OK, that's what I figured. Thanks and sorry for the bother. I wasn't sure that it was OK to replace the kernel on a running system and thought it was safer to do it off-line.. Got it now, I hope.
↧
Re: Debian on Dell Kace M300
↧
Re: Debian on Dell Kace M300
So far, so good:
And here are my stress tests:
Are we at point yet where we feel confident setting the uBoot environment variables to make it boot without having to have a serial console connection? I assume we still need to set them from stock OS...
Neal
root@debian:~# uname -a
Linux debian 5.2.9-kirkwood-tld-2 #1 PREEMPT Wed Aug 28 03:55:21 PDT 2019 armv5tel GNU/Linux
root@debian:~# free -h
total used free shared buff/cache available
Mem: 1.7Gi 15Mi 1.7Gi 0.0Ki 44Mi 1.7Gi
Swap: 0B 0B 0B
And here are my stress tests:
root@debian:~# stress-ng --vm 2 --vm-bytes 1700M --timeout 60s stress-ng: info: [1618] dispatching hogs: 2 vm stress-ng: info: [1618] cache allocate: using defaults, can't determine cache details from sysfs stress-ng: info: [1618] successful run completed in 60.36s (1 min, 0.36 secs) root@debian:~# stress-ng --cpu 64 --cpu-method all --verify -t 5m --metrics-brief stress-ng: info: [1624] dispatching hogs: 64 cpu stress-ng: info: [1624] cache allocate: using defaults, can't determine cache details from sysfs stress-ng: info: [1624] successful run completed in 306.70s (5 mins, 6.70 secs) stress-ng: info: [1624] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s stress-ng: info: [1624] (secs) (secs) (secs) (real time) (usr+sys time) stress-ng: info: [1624] cpu 1935 303.16 305.97 0.13 6.38 6.32 root@debian:~# stress-ng --cpu 64 --cpu-method all --verify -t 5m --metrics-brief --vm 2 --vm-bytes 1700M stress-ng: info: [1690] dispatching hogs: 64 cpu, 2 vm stress-ng: info: [1690] cache allocate: using defaults, can't determine cache details from sysfs stress-ng: info: [1690] successful run completed in 303.74s (5 mins, 3.74 secs) stress-ng: info: [1690] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s stress-ng: info: [1690] (secs) (secs) (secs) (real time) (usr+sys time) stress-ng: info: [1690] cpu 1241 301.90 293.83 0.05 4.11 4.22 stress-ng: info: [1690] vm 0 303.20 7.74 1.49 0.00 0.00
Are we at point yet where we feel confident setting the uBoot environment variables to make it boot without having to have a serial console connection? I assume we still need to set them from stock OS...
Neal
↧
↧
Re: Debian on Dell Kace M300
All good now, running 5.2.9 -tld-2 kernel and 5.2.9 filesystem. Took less time than my screwing around yesterday trying to save time. Story of my Linux experience.
I'm also ready to get rid of the serial connection if we can. I have root access to the stock OS.
I'm also ready to get rid of the serial connection if we can. I have root access to the stock OS.
↧
Re: Debian on Dell Kace M300
All,
Did you use the box web page during first time installation? and how did you gain root to stock OS without serial console (if you did)?
Did you use the box web page during first time installation? and how did you gain root to stock OS without serial console (if you did)?
↧
Re: Debian on Dell Kace M300
Bohdi posted this in June 2018:
With this clue, I did the same and can log in as root with my password, via serial, of course. I think the stock OS does not do ssh; I think this was discussed upthread, but couldn't find it quickly.
I did not do anything with the stock web interface beyond a quick look.
Quote
I did it the easy way. After I've booted with the Debian-4.12.1-kirkwood-tld-1-rootfs-bodhi.tar.bz2 rootfs, I mounted the internal SATA drive root partition and chroot into it, and reset root password. So now I can login to the box stock OS as root, if needed.
With this clue, I did the same and can log in as root with my password, via serial, of course. I think the stock OS does not do ssh; I think this was discussed upthread, but couldn't find it quickly.
I did not do anything with the stock web interface beyond a quick look.
↧
↧
Re: Debian buster 10.0 on Armada boxes -- udev problem with MAC address
Hi 1000001101000,
> On the Armada 370/XP devices I've worked with the
> mainline kernel isn't able to read the ATAG values
> containing the MAC address(es) though there is a
> patch available that adds that support. Is that
> still the case or has that feature made it into
> the mainline kernel? or is that something you've
> resolved with a newer u-boot?
I don't think ATAG is a problem in the kernel anymore for a while. Here is a data point for the Armada 385 (the reason I think systemd-udev is the problem, i.e. perhaps a race condition).
This configuration works fine
> I'm currently using a script in if-pre-up.d to
> grab the MAC address out of a u-boot environment
> variable and set them for the interface when
> needed. I doubt that approach makes any sense for
> what you're doing, it wouldn't make sense for me
> either if i had a better solution that could still
> work with the vanilla Debian kernel.
What if you run your script this way in /etc/rc.local (or use a systemd equivalence). That way an upgrade to vanilla Debian kernel has no effect on it.
I think if I can't make any progress on this, I will just put this work around in /etc/rc.local. Hopefully this will not cause too much network interruption.
> On the Armada 370/XP devices I've worked with the
> mainline kernel isn't able to read the ATAG values
> containing the MAC address(es) though there is a
> patch available that adds that support. Is that
> still the case or has that feature made it into
> the mainline kernel? or is that something you've
> resolved with a newer u-boot?
I don't think ATAG is a problem in the kernel anymore for a while. Here is a data point for the Armada 385 (the reason I think systemd-udev is the problem, i.e. perhaps a race condition).
Thecus N2350 Linux version 5.2.9-mvebu-tld-1 (root@tldDebianVM) (gcc version 7.4.0 (Ubuntu/Linaro 7.4.0-1ubuntu1~18.04.1)) #1 SMP PREEMPT Mon Aug 19 12:47:39 PDT 2019 Debian 9.8
ii udev 232-25+deb9u11 armel /dev/ and hotplug management daemon
This configuration works fine
[ 4.140060] mvneta f1070000.ethernet eth0: Using hardware mac address ......
> I'm currently using a script in if-pre-up.d to
> grab the MAC address out of a u-boot environment
> variable and set them for the interface when
> needed. I doubt that approach makes any sense for
> what you're doing, it wouldn't make sense for me
> either if i had a better solution that could still
> work with the vanilla Debian kernel.
What if you run your script this way in /etc/rc.local (or use a systemd equivalence). That way an upgrade to vanilla Debian kernel has no effect on it.
ifdown #run script that reads ethaddr ifup
I think if I can't make any progress on this, I will just put this work around in /etc/rc.local. Hopefully this will not cause too much network interruption.
↧
Re: Debian on Dell Kace M300
bodhi Wrote:
-------------------------------------------------------
> All,
>
> Did you use the box web page during first time
> installation? and how did you gain root to stock
> OS without serial console (if you did)?
I used serial console. I also mounted the SATA and chroot to reset the root password. I never looked at the web page.
Neal
-------------------------------------------------------
> All,
>
> Did you use the box web page during first time
> installation? and how did you gain root to stock
> OS without serial console (if you did)?
I used serial console. I also mounted the SATA and chroot to reset the root password. I never looked at the web page.
Neal
↧
Re: Debian on Dell Kace M300
I'm busy with other activities currently, so here are some questions pehaps some of us can investigate.
It would be nice to see what the web page provides. Such as whether SSH can be enabled from there.
Also, is the telnet port opened? if it is, what is the default credential to log in as root?
It would be nice to see what the web page provides. Such as whether SSH can be enabled from there.
Also, is the telnet port opened? if it is, what is the default credential to log in as root?
↧
Re: Linux Kernel 5.2.9 Kirkwood package and Debian rootfs
Thanks for the great work you have done. I just successfully turn my long bricked pogoplug v4 back to life.
But I do notice one thing. Unlike the old kernel, on these new kernel the BogoMIPS seems always be 333.33. Is that normal? The frequency info from sysfs looks alright though.
But I do notice one thing. Unlike the old kernel, on these new kernel the BogoMIPS seems always be 333.33. Is that normal? The frequency info from sysfs looks alright though.
↧
↧
Re: apt-get upgrade, systemd refuses to reload daemons
On the u-boot with uEnv.txt support you can simply add
initramfs.runsize=23MTo the boot parameter. Since I'm using the bodhi's u-boot this will work for me
custom_params=initramfs.runsize=23M
↧
Re: Linux Kernel 5.2.9 Kirkwood package and Debian rootfs
yuko1225,
> Thanks for the great work you have done. I just
> successfully turn my long bricked pogoplug v4 back
> to life.
Cool!
> But I do notice one thing. Unlike the old kernel,
> on these new kernel the BogoMIPS seems always be
> 333.33. Is that normal? The frequency info from
> sysfs looks alright though.
BogoMIPS has been unreliable for a long time. I don't think BogoMIPS is maintained anymore. I remmember a few years ago I had a patch to correct that (for SMP) but decided not to add it into the kernel.
> Thanks for the great work you have done. I just
> successfully turn my long bricked pogoplug v4 back
> to life.
Cool!
> But I do notice one thing. Unlike the old kernel,
> on these new kernel the BogoMIPS seems always be
> 333.33. Is that normal? The frequency info from
> sysfs looks alright though.
BogoMIPS has been unreliable for a long time. I don't think BogoMIPS is maintained anymore. I remmember a few years ago I had a patch to correct that (for SMP) but decided not to add it into the kernel.
↧
Re: Linux Kernel 5.2.9 Kirkwood package and Debian rootfs
yuko1225,
That reminded me, do you have cpufrequtils package install? that would be much more accurate to see the CPU clock.
That reminded me, do you have cpufrequtils package install? that would be much more accurate to see the CPU clock.
↧
Re: Linux Kernel 5.2.9 Kirkwood package and Debian rootfs
bodhi Wrote:
-------------------------------------------------------
> yuko1225,
>
> That reminded me, do you have cpufrequtils
> package install? that would be much more accurate
> to see the CPU clock.
Thanks for the reply.
https://lkml.org/lkml/2017/3/2/640
It's seems as long as the clock source is fine there is nothing need to worry about.
Thanks.
-------------------------------------------------------
> yuko1225,
>
> That reminded me, do you have cpufrequtils
> package install? that would be much more accurate
> to see the CPU clock.
Thanks for the reply.
cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to cpufreq@vger.kernel.org, please.
analyzing CPU 0:
driver: kirkwood-cpufreq
CPUs which run at the same hardware frequency: 0
CPUs which need to have their frequency coordinated by software: 0
maximum transition latency: 5.0 us.
hardware limits: 200 MHz - 800 MHz
available frequency steps: 800 MHz, 200 MHz
available cpufreq governors: powersave, ondemand, conservative, userspace, performance
current policy: frequency should be within 200 MHz and 800 MHz.
The governor "performance" may decide which speed to use
within this range.
current CPU frequency is 800 MHz (asserted by call to hardware).
It's looks fine. And I find a thread about it.
https://lkml.org/lkml/2017/3/2/640
[ 0.000000] clocksource: orion_clocksource: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 11467562657 ns [ 0.000013] sched_clock: 32 bits at 166MHz, resolution 6ns, wraps every 12884901885ns [ 0.000072] Switching to timer-based delay loop, resolution 6ns [ 0.001196] Console: colour dummy device 80x30 [ 0.001274] Calibrating delay loop (skipped), value calculated using timer frequency.. 333.33 BogoMIPS (lpj=1666666)
It's seems as long as the clock source is fine there is nothing need to worry about.
Thanks.
↧
↧
Re: Debian on Dell Kace M300
Post last year on ssh from the web interface:
Do we need to scan more ports than the just the normal telnet port 23?
Quote
JDS420
In order to enable ssh the way Dell intended, you have to enter a "tether key" which allows customer support to ssh into the box and undo/fix anything someone may have broken. Without a legitimate key, you have you purge all openssh packages and reinstall them. You might have to remove the kace-tether app too, but it's been months since I messed with the internals so I don't remember for sure
Do we need to scan more ports than the just the normal telnet port 23?
↧
Re: Debian on Dell Kace M300
> Do we need to scan more ports than the just the
> normal telnet port 23?
If you are inside stock OS, it is easier to see which daemon is running, and what ports are opened.
If no SSH or telnet, we'll just have to live with serial console installation for now.
I'll look at stock u-boot while it is running to see what else we can do.
> normal telnet port 23?
If you are inside stock OS, it is easier to see which daemon is running, and what ports are opened.
If no SSH or telnet, we'll just have to live with serial console installation for now.
I'll look at stock u-boot while it is running to see what else we can do.
↧
Re: Debian on Dell Kace M300
Looks like there is an sshd running:
If you kill it and restart it with -ddd to get some debug info you can see it is listening on port 22 but it's binding to loopback only:
If you comment these two lines out in /etc/ssh/sshd_config and restart it you can then log in:
Of course this doesn't really help us get access to a pristine box without serial console.
Neal
root@gilrain:~# ps -ef | grep sshd root 725 1 0 16:36 ? 00:00:00 /usr/sbin/sshd root 22517 21824 0 23:32 ttyS0 00:00:00 grep sshd root@gilrain:~#
If you kill it and restart it with -ddd to get some debug info you can see it is listening on port 22 but it's binding to loopback only:
root@gilrain:/usr/sbin# /usr/sbin/sshd -ddd
debug2: load_server_config: filename /etc/ssh/sshd_config
debug2: load_server_config: done config len = 688
debug2: parse_server_config: config /etc/ssh/sshd_config len 688
debug3: /etc/ssh/sshd_config:5 setting Port 22
debug3: /etc/ssh/sshd_config:7 setting ListenAddress ::1
debug3: /etc/ssh/sshd_config:8 setting ListenAddress 127.0.0.1
debug3: /etc/ssh/sshd_config:10 setting Protocol 2
debug3: /etc/ssh/sshd_config:12 setting HostKey /etc/ssh/ssh_host_rsa_key
debug3: /etc/ssh/sshd_config:13 setting HostKey /etc/ssh/ssh_host_dsa_key
debug3: /etc/ssh/sshd_config:15 setting UsePrivilegeSeparation yes
debug3: /etc/ssh/sshd_config:18 setting KeyRegenerationInterval 3600
debug3: /etc/ssh/sshd_config:19 setting ServerKeyBits 768
debug3: /etc/ssh/sshd_config:22 setting SyslogFacility AUTH
debug3: /etc/ssh/sshd_config:23 setting LogLevel INFO
debug3: /etc/ssh/sshd_config:26 setting LoginGraceTime 120
debug3: /etc/ssh/sshd_config:27 setting PermitRootLogin yes
debug3: /etc/ssh/sshd_config:28 setting StrictModes yes
debug3: /etc/ssh/sshd_config:30 setting RSAAuthentication yes
debug3: /etc/ssh/sshd_config:31 setting PubkeyAuthentication yes
debug3: /etc/ssh/sshd_config:35 setting IgnoreRhosts yes
debug3: /etc/ssh/sshd_config:37 setting RhostsRSAAuthentication no
debug3: /etc/ssh/sshd_config:39 setting HostbasedAuthentication no
debug3: /etc/ssh/sshd_config:44 setting PermitEmptyPasswords no
debug3: /etc/ssh/sshd_config:48 setting ChallengeResponseAuthentication no
debug3: /etc/ssh/sshd_config:63 setting X11Forwarding yes
debug3: /etc/ssh/sshd_config:64 setting X11DisplayOffset 10
debug3: /etc/ssh/sshd_config:65 setting PrintMotd no
debug3: /etc/ssh/sshd_config:66 setting PrintLastLog yes
debug3: /etc/ssh/sshd_config:67 setting TCPKeepAlive yes
debug3: /etc/ssh/sshd_config:74 setting AcceptEnv LANG LC_*
debug3: /etc/ssh/sshd_config:76 setting Subsystem sftp /usr/lib/openssh/sftp-server
debug3: /etc/ssh/sshd_config:87 setting UsePAM yes
debug1: sshd version OpenSSH_5.5p1 Debian-6
debug3: Not a RSA1 key file /etc/ssh/ssh_host_rsa_key.
debug1: read PEM private key done: type RSA
debug1: Checking blacklist file /usr/share/ssh/blacklist.RSA-2048
debug1: Checking blacklist file /etc/ssh/blacklist.RSA-2048
debug1: private host key: #0 type 1 RSA
debug3: Not a RSA1 key file /etc/ssh/ssh_host_dsa_key.
debug1: read PEM private key done: type DSA
debug1: Checking blacklist file /usr/share/ssh/blacklist.DSA-1024
debug1: Checking blacklist file /etc/ssh/blacklist.DSA-1024
debug1: private host key: #1 type 2 DSA
debug1: rexec_argv[0]='/usr/sbin/sshd'
debug1: rexec_argv[1]='-ddd'
debug3: oom_adjust_setup
Set /proc/self/oom_adj from 0 to -17
debug2: fd 3 setting O_NONBLOCK
debug1: Bind to port 22 on 127.0.0.1.
Server listening on 127.0.0.1 port 22.
socket: Address family not supported by protocol
If you comment these two lines out in /etc/ssh/sshd_config and restart it you can then log in:
# Use these options to restrict which interfaces/protocols sshd will bind to
#ListenAddress ::1
#ListenAddress 127.0.0.1
Of course this doesn't really help us get access to a pristine box without serial console.
Neal
↧
Re: Debian on Dell Kace M300
Good work, Neal!
OK, got ssh working in stock OS per the above, but also had to edit the /etc/network/interfaces file to set the the eth0 interface to dhcp. It was static 192.168.2.1 or something like that. My network is 192.168.1.x.
OK, got ssh working in stock OS per the above, but also had to edit the /etc/network/interfaces file to set the the eth0 interface to dhcp. It was static 192.168.2.1 or something like that. My network is 192.168.1.x.
↧
↧
Re: Debian on Dell Kace M300
Neal,
> Of course this doesn't really help us get access
> to a pristine box without serial console.
Right :)
Is port 23 open?
> Of course this doesn't really help us get access
> to a pristine box without serial console.
Right :)
Is port 23 open?
↧
Re: Debian on Dell Kace M300
U-boot tools seem to be in stock OS as fw_printenv works.
Summarizing where we are:
-Prepare a USB with bohdi's latest 5.2.9-tld-2 kernel and file system
-Connect serial to M300
-Boot into bohdi's file system setting u-boot commands through serial
-Change root password in stock OS by mounting internal SSD and using CHROOT
-Edit /etc/ssh/sshd_config to enable listening in other addresses
-Edit /etc/network/interfaces to set eth0 to DHCP (maybe)
-Reboot stock OS with ssh
Clearly, more detailed instructions are needed.
Summarizing where we are:
-Prepare a USB with bohdi's latest 5.2.9-tld-2 kernel and file system
-Connect serial to M300
-Boot into bohdi's file system setting u-boot commands through serial
-Change root password in stock OS by mounting internal SSD and using CHROOT
-Edit /etc/ssh/sshd_config to enable listening in other addresses
-Edit /etc/network/interfaces to set eth0 to DHCP (maybe)
-Reboot stock OS with ssh
Clearly, more detailed instructions are needed.
↧
Re: Debian on Dell Kace M300
Only ports 22 and 80 are open.
↧