hyper-V in Azure after 2019 goes EOL/EOS.

@scottalanmiller Because it is government and they are the vendor we are approved to use because they had the best pricing and are under contract. We have another vendor we got pricing from as well and are under contract with. They finally agreed to drop the VMWare from the quotes.

Greetings everyone.
A coworker and I have been working for a while getting quotes for hardware to run VMs. The vendor is pushing harder than ever to purchase the full VMWare suite and is clearly not taking NO for an answer. We finally got our point across.

For those of you using Proxmox now - what are your general impressions? We have KVM hosting our current VMs on hardware that is due to be replaced. Were planning on running vanilla KVM again, but are curious as to how PVE is holding up. We would benefit from the backup appliance as well. For backups now, we just rsync the sparse files. to a NAS.
I have years of experience with Hyper-V and Veeam, but it is not worth a Windows server license to run Veeam, and all the VMs will be Linux. There is no domain. Hyper-V just does not make sense.

@dafyre During my rsync it dropped to 101kB/s. Very strange. No other network traffic.

@scottalanmiller I did the same test over the network. With the same file. 27G mkv file. From the raid disk it would start out at 400 MB/s and then go to 120kB/s

From the same server and the same file over the network, it sustained 465 MB/s from an internal SSD instead of the raid array.

The network is SFP+ from box to box. No other traffic. Copying the file internally in the server from raid to SSD was approximately 250 MB/s. What would cause the network to copy the same file but slow to a crawl? It gets more confusing the longer I look at it.

I have a NAS that is running an Intel Core i3 with 8 cores, 16G RAM, and a 16Tb R10 (mdadm) with 4 x 8Tb Seagate Barracuda disks. OS is Fedora Server 34 minimal. Only used as an NFS store.

I noticed some VERY slow speeds when reading from the server over 10Gb network. I copied an mkv file to the server as a test. Doing an rsync from the raid array to a separate F34 server with only and an SSD disk, the rsync was transferring at 3MB/s. Estimated transfer time was approx 5 hours.

I moved the same file from the raid disk to an internal SSD and the rsync completed in 57 seconds to a 27G file. Transfer rate was 465MB/s. I realize the SSD is faster, but what caused the R10 disk to suddenly have very poor performance?

I ran SMART tests on the disks and they all passed. MDSTAT shows the array in good health.

Any ideas on what could have changed? I am running out of options. All the testing is looking good except for transfer speed.

@jaredbusch These are not dry pair. They are separate POTS lines. I verified that as well. They are originating in an Ericsson AXE switch. All 3 endpoints are within 1.5 miles of the ATT Central Office. They are all on the same BTN. I personally saw the monthly statement and the letter informing of the price increase.

It is 100% their fault and they realize it. It is one of those instances where the bill just kept going up each year and they kept paying it. They have had this service for decades.

I may ask them if they want me to have someone else manage it for them. It is one of the most important parts of their business.There are only 3 or 4 users, but there will be more than 3 phones including the front and rear doors. No video, just intercom.

@dashrender It is crazy to say the least. To top it off - the off-prem lines are only used on evenings and weekends. They are not used M-F 8-5.
It appears ATT is trying to price everyone out of POTS and other legacy services and push them to hosted services. ATT was going to move them to their managed VOIP service which now is apparently RingCentral for $33/mo per device. Originally they were quoted something lower but it changed. Strange any way you look at it.

It could probably be made to work by forwarding from the old pbx, but that still requires them to keep POTS service at 3 locations. I think the features of SIP and a new system would give many advantages.