@biggen said in Any idea why Debian complains about a start job failure on boot after a new install?:
@Pete-S I’m not there now but shows disklabel as “dos” if I remember correctly.
Please post it when you have access to the system.
And also the exact error you get in the log.
@biggen said in Any idea why Debian complains about a start job failure on boot after a new install?:
When I say “disk label” I mean partition type. So DOS = MBR. That is how the disk is partitioned now.
What you say is confusing to me. What does fdisk -l look like on the system?
@biggen said in Any idea why Debian complains about a start job failure on boot after a new install?:
I’ve tried to zero the md superblock, after that fact, I’m not sure it works anymore. If I boot into Debian (after waiting for the failed job start) on that disk and run that command, I get “couldn’t open for write. Not zeroing” for that drive /dev/sda”.
If you created the raid on the device, I think you should zero the superblocks on /dev/sda
But if you created the raid on the partition I think you need to zero the superblocks on /dev/sda1
@biggen said in Any idea why Debian complains about a start job failure on boot after a new install?:
@Pete-S Yeah I tested it with a VM yesterday on my desktop in Virtual Box after I had my problems on the physical machine. Worked fine. I even did my md -> lvm > XFS setup using two VHDs. Installed and fired right up with a nice RAID 1 array.
I can’t figure it out. It’s like there is something up with both disks. I’ve blown them out with Gparted and DD. I guess I can change the disk label to GPT and see if that makes a difference. I’m at a total loss...
Label shouldn't make any difference.
When reusing drives that has been in md raid you can use mdadm --zero-superblock /dev/sdX to wipe the raid info from it.
I did another try with setting things up manually in the installer with raid1 as well.
Also works.
If I were you I would have a look at uefi settings in your bios.
I usually just disable it so I don't have to deal with any problems, but maybe you need it.
Some BIOS also have bugs in their uefi implementation. So maybe upgrade the BIOS as well.
I installed debian 10 on a VM with LVM and XFS.
Debian puts a swap partition in by default and runs EXT4 by default but after changing that I had this:
This is the block list:
Only a 10GB disk for the VM but it works fine.
Normally I don't mess with the swap partition though.
@biggen
What happens if you just install it with the default partitioning (everything in one partition)?
@WrCombs said in Windows 7 File and Printer Sharing wont turn on:
@Pete-S said in Windows 7 File and Printer Sharing wont turn on:
Check that you have Client for Microsoft Network enabled under networking properties for the LAN adapter.
Thanks @Pete-S I already checked and it is installed and enabled. Still the problem persists.
That was actually for the computer connecting to the share (client), not the one serving the share (server).
But you want to share out a folder on the machine you are having problems with, correct?
Check that you have Client for Microsoft Network enabled under networking properties for the LAN adapter. Same place where you find tcp/ip enabled, QoS etc.
@IRJ said in Custom web apps for SMB:
@scottalanmiller said in Custom web apps for SMB:
@Pete-S said in Custom web apps for SMB:
Now, 50 hours that is a very small project. So your potential customers should be willing to spend at the very least $5K and often much more.
Tiny. Figure 200 hours for the smallest real work project and likely more like 1,200. And that's just to get into production!
50 hours would be more like branding or customization. Not really a build. This is the kind of hours you would expect to have after buying a works out of the box product, but have minor customization for your needs.
Well, you could complete a small project in that time if it's very straight forward and you've done something similar in the past.
@scottalanmiller said in Custom web apps for SMB:
@Pete-S said in Custom web apps for SMB:
Now, 50 hours that is a very small project. So your potential customers should be willing to spend at the very least $5K and often much more.
Tiny. Figure 200 hours for the smallest real work project and likely more like 1,200. And that's just to get into production!
Yes, a tiny project but that is all you are going to get when you are doing it on the side. If you're new, it's likely going to end up being a 50 hour project that will take 200 hours to complete - but without being able to charge the additional 150 hours for learning on the job 
@RojoLoco said in What Are You Doing Right Now:
Anyone else getting this error when clicking in the "Linkedin Endorsements" topic?
Are you talking about this: https://mangolassi.it/post/480737
Also how big of an SMB are we talking about? A company with 10 employees is very different from one with 100 employees.
The cost of something custom is the same but a larger company can make more use of it and as such it creates more value for them, hence they are more inclined to invest in it.
@DustinB3403 said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
@Pete-S At my workstation I'm getting 225Mbit/s down and 155Mbit/s up (clearly not symmetrical there. . .) but not bad either considering I have nothing special configured for my workstation.
On a second test I noticed this A connection of 152.8 Mbps upload would backup 1,650 GB in a day
So this very well could be feasible to do.
Yes, that's not too bad. It could work. As @dafyre and other mentioned you should give it a try.
$.005 per GB is $5 per TB. So get an account and upload 2TB of random data to see how long it takes. Only going to cost you 10 bucks to find out.
@JaredBusch said in Custom web apps for SMB:
@flaxking said in Custom web apps for SMB:
I was thinking about getting into creating custom web apps on the side for SMB, but I'm wondering what you guys think about the market. If I was running an SMB, I would think this is critical in order to keep a competitive edge. But my experience, SMBs drift towards cookie cutter solutions that are trying to be sold to all of their competitors too.
One thing that I think is pretty important is integration with their current LoB applications, which would increase the cost and is probably a reason why they drift towards non-custom solutions.
In my plan, they would own the code (unless there is something we agree on a discounting for making it open source), and I would be using a popular web framework, so they would still have control.
As someone that wrote custom software for a while, custom software is expensive. That is why SMB doesn't do it.
It is too expensive to buy something specific to them. They have to pay to have it wrote. Pay to have it debugged. Pay to have it maintained and updated.
That's no different than general software, but general software spreads the development cost and debug cost across all the customers.
I agree. Just think about how long something takes to do. Say you want to make a custom web application. You need to find out what the customer needs, figure out how to implement it, do the actual work, test that it works, implement it for the customer and hold their hand while they learn how to use it and do whatever changes needed for it to work as they wanted.
Can you do that in 5 days of work? 5 days, 10 hours per day, that's 50 hours total. Say $100 per hour for simplicity, that makes it a $5000 project for the customer.
Now, 50 hours that is a very small project. So your potential customers should be willing to spend at the very least $5K and often much more.
Now given that this is a side job, you can't spend 50 hours in one week. So how many hours do you have where you do nothing in a week, 10 hours? That means that this small 50 hour project will take at least 5 weeks to implement. Or 10 weeks for a $10K job. Almost half a year for a $25K project.
There is only one way to find out if it works.
@DustinB3403 said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
@Pete-S said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
@DustinB3403 said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
@Pete-S 1Gbe symmetric 24/7
So when you upload to Backblaze you get 1Gbit/s?
I haven't specifically checked, but when we get to L3 were do have 1GBe.
You could do a simple test here:
https://www.backblaze.com/speedtest/
I'm not sure it will tell the complete story though. I understand that Backblaze only has one datacenter i Sacramento, California. I don't know how many hops away that is for you. Any congestion, traffic shaping etc on the way will lower your bandwidth.
@DustinB3403 said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
@Pete-S 1Gbe symmetric 24/7
So when you upload to Backblaze you get 1Gbit/s?
@DustinB3403 said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
@Pete-S said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
@DustinB3403 said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
There are a multiple ways of doing this with tools like Duplicati, CloudBerry etc etc etc (the integrations page goes on forever).
In any scenario, if you had a high turnover SMB share with large files (some of which might be 10GB+ individual files) and multiple terabytes worth of change (in a week) - how would you go about getting the data to B2.
Down is the other half of the battle, which can be discussed afterwards.
Using a command line tool like rsync is one option, although I'm not sure how effective it would over a long duration.
To me it does sound like a cloud backup solution is the wrong solution for that use case.
Have you done the math? Multiple of TBs each week, say 5 TB per week, that is 700 GB per day, 30 GB per hour, 500 MB per minute or or 8 MB per second. So you need an average of 80 Megabit per sec 24/7 to upload that amount of data.
Bandwidth isn't an issue, the goal is to offload the data once the working files are collected and to simply store them in a safe relatively low cost space without having to build something.
I too immediately understand onsite backup would be great, but also unrealistic to build as the cost of the storage alone would be far too high.
I do understand what you're saying but I do think bandwidth is an issue. You might have the bandwidth but do you have that bandwidth consistently 24/7 all the way to Backblaze servers?
@DustinB3403 said in Syncing massive amounts of changing data to BackBlaze B2 via Linux:
There are a multiple ways of doing this with tools like Duplicati, CloudBerry etc etc etc (the integrations page goes on forever).
In any scenario, if you had a high turnover SMB share with large files (some of which might be 10GB+ individual files) and multiple terabytes worth of change (in a week) - how would you go about getting the data to B2.
Down is the other half of the battle, which can be discussed afterwards.
Using a command line tool like rsync is one option, although I'm not sure how effective it would over a long duration.
To me it does sound like a cloud backup solution is the wrong solution for that use case.
Have you done the math? Multiple of TBs each week, say 5 TB per week, that is 700 GB per day, 30 GB per hour, 500 MB per minute or or 8 MB per second. So you need an average of 80 Megabit per sec 24/7 to upload that amount of data.
I looking at a software raid6 array of spinning rust for storing backups. Drives will be something like 3.5" Exos 12TB (7200 rpm).
Is it possible to use an SSD drive as a write cache for linux md raid?
I've read somewhere that LVM has that feature.
Is it worthwhile? Anyone here that have experience with it?
Network is 10GbE so having up to 1 Gbyte/s transfer rate is what I'm after.
@scotth said in What Are You Doing Right Now:
@Pete-S said in What Are You Doing Right Now:
@scotth said in What Are You Doing Right Now:
@Pete-S said in What Are You Doing Right Now:
@scotth said in What Are You Doing Right Now:
What am I doing right now?
Arguing with our tech about attaching the shield in shielded wire to source ground while leaving the other end open.I can see the problem as both shielding and grounding is complicated. Probably because there are several different reasons for both and the best option in one situation is the worst in another. Was it network cabling or something else?
Just straight up comms. tx rx common shield.
So RS232 I assume. What did he want to do? Ground the shield in both sides?
1st -- since "the shield doesn't really do anything, anyway", he wouldn't connect it to anything.
2nd -- connect both ends
3rd -- connect it in the endpoint device.
4th -- after I passed along as much information as I could, they finally connected it correctly. I'm sure that someone will take credit for it. I'm ok with that. Just get the damned job done.
And, oh yes, they blew the comm driver chip on the 1st try in the endpoint device. Nice little hole in it.I'm done with this. Hopefully, they'll have learned. Time will tell.
EDIT: They blamed the issue on overhead wires. The owner and I had a little talk. I'm not interested in getting anyone fired, but I won't let the same bunch hook up the next one without supervision.
I wouldn't keep my hopes up. We do a lot of work in factories and such so we have seen a lot of serial communication. Usually avoiding RS232 though and using RS422 or 485 because of all the interference. Nowadays it's a lot of industrial ethernet and fiber. It's easier to install but we still have double check that the shielding on CAT6 is connected - every time.
@scotth said in What Are You Doing Right Now:
@Pete-S said in What Are You Doing Right Now:
@scotth said in What Are You Doing Right Now:
What am I doing right now?
Arguing with our tech about attaching the shield in shielded wire to source ground while leaving the other end open.I can see the problem as both shielding and grounding is complicated. Probably because there are several different reasons for both and the best option in one situation is the worst in another. Was it network cabling or something else?
Just straight up comms. tx rx common shield.
So RS232 I assume. What did he want to do? Ground the shield in both sides?
Network share doesn't mean it has to be on prem. It could be anywhere.
So there are two questions to answer
@scotth said in What Are You Doing Right Now:
What am I doing right now?
Arguing with our tech about attaching the shield in shielded wire to source ground while leaving the other end open.
I can see the problem as both shielding and grounding is complicated. Probably because there are several different reasons for both and the best option in one situation is the worst in another. Was it network cabling or something else?
@black3dynamite said in What Are You Watching Now:
Looks like Gravity but with Brad Pitt instead of Sandra Bullock.
@Obsolesce said in Managing spam posting:
@marcinozga said in Managing spam posting:
@Obsolesce said in Managing spam posting:
Captcha will help against bots. Try that first. Also, start blocking posts if they contain certain spam phrases. Block known ip spam lists.
Um, it won't. Well, maybe just the ones written by script kiddies. Bots have built-in captcha solving, either with services like anti-captcha or some even include OCR and other advanced logic to solve captcha. Google xrumer or scrapebox.
I was referring to these on here, just assuming it's a single dumb bot.
Otherwise blocking threads with those phrases will help likely perfectly in these cases.
For instance any posts having the word "carpet" will be sent to a moderator before being posted.
@marcinozga said in Managing spam posting:
Phone # verification. I'm not sure if anything else is effective against spam. Captcha has been ineffective pretty much since introduction, you can buy services in India where they have real people solving captcha in real time for peanuts. Like 2-3k captchas for $2.
If you can buy services in India to solve captchas, maybe you can buy services to identify and remove spam manually as well. And an infinite loop of work is created. 
OK, this is what I ended up doing.
I wanted the remote server to have access to a local repository served over http. This works with any kind of traffic over tcp though, as it's not a web proxy but tcp forwarding.
I set up the files and served the website on the ssh client machine with PHPs build-in server. It's easy to use and requires no setup. You just start it in the base directory you want to serve. I used port 8000.
php -S localhost:8000
Then access the remove server from the client with the reverse tcp forwarding active.
Basically forwarding port 8000 on the remote host to port 8000 on the local ssh client.
ssh -R 8000:localhost:8000 remote_ip
But since I was connecting with a windows machine I used putty instead.
This is how you set up the tcp forwarding:
It seems like you can not only forward one port, but many ports and in whatever direction you want at the same time.
To try that you have things working:
wget localhost:8000
In my case I wanted apt package manager to use the forwarded port so I just changed it to use http://localhost:8000 to access the packages.
@scottalanmiller said in PATH Error with TimeScaleDB-Tune on Fedora 30:
@Pete-S said in PATH Error with TimeScaleDB-Tune on Fedora 30:
@scottalanmiller said in PATH Error with TimeScaleDB-Tune on Fedora 30:
@Pete-S said in PATH Error with TimeScaleDB-Tune on Fedora 30:
@scottalanmiller I've had my eye on timescaledb for a while looking for a chance to test it. What do you use it for?
We are prepping to move Zabbix to it.
Yes, I saw that they had support for that but wasn't that feature experimental? Or is it mature now?
Experimental but why would we care
But since you make money with your monitoring system, doesn't that make it financial data. 
@scottalanmiller said in Application Virtualization in Linux Environment:
@Pete-S said in Application Virtualization in Linux Environment:
@Dashrender said in Application Virtualization in Linux Environment:
And it requires windows 10 Pro, not home!
Why would you use home edition for anything? Aren't we pros?
We actually use Home most of the time as we don't use AD 50% of the time. Pro has very few useful features other than AD joining.
Hmm, I never given it that much thought since the machines we buy usually comes with pro or they are servers. For machines that doesn't come with windows, for instance NUCs, we just get the oem pro version. Looking at it now, we would pay $30 less for the home version. Makes no real difference for us but we don't buy many machines either.