Greetings;
And I had just had it update everything, 162 of the 165 packages being from TDE.
From here, all indications of an extended to 20 minutes e2fsck or something. No response to a ping, no response to an attempted ssh -Y alias login. So I slip on a t-shirt and stumble out to the machine at 3AM, to be greeted by its own login greeter. So I logged in, all normal.
And now that I'm logged in there, everything else is also working normally.
So its own login has been moved up early so nothing in the way of services is running until after the local login.
This. to be blunt, is a disaster for the way I have normally worked to maintain these machines for over a decade.
Can this be unfixed? Its a right pain in the ass to be locked out of a machine because the local login hasn't been done.
Cheers, Gene Heskett
Blind guess: you are using debian, systemd killed sshd 'cause ssh did not get enough random data ... guess you run it on a rpi?
Nik
Anno domini 2019 Thu, 8 Aug 03:32:14 -0400 Gene Heskett scripsit:
Greetings;
And I had just had it update everything, 162 of the 165 packages being from TDE.
From here, all indications of an extended to 20 minutes e2fsck or something. No response to a ping, no response to an attempted ssh -Y alias login. So I slip on a t-shirt and stumble out to the machine at 3AM, to be greeted by its own login greeter. So I logged in, all normal.
And now that I'm logged in there, everything else is also working normally.
So its own login has been moved up early so nothing in the way of services is running until after the local login.
This. to be blunt, is a disaster for the way I have normally worked to maintain these machines for over a decade.
Can this be unfixed? Its a right pain in the ass to be locked out of a machine because the local login hasn't been done.
Cheers, Gene Heskett
On Thursday 08 August 2019 04:21:56 Dr. Nikolaus Klepp wrote:
Blind guess: you are using debian, systemd killed sshd 'cause ssh did not get enough random data ... guess you run it on a rpi?
Nik
Nope, debian wheezy yet on an old dell dimension 745 with a 2 core brain. Thats what LCNC runs on, although there is a stretch testing, but it comes with a broken kernel. The keyboard goes away, sometimes several times a day for the rt-preempt version, and in a week or so for the non-rt version of the same kernel. Thats not safe for running machinery.
But ATM they're working on a pi4 version. And I've a pi-4, 2G version on order. But ATM if you want buster on amd64, you still have to build your own rt kernel, and then build LCNC. AFAIK. I like debians net-installed buster but so far have been unable to prevent its self destruction after 5 or 6 reboots. With, or without a root pw set it invalidates your passwds and you are locked out till you bring the card back in and rewrite the netinstall stuff and start all over. 4 times now. IF it can be made to work, its not a damned u-booter, debians net-install actually uses grub to boot a pi-3b!. And that opens up lots of possibilities. I supposed I should go for loop 5 eventually on the pi-buster, & see if its been fixed. But theres no clue when its going to fail the next reboot.
This dell is the only machine I felt had enough iron in its butt to run TDE, but this may force me to go back to xfce4 which is working fine on 2 other intel atom boxes. If similar problem on the next reboot, TDE will be gone and xfce4 will replace it. It really is that simple for me.
That, and my ticker is threatening to quit despite a pacemaker, I spent the night before in the shop with angina pain, but the heart guy is out of pocket this week. So against their better judgement, I checked my self out and came home and am not exerting myself, basically taking care of Dee & the housework. And I feel pretty good so far. Ignore that faint knocking sound.
Thanks Nik.
Cheers, Gene Heskett