Posts made by scottalanmiller

The most effective meetings are the ones that you don't have. - Quote (by me)

Turns out it is an App Armor restriction that Unity hasn't mentioned anywhere in their documentation even though we are at LEAST two releases into this problem. The fix is simple, but should be part of the install docs.

This is the error for the search engines:

No usable sandbox! Update your kernel or see Chromium Docs - Linux SUID Sandbox Development 48 for more information on developing with the SUID sandbox. If you want to live dangerously and need an immediate workaround, you can try using --no-sandbox.

To Test:

sudo sysctl -w kernel.apparmor_restrict_unprivileged_userns=0

And then for the real fix: /etc/sysctl.d/unity.conf

kernel.apparmor_restrict_unprivileged_userns=0

But my CEO always needs help with EVERYTHING

@EddieJennings said in What Are You Doing Right Now:

Reading about trusts between Red Hat Identity Management and Active Directory.

YOu'll enjoy this... So a mutual old customer of ours (wink wink) that fired us over three years ago because we made it obvious that their new CEO was laundering, called us to beg for us to rebuild their infrastructure because they've not maintained anything for three years (even though money flows to an MSP every month AND they have a whole internal IT team now) and they don't know how to get their servers online. They literally called to ask us HOW TO SIGN IN to the servers that THEY have "maintained" with TWO companies for THREE years. LOL

We said sure, but I don't think you are in our billing system anymore (we already wrote down the $38K that they stole from us), we just need to set that up. And they literally acted shocked that they'd have to PAY us to do their jobs for them. They actually thought that we'd do all the work of their IT department, that they happily pay two organizations to not do already, for free!

Just back from two weeks in Argentina with @valentina

@IThomeboy80 said in Proxmox os backup and restore:

Restore:

Log in to the Proxmox web interface.

Select the node or server where you want to restore the backup.

That's for restoring VMs, not ProxMox. By the time you have this interface available, you've already restored ProxMox.

@IThomeboy80 said in Proxmox os backup and restore:

Note: It is essential to follow the backup and restore process carefully to avoid data loss or system failure. Always test the backup and restore process before performing it on a production system.

Actually, it should have no data loss. No data should ever be stored outside of the VM files.

@gjacobse said in Proxmox os backup and restore:

@ronneyb said in Proxmox os backup and restore:

So what is the best way to setup a single pve so that we can backup and restore the os. I have not been able to backup and restore using clonezilla. Maybe the proxmox backup server ? Any ideas and answers will be appreciated

Thanks
Ronney

Just for my own clarification, are you saying you want to backup/restore the Proxmox system?

Yeah, that's what he is doing.

@ronneyb So the main question is... why would you want to? There isn't much documentation on this because it's not generally considered a good thing to do. You normally just want to reinstall fresh.

Youtube Video

@WrCombs said in Outage 7/19:

@Obsolesce said in Outage 7/19:

This is something a single test would have caught prior to releasing the update...

You'd think more would be implementing automated testing... apparently, some do no testing whatsoever.

Testing is overrated apparently - lol

Imagine how hard that would be to do in a place using third party AV. It might not even be provided as an option.

@Obsolesce testing AV updates is uncommon. Compared to other kinds of things. Outside of Wall St. I've never known any shop anywhere to both use third party AV and do testing of it.

@WrCombs It's really super minor. Companies with a clue about security and basic system design aren't really impacted. It requires running Windows on mission critical systems AND not trusting Microsoft AND choosing ClowdStrike as your MS security replacement (which CS seems fine on its own but it's a problematic category of products to start with) to end up in a "no sane IT department should arrive at this decision" point.

By definition, those impacted, didn't see these systems as truly critical.

Youtube Video

@gjacobse Most likely this is an SSH timeout. Might be on the client end. Is the timeout the same regardless of where you connect from? What if you connect via SSH instead of NX?

@Obsolesce said in What Are You Doing Right Now:

The forums are alive!

Yup, at lesat a little bit, jaja.

If you want to use Castopod, you almost certainly want to connect it to an S3 bucket. But the documentation on the site is missing a couple critical elements. The bucket must be specified and the format of the host needs the header. Here is an example configuration from the end of the .env file.

media.fileManager="s3"
media.s3.endpoint="https://s3.us-east-001.backblazeb2.com"
media.s3.bucket="bucketname"
media.s3.key="mykeyID"
media.s3.secret="mykeySecret"
media.s3.region="us-east-001"

This took some tracking down. The media.s3.bucket is absolutely critical, but does not appear in the documentation. The https:// I'm told is required as it prefixes bad data without it.

I'm been on a media creation splurge this week. So busy, but lots of good stuff happening.

@MattSpeller said in What Are You Doing Right Now:

Revisiting an old haunt to say hello

Hope y'all are doing awesome, greetings from the great white north.

OMG been a while, how are you?

@scottalanmiller said in MeshCentral Future:

@EddieJennings said in MeshCentral Future:

The one ask I have of MeshCentral is Wayland support.

Maybe, at some point, it's Wayland that needs to step it up.

1.24 is out now.