I remember removing the IPX route entries from our Cat65 MSFC back in 2006 and from the ATM/Framerelay WAN Equipment. Wasn't very popular with the customers.
I also remember the first IPv6 Workshop on W2k SP3 back in 2002. Not that long ago.
Thanks to the trend to SASE like Palo Alto GlobalProtect or ZScsler this practice is not a good idea anymore. Speaking of ZScaler, they are still IPv4 only, right?
Had Google trying to send me mails to non-existing mail-addresses over months. You would think their logs might catch something like that or they would react to my complaints ... they don't and they just dont care.
It sometimes stops for weeks, then it continiues.
from my logs as an example:
Nov 13 22:10:51 bert postfix/smtpd[2693931]: NOQUEUE: reject: RCPT from mail-oi1-x248.google.com[2607:f8b0:4864:20::248]: 450 4.1.8 <ki+bncBD77RLFFQACRBZOX3DEAMGQEU5V3LXY@zf.thesparklebar.com>: Sender address rejected: Domain not found; from=<ki+bncBD77RLFFQACRBZOX3DEAMGQEU5V3LXY@zf.thesparklebar.com> to=<rmayer13@nerd-residenz.de> proto=ESMTP helo=<mail-oi1-x248.google.com>
Nov 13 22:12:07 bert postfix/smtpd[2696594]: NOQUEUE: reject: RCPT from mail-ua1-x948.google.com[2607:f8b0:4864:20::948]: 450 4.1.8 <ki+bncBD77RLFFQACRBZOX3DEAMGQEU5V3LXY@zf.thesparklebar.com>: Sender address rejected: Domain not found; from=<ki+bncBD77RLFFQACRBZOX3DEAMGQEU5V3LXY@zf.thesparklebar.com> to=<rmayer1000@nerd-residenz.de> proto=ESMTP helo=<mail-ua1-x948.google.com>
Nov 13 22:12:18 bert postfix/smtpd[2696594]: NOQUEUE: reject: RCPT from mail-wm1-x346.google.com[2a00:1450:4864:20::346]: 450 4.1.8 <ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com>: Sender address rejected: Domain not found; from=<ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com> to=<rmayer13@nerd-residenz.de> proto=ESMTP helo=<mail-wm1-x346.google.com>
Nov 13 22:12:37 bert postfix/smtpd[2696594]: NOQUEUE: reject: RCPT from mail-lf1-x146.google.com[2a00:1450:4864:20::146]: 450 4.1.8 <ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com>: Sender address rejected: Domain not found; from=<ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com> to=<rmayer333@nerd-residenz.de> proto=ESMTP helo=<mail-lf1-x146.google.com>
Nov 13 22:13:08 bert postfix/smtpd[2696594]: NOQUEUE: reject: RCPT from mail-lj1-x248.google.com[2a00:1450:4864:20::248]: 450 4.1.8 <hc+bncBDO2ZDH5DIIOXB6ZZADBUBB2QEZ74@zf.thesparklebar.com>: Sender address rejected: Domain not found; from=<hc+bncBDO2ZDH5DIIOXB6ZZADBUBB2QEZ74@zf.thesparklebar.com> to=<rmayer@nerd-residenz.de> proto=ESMTP helo=<mail-lj1-x248.google.com>
Nov 13 22:13:08 bert postfix/smtpd[2696594]: NOQUEUE: reject: RCPT from mail-wm1-x345.google.com[2a00:1450:4864:20::345]: 450 4.1.8 <ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com>: Sender address rejected: Domain not found; from=<ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com> to=<rmayerrmayer@nerd-residenz.de> proto=ESMTP helo=<mail-wm1-x345.google.com>
Nov 13 22:14:03 bert postfix/smtpd[2696594]: NOQUEUE: reject: RCPT from mail-lj1-x248.google.com[2a00:1450:4864:20::248]: 450 4.1.8 <ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com>: Sender address rejected: Domain not found; from=<ki+bncBDO2ZDH5DIIOXB6ZZADBUBFIYC6HQ@zf.thesparklebar.com> to=<rmayera@nerd-residenz.de> proto=ESMTP helo=<mail-lj1-x248.google.com>
As you can see, the to-address is generated and its different hosts at google trying to send mails.
Searching for zf.thesparklebar.com shows others having the same problem.
Wait till you learn about companies replacing the opensource parts of their stack/products with something an AI coding agent produced. They do this to get rid of all the burden that comes with using opensource like risking to get sued if they dont ship the source code according to the licence. This is why sboms are a hot topic right now. Also coding agents are now good and cheep enough to do this.
Like car.org has monthly tasks for check fluids, clean inside, clean outside. In April and October theres a task to remind me to change tires and as I'm German a task to remind me yearly to check the first aid kit (you get a fine if its expired and they spot it when they stop you).
So, almost every appliance has its org-file with routines. Same for me and my family, check vaccinations, see the dentist ...
So, if you are already using org, theres no need for an additional tool. Plus, its all text, just throw it in your git. Also, there's tools like beorg to have your org on your iPhone.
It varies, but in many places you couldn't sell a device without proper shielding, but unless you are causing disruption to a public service or safety equipment there is nothing that stops a person owning and running one. Just think of the world full of rPi units, other SBCs, and other PCs, running completely caseless or in cases without much/any thought towards EMC shielding - it obviously isn't a problem, or it would be a big problem.
Most rPi units and similar are fine as they can be argued to be sold as parts rather than devices just like any other motherboard¹. The Pi400 presumably gets away with it, as something this is conspicuously sold as a device not a part, because that chonky heatsink² is enough to disrupt any errant EM fields outside the ranges that it should be emitting (those around 2.4GHz and 5GHz).
There are many grey areas, and indeed those where the letter of the regs is broken but not enforced. To cut a long story short wrt “Is this "legal" to run a pc open like that?”: yes running a PC in a case like that with no extra shielding is legal pretty much everywhere, though selling a complete PC with a case like that probably breaks regs and maybe even laws.
----
[1] putting the responsibility with the purchaser, where it isn't enforced unless it is a problem (I chose not to shield my TV-box Pi4, not the company, and it isn't putting enough junk out to disrupt anyone else's anything else)
I'm not sure what customers Synology is targeting. Small office/home office (SoHo) was their original market, but these customers won't be willing to pay high prices per drive.
Medium-sized businesses? They mostly move their infrastructure to the cloud, which probably leads to low sales volumes. Plus, they're very price-sensitive too.
Large enterprises and corporations? This is the domain of established providers like NetApp. Synology might dream about the high prices that these major storage vendors can charge, but this market is difficult to enter without years and years and years of proven reliability in hardware and service.
I don't think this will work the way Synology imagines it.
I also remember the first IPv6 Workshop on W2k SP3 back in 2002. Not that long ago.
reply