Hi,
I am running into problems with starting foundation model lately.
1. I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
2. It was booting fine initially.
3. The problem started happening after I attempted to build Groff package, which was not getting build and the model was getting frozen (no response to Ctrl + C, etc.). However I was able to restart the model by closing it from the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the middle, doesn't get to the command prompt).
Anyone knows whom should I report to about this?
Thanks, Anil
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org wrote:
Hi,
I am running into problems with starting foundation model lately.
- I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
It was booting fine initially.
The problem started happening after I attempted to build Groff package,
which was not getting build and the model was getting frozen (no response to Ctrl + C, etc.). However I was able to restart the model by closing it from the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the middle, doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather than using the serial terminal. And if a command locks a session, I log in again via SSH and kill the old one, rather than pull the plug on the model.
Anyone knows whom should I report to about this?
This is a good place to discuss it.
Thanks, Anil
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Hi Ryan,
Thanks for your reply.
RootFS getting corrupted by repeated <Ctrl+C>: Does that mean we have some known limitations / stability issues with the foundation model (or rootfs)? This was not an issue for past 6 months and has only showed up recently after I used the custom image.
So far I have generally launched the the model and used the model window itself to do build etc.. I remember having issues with doing SSH to the model from the host initially when I started using it. There was no steps mentioned in the user guide and I had tried various things without success. Few things I could do with the model were:
1. wget works 2. scp from host pc into the model works. 3. pinging an ip address from inside the model doesn't work (e.g "ping google.com" fails)
So, I don't know completely if and how to connect to the model via ssh so that I can avoid having to close a hung process by <Ctrl+c>. Could you please point me to some resource?
Thanks, Anil
On 3 February 2014 22:03, Ryan Harkin ryan.harkin@linaro.org wrote:
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org wrote:
Hi,
I am running into problems with starting foundation model lately.
- I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
It was booting fine initially.
The problem started happening after I attempted to build Groff
package,
which was not getting build and the model was getting frozen (no
response to
Ctrl + C, etc.). However I was able to restart the model by closing it
from
the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the middle, doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather than using the serial terminal. And if a command locks a session, I log in again via SSH and kill the old one, rather than pull the plug on the model.
Anyone knows whom should I report to about this?
This is a good place to discuss it.
Thanks, Anil
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
On 4 February 2014 06:30, Anil Singhar anil.singhar@linaro.org wrote:
Hi Ryan,
Thanks for your reply.
RootFS getting corrupted by repeated <Ctrl+C>: Does that mean we have some known limitations / stability issues with the foundation model (or rootfs)?
No. If you CTRL-C the *model* - not a running process within the model - you are effectively pulling the plug on the machine. This is the equivalent of you booting up and powering off your computer without cleanly unmounting the disk, you will eventually corrupt the disk if you keep doing it.
This was not an issue for past 6 months and has only showed up recently after I used the custom image.
If you were killing the model every time, you got lucky.
So far I have generally launched the the model and used the model window itself to do build etc.. I remember having issues with doing SSH to the model from the host initially when I started using it. There was no steps mentioned in the user guide and I had tried various things without success. Few things I could do with the model were:
- wget works
- scp from host pc into the model works.
- pinging an ip address from inside the model doesn't work (e.g "ping
google.com" fails)
Linaro images have SSHD set up already. You have a networking setup problem if you don't have DNS resolving hostnames, but if you have access via IP address, you should be able to SSH into a model easily.
So, I don't know completely if and how to connect to the model via ssh so
Try this from your host PC:
ssh root@<ip address of model instance>
Eg, ssh root@192.168.1.88
that I can avoid having to close a hung process by <Ctrl+c>. Could you please point me to some resource?
More advanced networking setup information is available from ARM's website as linked from the release notes:
http://releases.linaro.org/14.01/openembedded/aarch64/
I've never been happy with NAT networking, so I use a bridged setup.
Thanks, Anil
On 3 February 2014 22:03, Ryan Harkin ryan.harkin@linaro.org wrote:
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org wrote:
Hi,
I am running into problems with starting foundation model lately.
- I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
It was booting fine initially.
The problem started happening after I attempted to build Groff
package, which was not getting build and the model was getting frozen (no response to Ctrl + C, etc.). However I was able to restart the model by closing it from the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the middle, doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather than using the serial terminal. And if a command locks a session, I log in again via SSH and kill the old one, rather than pull the plug on the model.
Anyone knows whom should I report to about this?
This is a good place to discuss it.
Thanks, Anil
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Hi Ryan,
1. If I have to close the model from inside, what is the command? (So that don't have to do Ctrl+C from the launch window.
I have tried the following two networking options (nat and bridged):
2. With the following launch command (networking option used bridged):
./Foundation_v8 --image img-foundation.axf --block-device vexpress64-openembedded_lamp-armv8-gcc-4.8_20140126-59g --network=bridged --network-bridge=eth0
I get the following on the model:
root@genericarmv8:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:02:F7:EF:20:B3 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:195 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 B) TX bytes:62094 (60.6 KiB) Interrupt:47 Base address:0x2000 DMA chan:ff
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:784 (784.0 B) TX bytes:784 (784.0 B)
root@genericarmv8:~#
This doesn't help, there is no ip address. Could not find anywhere what value should be used for flag network-bridge. Is eth0 correct? The website mentions default is ARM0, don't know what does that mean.
3. With the following command (networking option used nat):
./Foundation_v8 --image img-foundation.axf --block-device vexpress64-openembedded_lamp-armv8-gcc-4.8_20140126-596.img --network=nat --network-nat-ports=8022=22
I get the following on the model:
root@genericarmv8:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:02:F7:EF:20:B3 inet addr:172.20.51.1 Bcast:172.20.51.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:28 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:646 (646.0 B) TX bytes:4788 (4.6 KiB) Interrupt:47 Base address:0x2000 DMA chan:ff
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:784 (784.0 B) TX bytes:784 (784.0 B)
root@genericarmv8:~#
This doesn't help either. I can't ping anywhere to the outside world, not even to the host itself. And "ssh root@172.20.51.1" doesn't work from the host (outside the model). But surprisingly wget works. Here the value of "network-nat-ports", i.e. 8022=22 is provided in the document.
I don't know what is it in the instructions that I missing. I am not a networking expert at all, so it is actually too much work figuring out all these details just to be able to use the model.
Could you please let me know the set of commands you use to launch the model so that you can ping outside world and also do ssh to the model.
Thanks and regards, Anil
On 4 February 2014 12:27, Ryan Harkin ryan.harkin@linaro.org wrote:
On 4 February 2014 06:30, Anil Singhar anil.singhar@linaro.org wrote:
Hi Ryan,
Thanks for your reply.
RootFS getting corrupted by repeated <Ctrl+C>: Does that mean we have
some
known limitations / stability issues with the foundation model (or
rootfs)?
No. If you CTRL-C the *model* - not a running process within the model - you are effectively pulling the plug on the machine. This is the equivalent of you booting up and powering off your computer without cleanly unmounting the disk, you will eventually corrupt the disk if you keep doing it.
This was not an issue for past 6 months and has only showed up recently after I used the custom image.
If you were killing the model every time, you got lucky.
So far I have generally launched the the model and used the model window itself to do build etc.. I remember having issues with doing SSH to the model from the host initially when I started using it. There was no steps mentioned in the user guide and I had tried various things without
success.
Few things I could do with the model were:
- wget works
- scp from host pc into the model works.
- pinging an ip address from inside the model doesn't work (e.g "ping
google.com" fails)
Linaro images have SSHD set up already. You have a networking setup problem if you don't have DNS resolving hostnames, but if you have access via IP address, you should be able to SSH into a model easily.
So, I don't know completely if and how to connect to the model via ssh so
Try this from your host PC:
ssh root@<ip address of model instance>Eg, ssh root@192.168.1.88
that I can avoid having to close a hung process by <Ctrl+c>. Could you please point me to some resource?
More advanced networking setup information is available from ARM's website as linked from the release notes:
http://releases.linaro.org/14.01/openembedded/aarch64/
I've never been happy with NAT networking, so I use a bridged setup.
Thanks, Anil
On 3 February 2014 22:03, Ryan Harkin ryan.harkin@linaro.org wrote:
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org wrote:
Hi,
I am running into problems with starting foundation model lately.
- I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
It was booting fine initially.
The problem started happening after I attempted to build Groff
package, which was not getting build and the model was getting frozen (no response to Ctrl + C, etc.). However I was able to restart the model by closing it from the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the
middle,
doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather than using the serial terminal. And if a command locks a session, I log in again via SSH and kill the old one, rather than pull the plug on the model.
Anyone knows whom should I report to about this?
This is a good place to discuss it.
Thanks, Anil
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
On Tue, 2014-02-04 at 14:32 +0530, Anil Singhar wrote:
- If I have to close the model from inside, what is the command?
'poweroff' from a command-line in the model, same as on real hardware. However, after the system has done it's clean shutdown of services and filesystems it doesn't actually shut the model down, you will still need to kill it by hand.
Hi Jon,
Thanks for your reply.
I am trying to get the full networking on the model. Right now I am not able to do it. Could you please help me with that?
I have used nat but it opens in half-duplex mode and for some mysterious reason "wget" works but normal "ping" doesn't work. My intention is: 1. To start the model 2. ssh into the model from outside and use it instead of using it from the model's xterm.
Thanks & Regards, Anil
On 4 February 2014 14:43, Jon Medhurst (Tixy) tixy@linaro.org wrote:
On Tue, 2014-02-04 at 14:32 +0530, Anil Singhar wrote:
- If I have to close the model from inside, what is the command?
'poweroff' from a command-line in the model, same as on real hardware. However, after the system has done it's clean shutdown of services and filesystems it doesn't actually shut the model down, you will still need to kill it by hand.
-- Tixy
On 4 February 2014 09:02, Anil Singhar anil.singhar@linaro.org wrote:
Hi Ryan,
- If I have to close the model from inside, what is the command? (So that
don't have to do Ctrl+C from the launch window.
I have tried the following two networking options (nat and bridged):
- With the following launch command (networking option used bridged):
./Foundation_v8 --image img-foundation.axf --block-device vexpress64-openembedded_lamp-armv8-gcc-4.8_20140126-59g --network=bridged --network-bridge=eth0
I get the following on the model:
root@genericarmv8:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:02:F7:EF:20:B3 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:195 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 B) TX bytes:62094 (60.6 KiB) Interrupt:47 Base address:0x2000 DMA chan:ff
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:784 (784.0 B) TX bytes:784 (784.0 B)
root@genericarmv8:~#
This doesn't help, there is no ip address. Could not find anywhere what value should be used for flag network-bridge. Is eth0 correct? The website mentions default is ARM0, don't know what does that mean.
eth0 is not correct. It should be "ARM$USER" as per the release notes. First, you follow the link in the release notes to create a network TAP, this creates a new interface on your computer called "ARM$USER", eg, ARMryan in my case. Then you add the options as described on the release note.
If you have problems with the TAP setup from ARM's website, then you need to contact support@arm.com.
- With the following command (networking option used nat):
./Foundation_v8 --image img-foundation.axf --block-device vexpress64-openembedded_lamp-armv8-gcc-4.8_20140126-596.img --network=nat --network-nat-ports=8022=22
I get the following on the model:
root@genericarmv8:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:02:F7:EF:20:B3 inet addr:172.20.51.1 Bcast:172.20.51.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:28 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:646 (646.0 B) TX bytes:4788 (4.6 KiB) Interrupt:47 Base address:0x2000 DMA chan:ff
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:784 (784.0 B) TX bytes:784 (784.0 B)
root@genericarmv8:~#
This doesn't help either. I can't ping anywhere to the outside world, not even to the host itself. And "ssh root@172.20.51.1" doesn't work from the host (outside the model). But surprisingly wget works. Here the value of "network-nat-ports", i.e. 8022=22 is provided in the document.
I don't know what is it in the instructions that I missing. I am not a networking expert at all, so it is actually too much work figuring out all these details just to be able to use the model.
Could you please let me know the set of commands you use to launch the model so that you can ping outside world and also do ssh to the model.
Thanks and regards, Anil
On 4 February 2014 12:27, Ryan Harkin ryan.harkin@linaro.org wrote:
On 4 February 2014 06:30, Anil Singhar anil.singhar@linaro.org wrote:
Hi Ryan,
Thanks for your reply.
RootFS getting corrupted by repeated <Ctrl+C>: Does that mean we have some known limitations / stability issues with the foundation model (or rootfs)?
No. If you CTRL-C the *model* - not a running process within the model - you are effectively pulling the plug on the machine. This is the equivalent of you booting up and powering off your computer without cleanly unmounting the disk, you will eventually corrupt the disk if you keep doing it.
This was not an issue for past 6 months and has only showed up recently after I used the custom image.
If you were killing the model every time, you got lucky.
So far I have generally launched the the model and used the model window itself to do build etc.. I remember having issues with doing SSH to the model from the host initially when I started using it. There was no steps mentioned in the user guide and I had tried various things without success. Few things I could do with the model were:
- wget works
- scp from host pc into the model works.
- pinging an ip address from inside the model doesn't work (e.g "ping
google.com" fails)
Linaro images have SSHD set up already. You have a networking setup problem if you don't have DNS resolving hostnames, but if you have access via IP address, you should be able to SSH into a model easily.
So, I don't know completely if and how to connect to the model via ssh so
Try this from your host PC:
ssh root@<ip address of model instance>Eg, ssh root@192.168.1.88
that I can avoid having to close a hung process by <Ctrl+c>. Could you please point me to some resource?
More advanced networking setup information is available from ARM's website as linked from the release notes:
http://releases.linaro.org/14.01/openembedded/aarch64/
I've never been happy with NAT networking, so I use a bridged setup.
Thanks, Anil
On 3 February 2014 22:03, Ryan Harkin ryan.harkin@linaro.org wrote:
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org wrote:
Hi,
I am running into problems with starting foundation model lately.
- I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
It was booting fine initially.
The problem started happening after I attempted to build Groff
package, which was not getting build and the model was getting frozen (no response to Ctrl + C, etc.). However I was able to restart the model by closing it from the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the middle, doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather than using the serial terminal. And if a command locks a session, I log in again via SSH and kill the old one, rather than pull the plug on the model.
Anyone knows whom should I report to about this?
This is a good place to discuss it.
Thanks, Anil
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
On 4 February 2014 10:02, Anil Singhar anil.singhar@linaro.org wrote:
- With the following command (networking option used nat):
./Foundation_v8 --image img-foundation.axf --block-device vexpress64-openembedded_lamp-armv8-gcc-4.8_20140126-596.img --network=nat --network-nat-ports=8022=22
I get the following on the model:
root@genericarmv8:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:02:F7:EF:20:B3 inet addr:172.20.51.1 Bcast:172.20.51.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:28 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:646 (646.0 B) TX bytes:4788 (4.6 KiB) Interrupt:47 Base address:0x2000 DMA chan:ff
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:784 (784.0 B) TX bytes:784 (784.0 B)
root@genericarmv8:~#
This doesn't help either. I can't ping anywhere to the outside world, not even to the host itself. And "ssh root@172.20.51.1" doesn't work from the host (outside the model). But surprisingly wget works. Here the value of "network-nat-ports", i.e. 8022=22 is provided in the document.
I don't know what is it in the instructions that I missing. I am not a networking expert at all, so it is actually too much work figuring out all these details just to be able to use the model.
Could you please let me know the set of commands you use to launch the model so that you can ping outside world and also do ssh to the model.
I don't need to go outside from within the model, but that's probably of network configuration of the linux distro installed in the model.
However, I am able to ssh to the model: ssh -p 8022 root@HOSTNAME (where HOSTNAME=name of the machine on which you start the model)
HTH
Christophe.
Thanks and regards, Anil
On 4 February 2014 12:27, Ryan Harkin ryan.harkin@linaro.org wrote:
On 4 February 2014 06:30, Anil Singhar anil.singhar@linaro.org wrote:
Hi Ryan,
Thanks for your reply.
RootFS getting corrupted by repeated <Ctrl+C>: Does that mean we have some known limitations / stability issues with the foundation model (or rootfs)?
No. If you CTRL-C the *model* - not a running process within the model - you are effectively pulling the plug on the machine. This is the equivalent of you booting up and powering off your computer without cleanly unmounting the disk, you will eventually corrupt the disk if you keep doing it.
This was not an issue for past 6 months and has only showed up recently after I used the custom image.
If you were killing the model every time, you got lucky.
So far I have generally launched the the model and used the model window itself to do build etc.. I remember having issues with doing SSH to the model from the host initially when I started using it. There was no steps mentioned in the user guide and I had tried various things without success. Few things I could do with the model were:
- wget works
- scp from host pc into the model works.
- pinging an ip address from inside the model doesn't work (e.g "ping
google.com" fails)
Linaro images have SSHD set up already. You have a networking setup problem if you don't have DNS resolving hostnames, but if you have access via IP address, you should be able to SSH into a model easily.
So, I don't know completely if and how to connect to the model via ssh so
Try this from your host PC:
ssh root@<ip address of model instance>Eg, ssh root@192.168.1.88
that I can avoid having to close a hung process by <Ctrl+c>. Could you please point me to some resource?
More advanced networking setup information is available from ARM's website as linked from the release notes:
http://releases.linaro.org/14.01/openembedded/aarch64/
I've never been happy with NAT networking, so I use a bridged setup.
Thanks, Anil
On 3 February 2014 22:03, Ryan Harkin ryan.harkin@linaro.org wrote:
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org wrote:
Hi,
I am running into problems with starting foundation model lately.
- I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
It was booting fine initially.
The problem started happening after I attempted to build Groff
package, which was not getting build and the model was getting frozen (no response to Ctrl + C, etc.). However I was able to restart the model by closing it from the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the middle, doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather than using the serial terminal. And if a command locks a session, I log in again via SSH and kill the old one, rather than pull the plug on the model.
Anyone knows whom should I report to about this?
This is a good place to discuss it.
Thanks, Anil
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
That still doesn't work for me:
anilss@anilss:~/Linaro$ ssh -p 8022 root@anilss ssh: connect to host anilss port 8022: Connection refused
Per ARM, the bridge support is NOT available on foundation model..though their user guide mentions it. I am tried the above using nat.
On 5 February 2014 17:41, Christophe Lyon christophe.lyon@linaro.orgwrote:
On 4 February 2014 10:02, Anil Singhar anil.singhar@linaro.org wrote:
- With the following command (networking option used nat):
./Foundation_v8 --image img-foundation.axf --block-device vexpress64-openembedded_lamp-armv8-gcc-4.8_20140126-596.img
--network=nat
--network-nat-ports=8022=22
I get the following on the model:
root@genericarmv8:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:02:F7:EF:20:B3 inet addr:172.20.51.1 Bcast:172.20.51.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:28 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:646 (646.0 B) TX bytes:4788 (4.6 KiB) Interrupt:47 Base address:0x2000 DMA chan:ff
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:784 (784.0 B) TX bytes:784 (784.0 B)
root@genericarmv8:~#
This doesn't help either. I can't ping anywhere to the outside world, not even to the host itself. And "ssh root@172.20.51.1" doesn't work from
the
host (outside the model). But surprisingly wget works. Here the value of "network-nat-ports", i.e. 8022=22 is provided in the document.
I don't know what is it in the instructions that I missing. I am not a networking expert at all, so it is actually too much work figuring out
all
these details just to be able to use the model.
Could you please let me know the set of commands you use to launch the
model
so that you can ping outside world and also do ssh to the model.
I don't need to go outside from within the model, but that's probably of network configuration of the linux distro installed in the model.
However, I am able to ssh to the model: ssh -p 8022 root@HOSTNAME (where HOSTNAME=name of the machine on which you start the model)
HTH
Christophe.
Thanks and regards, Anil
On 4 February 2014 12:27, Ryan Harkin ryan.harkin@linaro.org wrote:
On 4 February 2014 06:30, Anil Singhar anil.singhar@linaro.org wrote:
Hi Ryan,
Thanks for your reply.
RootFS getting corrupted by repeated <Ctrl+C>: Does that mean we have some known limitations / stability issues with the foundation model (or rootfs)?
No. If you CTRL-C the *model* - not a running process within the model - you are effectively pulling the plug on the machine. This is the equivalent of you booting up and powering off your computer without cleanly unmounting the disk, you will eventually corrupt the disk if you keep doing it.
This was not an issue for past 6 months and has only showed up
recently
after I used the custom image.
If you were killing the model every time, you got lucky.
So far I have generally launched the the model and used the model
window
itself to do build etc.. I remember having issues with doing SSH to
the
model from the host initially when I started using it. There was no steps mentioned in the user guide and I had tried various things without success. Few things I could do with the model were:
- wget works
- scp from host pc into the model works.
- pinging an ip address from inside the model doesn't work (e.g "ping
google.com" fails)
Linaro images have SSHD set up already. You have a networking setup problem if you don't have DNS resolving hostnames, but if you have access via IP address, you should be able to SSH into a model easily.
So, I don't know completely if and how to connect to the model via ssh so
Try this from your host PC:
ssh root@<ip address of model instance>Eg, ssh root@192.168.1.88
that I can avoid having to close a hung process by <Ctrl+c>. Could you please point me to some resource?
More advanced networking setup information is available from ARM's website as linked from the release notes:
http://releases.linaro.org/14.01/openembedded/aarch64/
I've never been happy with NAT networking, so I use a bridged setup.
Thanks, Anil
On 3 February 2014 22:03, Ryan Harkin ryan.harkin@linaro.org wrote:
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org
wrote:
Hi,
I am running into problems with starting foundation model lately.
- I had created an image using
linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz
It was booting fine initially.
The problem started happening after I attempted to build Groff
package, which was not getting build and the model was getting frozen (no response to Ctrl + C, etc.). However I was able to restart the model by closing it from the launch window and then restarting it.
Now, the model simply fails to launch successfully (hangs in the middle, doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather
than
using the serial terminal. And if a command locks a session, I log
in
again via SSH and kill the old one, rather than pull the plug on the model.
Anyone knows whom should I report to about this?
This is a good place to discuss it.
Thanks, Anil
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
On 5 February 2014 17:34, Anil Singhar anil.singhar@linaro.org wrote:
That still doesn't work for me:
anilss@anilss:~/Linaro$ ssh -p 8022 root@anilss ssh: connect to host anilss port 8022: Connection refused
Per ARM, the bridge support is NOT available on foundation model..
Please explain what you mean. Bridge support works on the Foundation model.
though their user guide mentions it. I am tried the above using nat.
On 5 February 2014 17:41, Christophe Lyon christophe.lyon@linaro.org wrote:
On 4 February 2014 10:02, Anil Singhar anil.singhar@linaro.org wrote:
- With the following command (networking option used nat):
./Foundation_v8 --image img-foundation.axf --block-device vexpress64-openembedded_lamp-armv8-gcc-4.8_20140126-596.img --network=nat --network-nat-ports=8022=22
I get the following on the model:
root@genericarmv8:~# ifconfig eth0 Link encap:Ethernet HWaddr 00:02:F7:EF:20:B3 inet addr:172.20.51.1 Bcast:172.20.51.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:28 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:646 (646.0 B) TX bytes:4788 (4.6 KiB) Interrupt:47 Base address:0x2000 DMA chan:ff
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:784 (784.0 B) TX bytes:784 (784.0 B)
root@genericarmv8:~#
This doesn't help either. I can't ping anywhere to the outside world, not even to the host itself. And "ssh root@172.20.51.1" doesn't work from the host (outside the model). But surprisingly wget works. Here the value of "network-nat-ports", i.e. 8022=22 is provided in the document.
I don't know what is it in the instructions that I missing. I am not a networking expert at all, so it is actually too much work figuring out all these details just to be able to use the model.
Could you please let me know the set of commands you use to launch the model so that you can ping outside world and also do ssh to the model.
I don't need to go outside from within the model, but that's probably of network configuration of the linux distro installed in the model.
However, I am able to ssh to the model: ssh -p 8022 root@HOSTNAME (where HOSTNAME=name of the machine on which you start the model)
HTH
Christophe.
Thanks and regards, Anil
On 4 February 2014 12:27, Ryan Harkin ryan.harkin@linaro.org wrote:
On 4 February 2014 06:30, Anil Singhar anil.singhar@linaro.org wrote:
Hi Ryan,
Thanks for your reply.
RootFS getting corrupted by repeated <Ctrl+C>: Does that mean we have some known limitations / stability issues with the foundation model (or rootfs)?
No. If you CTRL-C the *model* - not a running process within the model - you are effectively pulling the plug on the machine. This is the equivalent of you booting up and powering off your computer without cleanly unmounting the disk, you will eventually corrupt the disk if you keep doing it.
This was not an issue for past 6 months and has only showed up recently after I used the custom image.
If you were killing the model every time, you got lucky.
So far I have generally launched the the model and used the model window itself to do build etc.. I remember having issues with doing SSH to the model from the host initially when I started using it. There was no steps mentioned in the user guide and I had tried various things without success. Few things I could do with the model were:
- wget works
- scp from host pc into the model works.
- pinging an ip address from inside the model doesn't work (e.g
"ping google.com" fails)
Linaro images have SSHD set up already. You have a networking setup problem if you don't have DNS resolving hostnames, but if you have access via IP address, you should be able to SSH into a model easily.
So, I don't know completely if and how to connect to the model via ssh so
Try this from your host PC:
ssh root@<ip address of model instance>Eg, ssh root@192.168.1.88
that I can avoid having to close a hung process by <Ctrl+c>. Could you please point me to some resource?
More advanced networking setup information is available from ARM's website as linked from the release notes:
http://releases.linaro.org/14.01/openembedded/aarch64/
I've never been happy with NAT networking, so I use a bridged setup.
Thanks, Anil
On 3 February 2014 22:03, Ryan Harkin ryan.harkin@linaro.org wrote:
On 3 February 2014 13:38, Anil Singhar anil.singhar@linaro.org wrote: > Hi, > > I am running into problems with starting foundation model lately. > > 1. I had created an image using > > linaro-image-lamp-genericarmv8-20140127-635.rootfs.tar.gz > > 2. It was booting fine initially. > > 3. The problem started happening after I attempted to build Groff > package, > which was not getting build and the model was getting frozen (no > response to > Ctrl + C, etc.). However I was able to restart the model by > closing > it > from > the launch window and then restarting it. > > Now, the model simply fails to launch successfully (hangs in the > middle, > doesn't get to the command prompt).
I suspect you've corrupted your rootfs by killing the model repeatedly. Have you tried starting again with a "clean" rootfs, built with linaro-media-create? If not, you should.
Generally, I tend to SSH into my machines to run commands, rather than using the serial terminal. And if a command locks a session, I log in again via SSH and kill the old one, rather than pull the plug on the model.
> > Anyone knows whom should I report to about this?
This is a good place to discuss it.
> > Thanks, > Anil > > _______________________________________________ > linaro-dev mailing list > linaro-dev@lists.linaro.org > http://lists.linaro.org/mailman/listinfo/linaro-dev >
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
-
Anil Singhar -
Christophe Lyon -
Jon Medhurst (Tixy) -
Ryan Harkin