After figuring out proxies and setting them correctly, I let run First QEMU job, described here: https://staging.validation.linaro.org/static/docs/v2/first-job.html
And have the following errors: https://pastebin.com/334xHvX8
For example, there are no elements requested from this job such as: E: Failed to fetch http://mirror.bytemark.co.uk/debian/pool/main/p/python3.5/python3.5-minimal_... Unable to connect to mirror.bytemark.co.uk:http: [IP: 2001:41c8:20:5e6::10 80] E: Failed to fetch http://mirror.bytemark.co.uk/debian/pool/main/p/python3-defaults/python3-min... Unable to connect to mirror.bytemark.co.uk:http: [IP: 2001:41c8:20:5e6::10 80]
But I am getting on ping mirror.bytemark.co.uk PING mirror.bytemark.co.uk (212.110.161.69)... ???
IP address seems to be IPV6 (which points that something is not correctly set)?!
I also checked a routing table, everything there within VM is set correctly. Do I need to set more of the proxies?!
What is the problem here???
Thank you, Zoran
Hi Zoran,
On Wed, Feb 14, 2018 at 03:45:12PM +0100, Zoran S wrote:
After figuring out proxies and setting them correctly, I let run First QEMU job, described here: https://staging.validation.linaro.org/static/docs/v2/first-job.html
And have the following errors: https://pastebin.com/334xHvX8
For example, there are no elements requested from this job such as: E: Failed to fetch http://mirror.bytemark.co.uk/debian/pool/main/p/python3.5/python3.5-minimal_... Unable to connect to mirror.bytemark.co.uk:http: [IP: 2001:41c8:20:5e6::10 80] E: Failed to fetch http://mirror.bytemark.co.uk/debian/pool/main/p/python3-defaults/python3-min... Unable to connect to mirror.bytemark.co.uk:http: [IP: 2001:41c8:20:5e6::10 80]
The image you're testing with is configured to use the Bytemark mirror as a source for Debian packages. It's trying to use that mirror to install packages listed in lava-test-shell/single-node/singlenode03.yaml - see the install: step.
But I am getting on ping mirror.bytemark.co.uk PING mirror.bytemark.co.uk (212.110.161.69)... ???
IP address seems to be IPV6 (which points that something is not correctly set)?!
2001:41c8:20:5e6::10 is mirror.bytemark.co.uk, so that seems correct. Do you expect your test VM to use v6?
Cheers,
Hello Steve,
I am not reaching mirror.bytemark.co.uk : Unable to connect to mirror.bytemark.co.uk:http: [IP:>2001:41c8:20:5e6::10 80]
No, my VM is set to use IPV4. And my VM does reach all other external sites, except mirror.bytemark.co.uk. You can see this from the pastebin log: https://pastebin.com/334xHvX8
Do I have authorization issue here? Do I need to have some more authorization to reach this server?
Zoran _______
On Wed, Feb 14, 2018 at 4:20 PM, Steve McIntyre steve.mcintyre@linaro.org wrote:
Hi Zoran,
On Wed, Feb 14, 2018 at 03:45:12PM +0100, Zoran S wrote:
After figuring out proxies and setting them correctly, I let run First QEMU job, described here: https://staging.validation.linaro.org/static/docs/v2/first-job.html
And have the following errors: https://pastebin.com/334xHvX8
For example, there are no elements requested from this job such as: E: Failed to fetch http://mirror.bytemark.co.uk/debian/pool/main/p/python3.5/python3.5-minimal_... Unable to connect to mirror.bytemark.co.uk:http: [IP: 2001:41c8:20:5e6::10 80] E: Failed to fetch http://mirror.bytemark.co.uk/debian/pool/main/p/python3-defaults/python3-min... Unable to connect to mirror.bytemark.co.uk:http: [IP: 2001:41c8:20:5e6::10 80]
The image you're testing with is configured to use the Bytemark mirror as a source for Debian packages. It's trying to use that mirror to install packages listed in lava-test-shell/single-node/singlenode03.yaml - see the install: step.
But I am getting on ping mirror.bytemark.co.uk PING mirror.bytemark.co.uk (212.110.161.69)... ???
IP address seems to be IPV6 (which points that something is not correctly set)?!
2001:41c8:20:5e6::10 is mirror.bytemark.co.uk, so that seems correct. Do you expect your test VM to use v6?
Cheers,
Steve McIntyre steve.mcintyre@linaro.org http://www.linaro.org/ Linaro.org | Open source software for ARM SoCs
On Wed, Feb 14, 2018 at 04:42:02PM +0100, Zoran S wrote:
Hello Steve,
I am not reaching mirror.bytemark.co.uk : Unable to connect to mirror.bytemark.co.uk:http: [IP:>2001:41c8:20:5e6::10 80]
No, my VM is set to use IPV4. And my VM does reach all other external sites, except mirror.bytemark.co.uk. You can see this from the pastebin log: https://pastebin.com/334xHvX8
Do I have authorization issue here? Do I need to have some more authorization to reach this server?
No, that's not your problem.
apt-get inside the VM is looking up "mirror.bytemark.co.uk" in DNS and getting back both the IPv4 and IPv6 addresses. It's preferring v6 (as most things will these days) and trying to connect to 2001:41c8:20:5e6::10. That's failing. It's then reporting that error. This is not normally a problem - apt will fall back to v4 if v6 doesn't work. But I'm guessing that also didn't work, so it's reporting the first error.
I've seen in your previous mails to the list that you've needed to configure a proxy for the dispatcher (so that works for downloading the image in the first place). Unfortunately, the proxy settings are not passed on into the network setup inside the VM and that's probably the problem you're seeing. It will be trying to connect directly to the remote web server. If your network doesn't allow that at all, then you'll either need to reconfigure the test image to use your proxy, or set up a transparent proxy to control things as needed.
I've specifically tried to reproduce this problem on my local network using the same image as you (the standard one from images.v.l.o) and I can't. I've tried various configurations on the host machine and VM, (no network, v4 only, v6 only, both v4 and v6) and I don't get the same error messages. Thst's what tells me it must be your network with the proxy setup, I'm afraid.
Hope that helps!
Cheers,
Hello Steve,
Thank you for the reply. I think I know what/why the mess is. NOT the proxies in Lava Server, they are set correctly.
The following exactly describes the situation: I have three separate entities, working altogether: [1] Host; [2] VM with Lava; [3] QEMU, which is EXTERNAL entity to Lava (inside the Lava, supposed to be/emulate HW target connected to the host)!
How I can login to QEMU from Lava/actually from VM? This is a path to the solution of this problem.
Zoran _______
On Wed, Feb 14, 2018 at 5:27 PM, Steve McIntyre steve.mcintyre@linaro.org wrote:
On Wed, Feb 14, 2018 at 04:42:02PM +0100, Zoran S wrote:
Hello Steve,
I am not reaching mirror.bytemark.co.uk : Unable to connect to mirror.bytemark.co.uk:http: [IP:>2001:41c8:20:5e6::10 80]
No, my VM is set to use IPV4. And my VM does reach all other external sites, except mirror.bytemark.co.uk. You can see this from the pastebin log: https://pastebin.com/334xHvX8
Do I have authorization issue here? Do I need to have some more authorization to reach this server?
No, that's not your problem.
apt-get inside the VM is looking up "mirror.bytemark.co.uk" in DNS and getting back both the IPv4 and IPv6 addresses. It's preferring v6 (as most things will these days) and trying to connect to 2001:41c8:20:5e6::10. That's failing. It's then reporting that error. This is not normally a problem - apt will fall back to v4 if v6 doesn't work. But I'm guessing that also didn't work, so it's reporting the first error.
I've seen in your previous mails to the list that you've needed to configure a proxy for the dispatcher (so that works for downloading the image in the first place). Unfortunately, the proxy settings are not passed on into the network setup inside the VM and that's probably the problem you're seeing. It will be trying to connect directly to the remote web server. If your network doesn't allow that at all, then you'll either need to reconfigure the test image to use your proxy, or set up a transparent proxy to control things as needed.
I've specifically tried to reproduce this problem on my local network using the same image as you (the standard one from images.v.l.o) and I can't. I've tried various configurations on the host machine and VM, (no network, v4 only, v6 only, both v4 and v6) and I don't get the same error messages. Thst's what tells me it must be your network with the proxy setup, I'm afraid.
Hope that helps!
Cheers,
Steve McIntyre steve.mcintyre@linaro.org http://www.linaro.org/ Linaro.org | Open source software for ARM SoCs
Yup... All Clear! I was able to run qemu image with qemu-system-x86_64 exe, and there I found the (networking/proxies) problem.
QEMU itself does not have proxies setup. ;-)
4 levels of redirections: from [1] corporate firewall; [2] host notebook connected to it, running VM; [3] VM running Lava inside; [4] Lava inside running qemu; [5] qemu itself with proper routing table set (but sans proxies)! :-))
So, the obvious solution is to set up the local mirrors (from http://mirror.bytemark.co.uk/debian/) in host and/or VM.
Best Regards, Zoran _______
On Thu, Feb 15, 2018 at 9:51 AM, Zoran S zoran.stojsavljevic.de@gmail.com wrote:
Hello Steve,
Thank you for the reply. I think I know what/why the mess is. NOT the proxies in Lava Server, they are set correctly.
The following exactly describes the situation: I have three separate entities, working altogether: [1] Host; [2] VM with Lava; [3] QEMU, which is EXTERNAL entity to Lava (inside the Lava, supposed to be/emulate HW target connected to the host)!
How I can login to QEMU from Lava/actually from VM? This is a path to the solution of this problem.
Zoran _______
On Wed, Feb 14, 2018 at 5:27 PM, Steve McIntyre steve.mcintyre@linaro.org wrote:
On Wed, Feb 14, 2018 at 04:42:02PM +0100, Zoran S wrote:
Hello Steve,
I am not reaching mirror.bytemark.co.uk : Unable to connect to mirror.bytemark.co.uk:http: [IP:>2001:41c8:20:5e6::10 80]
No, my VM is set to use IPV4. And my VM does reach all other external sites, except mirror.bytemark.co.uk. You can see this from the pastebin log: https://pastebin.com/334xHvX8
Do I have authorization issue here? Do I need to have some more authorization to reach this server?
No, that's not your problem.
apt-get inside the VM is looking up "mirror.bytemark.co.uk" in DNS and getting back both the IPv4 and IPv6 addresses. It's preferring v6 (as most things will these days) and trying to connect to 2001:41c8:20:5e6::10. That's failing. It's then reporting that error. This is not normally a problem - apt will fall back to v4 if v6 doesn't work. But I'm guessing that also didn't work, so it's reporting the first error.
I've seen in your previous mails to the list that you've needed to configure a proxy for the dispatcher (so that works for downloading the image in the first place). Unfortunately, the proxy settings are not passed on into the network setup inside the VM and that's probably the problem you're seeing. It will be trying to connect directly to the remote web server. If your network doesn't allow that at all, then you'll either need to reconfigure the test image to use your proxy, or set up a transparent proxy to control things as needed.
I've specifically tried to reproduce this problem on my local network using the same image as you (the standard one from images.v.l.o) and I can't. I've tried various configurations on the host machine and VM, (no network, v4 only, v6 only, both v4 and v6) and I don't get the same error messages. Thst's what tells me it must be your network with the proxy setup, I'm afraid.
Hope that helps!
Cheers,
Steve McIntyre steve.mcintyre@linaro.org http://www.linaro.org/ Linaro.org | Open source software for ARM SoCs
Hello,
you cans et environment variable for the DUT (Device Under Test) by editing /etc/lava-server/env.dut.yaml (exact same syntax as env.yaml)
2018-02-15 13:11 GMT+01:00 Zoran S zoran.stojsavljevic.de@gmail.com:
Yup... All Clear! I was able to run qemu image with qemu-system-x86_64 exe, and there I found the (networking/proxies) problem.
QEMU itself does not have proxies setup. ;-)
4 levels of redirections: from [1] corporate firewall; [2] host notebook connected to it, running VM; [3] VM running Lava inside; [4] Lava inside running qemu; [5] qemu itself with proper routing table set (but sans proxies)! :-))
So, the obvious solution is to set up the local mirrors (from http://mirror.bytemark.co.uk/debian/) in host and/or VM.
Best Regards, Zoran _______
On Thu, Feb 15, 2018 at 9:51 AM, Zoran S zoran.stojsavljevic.de@gmail.com wrote:
Hello Steve,
Thank you for the reply. I think I know what/why the mess is. NOT the proxies in Lava Server, they are set correctly.
The following exactly describes the situation: I have three separate entities, working altogether: [1] Host; [2] VM with Lava; [3] QEMU, which is EXTERNAL entity to Lava (inside the Lava, supposed to be/emulate HW target connected to the host)!
How I can login to QEMU from Lava/actually from VM? This is a path to the solution of this problem.
Zoran _______
On Wed, Feb 14, 2018 at 5:27 PM, Steve McIntyre steve.mcintyre@linaro.org wrote:
On Wed, Feb 14, 2018 at 04:42:02PM +0100, Zoran S wrote:
Hello Steve,
I am not reaching mirror.bytemark.co.uk : Unable to connect to mirror.bytemark.co.uk:http:
[IP:>2001:41c8:20:5e6::10 80]
No, my VM is set to use IPV4. And my VM does reach all other external sites, except mirror.bytemark.co.uk. You can see this from the pastebin log: https://pastebin.com/334xHvX8
Do I have authorization issue here? Do I need to have some more authorization to reach this server?
No, that's not your problem.
apt-get inside the VM is looking up "mirror.bytemark.co.uk" in DNS and getting back both the IPv4 and IPv6 addresses. It's preferring v6 (as most things will these days) and trying to connect to 2001:41c8:20:5e6::10. That's failing. It's then reporting that error. This is not normally a problem - apt will fall back to v4 if v6 doesn't work. But I'm guessing that also didn't work, so it's reporting the first error.
I've seen in your previous mails to the list that you've needed to configure a proxy for the dispatcher (so that works for downloading the image in the first place). Unfortunately, the proxy settings are not passed on into the network setup inside the VM and that's probably the problem you're seeing. It will be trying to connect directly to the remote web server. If your network doesn't allow that at all, then you'll either need to reconfigure the test image to use your proxy, or set up a transparent proxy to control things as needed.
I've specifically tried to reproduce this problem on my local network using the same image as you (the standard one from images.v.l.o) and I can't. I've tried various configurations on the host machine and VM, (no network, v4 only, v6 only, both v4 and v6) and I don't get the same error messages. Thst's what tells me it must be your network with the proxy setup, I'm afraid.
Hope that helps!
Cheers,
Steve McIntyre steve.mcintyre@linaro.org http://www.linaro.org/ Linaro.org | Open source software for ARM SoCs
Lava-users mailing list Lava-users@lists.linaro.org https://lists.linaro.org/mailman/listinfo/lava-users
lava-users@lists.lavasoftware.org
-
Remi Duraffort -
Steve McIntyre -
Zoran S