lanefu

Which distro? Is it part of a different package?

I just remember all the buzz with pov-ray, but it was like "write a text file" . was way over my head. but now i kinda wanna play with it on a farm of sbcs

I've been doing a little housekeeping with github issues and PR's to eliminate noise or bring closure. I've created a STALE label for issues that lack closure or don't have clear actionable items and closing. Hopefully that makes them easy to revisit later if one is interested. I've also posted some follow-ups on a few to try to get things moving and merged a few PRs while applying extreme discretion. Let me know if I do anything you disagree with or want me to back off. I'm trying to be mindful of everyone's work.

I moved outbound email to a new SMTP server. This should improve deliverability. My testing shows things being good. If you notice that you aren't receiving emails, or new spam flags just reply to this thread. Again, I just changed the server that the forum sends emails from, nothing else.

I feel like you missed the part where I said that netdata is awesome. Anyway, in all seriousness thanks for clarifying the purpose of RPI Monitor, its longer polling interval, lighter footprint, etc would definitely make more sense than netdata. armbianmonitor -m is my goto on all my SBCs for checking utilization. For the sake of providing a general defense for netdata, my set of the pants testing on my Opi Prime while glancing at armbianmonitor, htop, etc has shown its footprint to be pretty minimal-- and much less jarring than when Observium's snmp poller hits it. It looks like there are several fairly easy ways to add more metrics to it. I got consul and nomad data via its built in statsd collector. If you haven't used it on your own gear in a few years, its probably worth another look for other potential use cases.

I just recently came across netdata. Have you all seen it before... it's pretty great.... its kind of like a free version of datadog. I was wondering if there would be interest in replacing the armbianmonitor -r rpimontor with it.. or augmenting as another install option

man they're all super cool.

here's some NFS boot breadcrumbs that arent armbian specific but you'll need to update your initrams image with a few settings which uboot will pool. see /etc/initramfs-tools/initramfs.conf MODULES=netboot also you'll want to set your rootfs to be something like this # UNCONFIGURED FSTAB FOR BASE SYSTEM /dev/nfs / nfs defaults 1 1 /proc /proc proc defaults 0 0 /sys /sys sysfs defaults 0 0 /dev/fd0 /mnt/floppy auto user,noauto,sync,exec,umask=000 0 0 /dev/cdrom /mnt/cdrom auto user,noauto,exec,ro 0 0 none /tmp tmpfs defaults 0 0 none /var/run tmpfs defaults 0 0 none /var/lock tmpfs defaults 0 0 none /var/tmp tmpfs defaults 0 0

Just curious why you advise against it.. I use to run a PXE Boot a Ubuntu Desktop on a C2D and run over NFS and it worked pretty great... I even used docker with it? Is it the network stack on the SBC thats the concern?

They're two different flavors of Linux... Debian Stretch and Ubuntu Bionic Beaver. Re: desktop flavor ... Typically happens when the community has video drivers etc figured out.... it's always the weakest link in the SBC world... You'll see lots of conversations and questions about SBC video and GPU etc.

Step 1 get a serial console... do you have a TTL serial adapter? so those 3 or 4 holes in the bottom of the board..... get an output and then take it from there.

some more quick feedback. can you just clean up the code snippets that were added from the readme? right now you can't just cut and paste them and run because they have > on each line i will test the code out tho i promise

hey is that just for building u-boot or u-boot with the UEFI magic used on the new libre computer images?

Firs of all I admire everyone's quest for speed. Finding every bit of performance available in these boards is a lot of fun..... But allow me to provide a a brief sermon on link aggregation to manage expectations for those who many not have much experience. So I've gone down the NIC bonding rabbit hole many times on many pieces of equipment. I even have LACP trunks going to my garage on principle. It's really really hard to get a performance payoff on bonded gig links. Single TCP streams are still only sent down 1 link at a time. So things like iperf testing, most basic file IO tests etc, won't even send traffic on more than 1 link. The smarter hashing algorithms will load balance the links to a degree typically based on MAC or IP:port for service. Aka servers with dozens and dozens of clients work well because the bandwidth accross the 2 links can be distributed. Even protocols like NFS 4.1 and SMB3.0 that have concepts of parallism still don't perform well with just 2 endpoints. Typically it's been backup servers that are being absolutely pounded at the same time by many nodes at night, or VM hosts that I've ever gotten over 1 gig of traffic on LACP. The best performance I've been able to get using multiple gig links is by properly impementing iscsi with several initiators on a node and multipathing. Since iSCSI is multipathing SCSI IO calls and not tcp it's able to deal wit the out of order stuff and you can really hammer your links. Best performance bet probably is using 10G and maybe you'll hit a few gigsabits on that. IMHO I'd just focus on shaving latency down on your services and keep your networking stack simple. If performance is really an issue... i'd probaly just brute force with a junky intel box.. or by a more purpose built board. PS in defense of the 1 gig SGMII lane using for the topaz switch to the CPU.... that's still 1GIG of full duplex traffic.. aka.. that's 1GIG of potential packet forwarding/filtering between WAN and LAN on the topaz. In theory it's more potent than my edge router lite.... i'll have to test one day.

Yep. A positive spin is i can just set governors to performance and that big heat sink obliges. Something to be said for stability. It would be interesting to compare my opi prime to potato at some point Sent from my SM-G950U using Tapatalk

So i tried the nightly libre computer official image with the mali mail drivers, but the desktop was flaky. The unaccelerated Armbian image was fast and even faster than tritium h5. I just got a emmc module for potato yesterday, so im working on an ansible job to do my desktop setup quickly then ill cutover to le potato. Sent from my SM-G950U using Tapatalk

Okay I setup SES and DM'd you the info... your deliverability should go way up with being able to use their DKIM signing.

So I built some fresh Armbian images with dev kernels, and setup ubuntu mate and ran them on my tritium h5, and my le potato... and like I've got slack running in chromium n stuff, and desktop switching is pretty decent... and that's without acceleration.. just good 'ole brute force. Basically my mind is blown that its usable and I'm going to retire my c2d dell desktop that's on my KVM and finally use an ARM SBC as a desktop!

Thanks for sharing, dude. That is super cool! Sent from my SM-G950U using Tapatalk

Ya i think this can be done super cheap with amazon ses. I'll investigate more today and let you know what has to be done. How is your outbound mail for the forum currently configured? https://aws.amazon.com/ses/pricing/ Sent from my SM-G950U using Tapatalk

FYI even with a vendor you’ll still need to do those DNS records, but they’ll tell you what to enter. So about 1000 messages a day is the outbound volume? Sent from my iPad using Tapatalk

@nicod I saw it on the other thread. It’s worthy of its own post. That rig captures the essence of what makes hacking on SBCs so great. Lane Sent from my iPad using Tapatalk

As a whole your inquiry kind of pains me... but here's the truthiest answer: If you have data you want people to have, don't store it on the device. Have it pull from the cloud and store in memory. Or encrypt on disk, and decryption key is pulled from cloud and stored in memory. Luckily for you, that means it can be solved via network and application layers.

Ya you nailed what id been thinking. Some pretty big projects work out of github. So i think learning to master its functionality and even look at github bots if needed is a great path forward If the process above is properly captured in documentation and a clear policy for moderators to guide forum issues into github would be a huge step in the right direction without forklifting things to a new platform. I tried design a process a few years but, but i got it backwards and was trying to redirect issue dialog back to the forums. Sent from my SM-G950U using Tapatalk

Ya I kind of meant what's an open source project similar to armbian and how do they handle development vs community interaction...etc