* [PATCH] devshell: Use different termination test to avoid warnings
@ 2020-05-16 8:00 Jan Kiszka
2020-05-25 16:11 ` Baurzhan Ismagulov
0 siblings, 1 reply; 3+ messages in thread
From: Jan Kiszka @ 2020-05-16 8:00 UTC (permalink / raw)
To: isar-users
From: Jan Kiszka <jan.kiszka@siemens.com>
As we run the terminal via a sudo chroot, OE's way of testing for the
existence of the target PID does not work. It will rather fail early due
to missing permissions (user tries to send signal to root process), and
we will wait in the unmount loop, generating warnings like
WARNING: mc:qemuamd64-buster:example-raw-0.3-r0 do_devshell: /work/build/tmp/deploy/buildchroot-target/debian-buster-amd64//home/builder/example-raw: Couldn't unmount, retrying...
This replaces the kill(pid, 0) with a test for the existence of the
corresponding /proc/<pid> directory.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
meta/lib/oe/terminal.py | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/meta/lib/oe/terminal.py b/meta/lib/oe/terminal.py
index 87058040..a9fb62b9 100644
--- a/meta/lib/oe/terminal.py
+++ b/meta/lib/oe/terminal.py
@@ -245,9 +245,11 @@ def spawn(name, sh_cmd, title=None, env=None, d=None):
finally:
os.unlink(pidfile)
+ # Test for /proc entry so that target pid can also be privileged
+ procpid = "/proc/%d" % pid
while True:
try:
- os.kill(pid, 0)
+ os.stat(procpid)
time.sleep(0.1)
except OSError:
return
--
2.26.1
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] devshell: Use different termination test to avoid warnings
2020-05-16 8:00 [PATCH] devshell: Use different termination test to avoid warnings Jan Kiszka
@ 2020-05-25 16:11 ` Baurzhan Ismagulov
2020-05-25 16:13 ` Jan Kiszka
0 siblings, 1 reply; 3+ messages in thread
From: Baurzhan Ismagulov @ 2020-05-25 16:11 UTC (permalink / raw)
To: isar-users
On Sat, May 16, 2020 at 10:00:37AM +0200, Jan Kiszka wrote:
> As we run the terminal via a sudo chroot, OE's way of testing for the
> existence of the target PID does not work. It will rather fail early due
> to missing permissions (user tries to send signal to root process), and
> we will wait in the unmount loop, generating warnings like
>
> WARNING: mc:qemuamd64-buster:example-raw-0.3-r0 do_devshell: /work/build/tmp/deploy/buildchroot-target/debian-buster-amd64//home/builder/example-raw: Couldn't unmount, retrying...
>
> This replaces the kill(pid, 0) with a test for the existence of the
> corresponding /proc/<pid> directory.
Applied to next, thanks.
Is there anything against submitting this upstream, too?
With kind regards,
Baurzhan.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] devshell: Use different termination test to avoid warnings
2020-05-25 16:11 ` Baurzhan Ismagulov
@ 2020-05-25 16:13 ` Jan Kiszka
0 siblings, 0 replies; 3+ messages in thread
From: Jan Kiszka @ 2020-05-25 16:13 UTC (permalink / raw)
To: isar-users
On 25.05.20 18:11, Baurzhan Ismagulov wrote:
> On Sat, May 16, 2020 at 10:00:37AM +0200, Jan Kiszka wrote:
>> As we run the terminal via a sudo chroot, OE's way of testing for the
>> existence of the target PID does not work. It will rather fail early due
>> to missing permissions (user tries to send signal to root process), and
>> we will wait in the unmount loop, generating warnings like
>>
>> WARNING: mc:qemuamd64-buster:example-raw-0.3-r0 do_devshell: /work/build/tmp/deploy/buildchroot-target/debian-buster-amd64//home/builder/example-raw: Couldn't unmount, retrying...
>>
>> This replaces the kill(pid, 0) with a test for the existence of the
>> corresponding /proc/<pid> directory.
>
> Applied to next, thanks.
>
> Is there anything against submitting this upstream, too?
>
We need a reason, and that is - to my understanding - not in sight for
upstream OE.
Jan
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-05-25 16:13 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-16 8:00 [PATCH] devshell: Use different termination test to avoid warnings Jan Kiszka
2020-05-25 16:11 ` Baurzhan Ismagulov
2020-05-25 16:13 ` Jan Kiszka
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox