Message ID | 20180823112153.15279-1-peter.maydell@linaro.org |
---|---|
State | New |
Headers | show |
Series | tests/vm: Increase timeout waiting for VM to boot to 5 minutes | expand |
On Thu, 08/23 12:21, Peter Maydell wrote: > The VM tests currently have a timeout of 2 minutes for trying > to connect to ssh. Since the guest VM has to boot from cold > to the point of accepting inbound ssh during this time, if the > host machine is heavily loaded it can spuriously time out. > Increase the timeout from 2 to 5 minutes. > > Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Fam Zheng <famz@redhat.com> Again, please apply this yourself. :) Thanks, Fam
On 23 August 2018 at 15:11, Fam Zheng <famz@redhat.com> wrote: > On Thu, 08/23 12:21, Peter Maydell wrote: >> The VM tests currently have a timeout of 2 minutes for trying >> to connect to ssh. Since the guest VM has to boot from cold >> to the point of accepting inbound ssh during this time, if the >> host machine is heavily loaded it can spuriously time out. >> Increase the timeout from 2 to 5 minutes. >> >> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> > > Acked-by: Fam Zheng <famz@redhat.com> > > Again, please apply this yourself. :) Applied, thanks. -- PMM
On 24 August 2018 at 13:02, Peter Maydell <peter.maydell@linaro.org> wrote: > On 23 August 2018 at 15:11, Fam Zheng <famz@redhat.com> wrote: >> On Thu, 08/23 12:21, Peter Maydell wrote: >>> The VM tests currently have a timeout of 2 minutes for trying >>> to connect to ssh. Since the guest VM has to boot from cold >>> to the point of accepting inbound ssh during this time, if the >>> host machine is heavily loaded it can spuriously time out. >>> Increase the timeout from 2 to 5 minutes. >>> >>> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> >> >> Acked-by: Fam Zheng <famz@redhat.com> >> >> Again, please apply this yourself. :) > > Applied, thanks. I'm still seeing the ssh login for a netbsd VM sometimes sit there apparently indefinitely. Is there some way to debug what the guest VM is actually doing when the log just sits there doing DEBUG:root:ssh_cmd: ssh -q -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -o ConnectTimeout=1 -p 36947 -i /home/peter.maydell/qemu-netbsd/build/vm-test-g0SkVX.tmp/id_rsa qemu@127.0.0.1 exit 0 DEBUG:root:ssh_cmd: ssh -q -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -o ConnectTimeout=1 -p 36947 -i /home/peter.maydell/qemu-netbsd/build/vm-test-g0SkVX.tmp/id_rsa qemu@127.0.0.1 exit 0 (eg a capture of its terminal output) ? thanks -- PMM
On Fri, 08/24 13:45, Peter Maydell wrote: > On 24 August 2018 at 13:02, Peter Maydell <peter.maydell@linaro.org> wrote: > > On 23 August 2018 at 15:11, Fam Zheng <famz@redhat.com> wrote: > >> On Thu, 08/23 12:21, Peter Maydell wrote: > >>> The VM tests currently have a timeout of 2 minutes for trying > >>> to connect to ssh. Since the guest VM has to boot from cold > >>> to the point of accepting inbound ssh during this time, if the > >>> host machine is heavily loaded it can spuriously time out. > >>> Increase the timeout from 2 to 5 minutes. > >>> > >>> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> > >> > >> Acked-by: Fam Zheng <famz@redhat.com> > >> > >> Again, please apply this yourself. :) > > > > Applied, thanks. > > I'm still seeing the ssh login for a netbsd VM sometimes > sit there apparently indefinitely. Is there some way to > debug what the guest VM is actually doing when the > log just sits there doing > DEBUG:root:ssh_cmd: ssh -q -o StrictHostKeyChecking=no -o > UserKnownHostsFile=/dev/null -o ConnectTimeout=1 -p 36947 -i > /home/peter.maydell/qemu-netbsd/build/vm-test-g0SkVX.tmp/id_rsa > qemu@127.0.0.1 exit 0 > DEBUG:root:ssh_cmd: ssh -q -o StrictHostKeyChecking=no -o > UserKnownHostsFile=/dev/null -o ConnectTimeout=1 -p 36947 -i > /home/peter.maydell/qemu-netbsd/build/vm-test-g0SkVX.tmp/id_rsa > qemu@127.0.0.1 exit 0 > > (eg a capture of its terminal output) ? I need to investigate into it. Sometimes I see the timeout too with V=1, but not if I omit it (i.e. on DEBUG: lines are echoed). Fam
diff --git a/tests/vm/basevm.py b/tests/vm/basevm.py index d7149dea7da..7e58d9e0ca2 100755 --- a/tests/vm/basevm.py +++ b/tests/vm/basevm.py @@ -176,7 +176,7 @@ class BaseVM(object): raise Exception("Cannot find ssh port from 'info usernet':\n%s" % \ usernet_info) - def wait_ssh(self, seconds=120): + def wait_ssh(self, seconds=300): starttime = datetime.datetime.now() guest_up = False while (datetime.datetime.now() - starttime).total_seconds() < seconds:
The VM tests currently have a timeout of 2 minutes for trying to connect to ssh. Since the guest VM has to boot from cold to the point of accepting inbound ssh during this time, if the host machine is heavily loaded it can spuriously time out. Increase the timeout from 2 to 5 minutes. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- I see this ssh-connection-attempt timeout occasionally in my test builds... tests/vm/basevm.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)