USB3 back working in 4.0.2-1-ARCH

[technosf]

Thanks to moonman's prompt in another post, I can confirm that USB3 is working again in 4.0.2-1 on my NSA325

I did not upgrade systemd from 218-2, so cannot confirm the recent issues there are gone. Can anyone confirm?

Thank you!

[haiku]

Thanks, the USB3 works again on my pogo V4
ps: I run systemd 219

[maddog]

Is updating to systemd-219-6 safe ? Running a pogo e02 and ignored the updates for now because of the forum reports.

[technosf]

So just updated to 4.0.3-1-ARCH and everything is copacetic with the kernel... And I decided to venture on from Systemd 218 to Systemd 219 (ensuring I still had the 2.1.8 packages around). Here's what I found, with no analysis etc.

Up comes the OS... The internal network is up. Ufw is up. I can ping the outside world from the server. I can resolve from clients. What I can't do is ping outside addresses from clients. I.e. can't access the web from network clients - 219 has changed something that affect the network.

Rolled back to 218 and everything is AOK.

[maddog]

Thanks a lot for taking the time and posting the situation about systemd 219, I'll guess I'll wait on future revisions.

[moonman]

Read up on how networkd works https://wiki.archlinux.org/index.php/Systemd-networkd
And check for ipv6 nameservers in your config if you don't have ipv6. All my devices are up to date and work fine

[technosf]

No IPv6 addresses in resolve.conf... Not even ::1

It resolves just fine... On the server. It's only on clients that use the server as a router (which it is). I use dnamasq to dole out IPv4 addresses and to do name resolution, providing the external upstream name servers in the dnsmasq.conf.

Eveything fine in 218, brokey in 219....

So, as I write, it occurs to me that forwarding has been affected: I Googled and found this: https://bbs.archlinux.org/viewtopic.php?id=196815

Looks like poss. a new systemd setting. Will try and let y'all know.

[technosf]

And that was it... New setting in systemd 219. Judicous use of:

$this->bbcode_second_pass_code('', '[Network]
IPForward=ipv4
IPMasquerade=yes')

on all interface def files in /etc/systemd/network solved it. Will try removing setting from internal interfaces one by one to find the appropriate settings there.

Woo hoo!

Well, that was a surprise. But I still think systemd is the best thing since screw-cap wine