Change in bitbake behaviour?

Change in bitbake behaviour?

[Gary Thomas]

I just updated to the latest Poky/Yocto master
and I noticed this change in behaviour when I tried to stop
a build:
   Keyboard Interrupt, closing down...

   NOTE: Tasks Summary: Attempted 86 tasks of which 36 didn't need to be rerun and all succeeded.
   ERROR: Command execution failed: Stopped build

   Summary: There was 1 ERROR message shown, returning a non-zero exit code.
   Execution was interrupted, returning a non-zero exit code.
   timed out

It took more than a minute between the last two lines at which time
I thought it might be hung.  I left it and then found the "timed out"
line somewhat later.

Is this to be expected?

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------

Re: Change in bitbake behaviour?

[Paul Eggleton]

Hi Gary,

On Monday 21 September 2015 06:35:54 Gary Thomas wrote:
> I just updated to the latest Poky/Yocto master
> and I noticed this change in behaviour when I tried to stop
> a build:
>    Keyboard Interrupt, closing down...
> 
>    NOTE: Tasks Summary: Attempted 86 tasks of which 36 didn't need to be
> rerun and all succeeded. ERROR: Command execution failed: Stopped build
> 
>    Summary: There was 1 ERROR message shown, returning a non-zero exit code.
> Execution was interrupted, returning a non-zero exit code.
>    timed out
> 
> It took more than a minute between the last two lines at which time
> I thought it might be hung.  I left it and then found the "timed out"
> line somewhat later.
> 
> Is this to be expected?

No, this sounds like a bug - but was this somehow related to your other build 
issue or do you still see this?

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre

Re: Change in bitbake behaviour?

[Gary Thomas]

On 2015-09-23 07:46, Paul Eggleton wrote:
> Hi Gary,
>
> On Monday 21 September 2015 06:35:54 Gary Thomas wrote:
>> I just updated to the latest Poky/Yocto master
>> and I noticed this change in behaviour when I tried to stop
>> a build:
>>     Keyboard Interrupt, closing down...
>>
>>     NOTE: Tasks Summary: Attempted 86 tasks of which 36 didn't need to be
>> rerun and all succeeded. ERROR: Command execution failed: Stopped build
>>
>>     Summary: There was 1 ERROR message shown, returning a non-zero exit code.
>> Execution was interrupted, returning a non-zero exit code.
>>     timed out
>>
>> It took more than a minute between the last two lines at which time
>> I thought it might be hung.  I left it and then found the "timed out"
>> line somewhat later.
>>
>> Is this to be expected?
>
> No, this sounds like a bug - but was this somehow related to your other build
> issue or do you still see this?

No, I'm still seeing it, but only occasionally.  It's most
pronounced if I try and abort a build, but I have seen it
also with a successful run (it hangs when seemingly complete
the prints 'timed out' after a fairly long time)

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------

Re: Change in bitbake behaviour?

[Paul Eggleton]

On Wednesday 23 September 2015 07:49:34 Gary Thomas wrote:
> On 2015-09-23 07:46, Paul Eggleton wrote:
> > Hi Gary,
> > 
> > On Monday 21 September 2015 06:35:54 Gary Thomas wrote:
> >> I just updated to the latest Poky/Yocto master
> >> and I noticed this change in behaviour when I tried to stop
> >> 
> >> a build:
> >>     Keyboard Interrupt, closing down...
> >>     
> >>     NOTE: Tasks Summary: Attempted 86 tasks of which 36 didn't need to be
> >> 
> >> rerun and all succeeded. ERROR: Command execution failed: Stopped build
> >> 
> >>     Summary: There was 1 ERROR message shown, returning a non-zero exit
> >>     code.
> >> 
> >> Execution was interrupted, returning a non-zero exit code.
> >> 
> >>     timed out
> >> 
> >> It took more than a minute between the last two lines at which time
> >> I thought it might be hung.  I left it and then found the "timed out"
> >> line somewhat later.
> >> 
> >> Is this to be expected?
> > 
> > No, this sounds like a bug - but was this somehow related to your other
> > build issue or do you still see this?
> 
> No, I'm still seeing it, but only occasionally.  It's most
> pronounced if I try and abort a build, but I have seen it
> also with a successful run (it hangs when seemingly complete
> the prints 'timed out' after a fairly long time)

OK... is a PR server enabled in your configuration and if so is it being 
started automatically or running on a different machine?

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre

Re: Change in bitbake behaviour?

[Gary Thomas]

On 2015-09-23 07:52, Paul Eggleton wrote:
> On Wednesday 23 September 2015 07:49:34 Gary Thomas wrote:
>> On 2015-09-23 07:46, Paul Eggleton wrote:
>>> Hi Gary,
>>>
>>> On Monday 21 September 2015 06:35:54 Gary Thomas wrote:
>>>> I just updated to the latest Poky/Yocto master
>>>> and I noticed this change in behaviour when I tried to stop
>>>>
>>>> a build:
>>>>      Keyboard Interrupt, closing down...
>>>>
>>>>      NOTE: Tasks Summary: Attempted 86 tasks of which 36 didn't need to be
>>>>
>>>> rerun and all succeeded. ERROR: Command execution failed: Stopped build
>>>>
>>>>      Summary: There was 1 ERROR message shown, returning a non-zero exit
>>>>      code.
>>>>
>>>> Execution was interrupted, returning a non-zero exit code.
>>>>
>>>>      timed out
>>>>
>>>> It took more than a minute between the last two lines at which time
>>>> I thought it might be hung.  I left it and then found the "timed out"
>>>> line somewhat later.
>>>>
>>>> Is this to be expected?
>>>
>>> No, this sounds like a bug - but was this somehow related to your other
>>> build issue or do you still see this?
>>
>> No, I'm still seeing it, but only occasionally.  It's most
>> pronounced if I try and abort a build, but I have seen it
>> also with a successful run (it hangs when seemingly complete
>> the prints 'timed out' after a fairly long time)
>
> OK... is a PR server enabled in your configuration and if so is it being
> started automatically or running on a different machine?

Started automatically on this machine.  I routinely have this
in my local.conf:
   # Enable local PR server - edit this to use a shared server
   PRSERV_HOST = "localhost:0"

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------