Skip to content
This repository has been archived by the owner on Feb 23, 2021. It is now read-only.

On certain systems, systemd fails to use ttyS0 as Standard Output #130

Open
ricardon opened this issue Feb 16, 2018 · 0 comments
Open

On certain systems, systemd fails to use ttyS0 as Standard Output #130

ricardon opened this issue Feb 16, 2018 · 0 comments
Assignees
Labels

Comments

@ricardon
Copy link
Contributor

ricardon commented Feb 16, 2018

This is serious because the system appears to hang. The splash screen only shows the status of telemetrics.

After debugging, systemctl status luv-test-manager.service shows

luv-test-manager.service: control process exited, code=exited status=209

According to [1], this means that systemd was unable to setup the file descriptor for stdout.

In a recent patch [2], the StandardOuput is set to tty with TTYPath as ttyS0 for x86. Oddly, this system does have a /dev/ttyS0.

Replacing the StandardOutput with journal+console causes the LUV test manager to print its output on the virtual terminals (tty0) but not the serial console at boot. However, if the LUV test manager is started from the login screen, the output goes to the serial console as expected.

I verified that the the file /sys/classes/tty/console/active has ttyS0 in it.

Thanks and BR,
Ricardo

[1] https://bugzilla.redhat.com/show_bug.cgi?id=737292
[2] b12bf16

@ricardon ricardon added the bug label Feb 16, 2018
@meghadey meghadey self-assigned this May 23, 2018
meghadey pushed a commit that referenced this issue Feb 20, 2019
Commit b12bf16 "luv-test: update standard output for luv" updated
the default standard output of LUV to ttyS0 for x86 and ttyAMA0
for ARM architectures. However, after this change it was observed that
on some x86 systems, systemd fails to use ttyS0 as Standard Output and
the system appears to hang.

By reverting the above commit, we no longer see this issue, but we do
not see test-suite logs for ARM architecture on the serial console.
Hence, selectively revert the above patch for x86 architecture.

This patch resolves the issue:
#130

Signed-off-by: Megha Dey <megha.dey@linux.intel.com>
meghadey pushed a commit that referenced this issue Jun 25, 2019
Commit b12bf16 "luv-test: update standard output for luv" updated
the default standard output of LUV to ttyS0 for x86 and ttyAMA0
for ARM architectures. However, after this change it was observed that
on some x86 systems, systemd fails to use ttyS0 as Standard Output and
the system appears to hang.

By reverting the above commit, we no longer see this issue, but we do
not see test-suite logs for ARM architecture on the serial console.
Hence, selectively revert the above patch for x86 architecture.

This patch resolves the issue:
#130

Signed-off-by: Megha Dey <megha.dey@linux.intel.com>
meghadey pushed a commit that referenced this issue Jun 25, 2019
Commit b12bf16 "luv-test: update standard output for luv" updated
the default standard output of LUV to ttyS0 for x86 and ttyAMA0
for ARM architectures. However, after this change it was observed that
on some x86 systems, systemd fails to use ttyS0 as Standard Output and
the system appears to hang.

By reverting the above commit, we no longer see this issue, but we do
not see test-suite logs for ARM architecture on the serial console.
Hence, selectively revert the above patch for x86 architecture.

This patch resolves the issue:
#130

Signed-off-by: Megha Dey <megha.dey@linux.intel.com>
meghadey pushed a commit that referenced this issue Jun 25, 2019
Commit b12bf16 "luv-test: update standard output for luv" updated
the default standard output of LUV to ttyS0 for x86 and ttyAMA0
for ARM architectures. However, after this change it was observed that
on some x86 systems, systemd fails to use ttyS0 as Standard Output and
the system appears to hang.

By reverting the above commit, we no longer see this issue, but we do
not see test-suite logs for ARM architecture on the serial console.
Hence, selectively revert the above patch for x86 architecture.

This patch resolves the issue:
#130

Reviewed-By/Tested-By: Naresh Bhat <naresh.bhat at linaro.org>
Signed-off-by: Megha Dey <megha.dey@linux.intel.com>
meghadey pushed a commit that referenced this issue Jun 25, 2019
Commit b12bf16 "luv-test: update standard output for luv" updated
the default standard output of LUV to ttyS0 for x86 and ttyAMA0
for ARM architectures. However, after this change it was observed that
on some x86 systems, systemd fails to use ttyS0 as Standard Output and
the system appears to hang.

By reverting the above commit, we no longer see this issue, but we do
not see test-suite logs for ARM architecture on the serial console.
Hence, selectively revert the above patch for x86 architecture.

This patch resolves the issue:
#130

Reviewed-By/Tested-By: Naresh Bhat <naresh.bhat at linaro.org>
Signed-off-by: Megha Dey <megha.dey@linux.intel.com>
meghadey pushed a commit that referenced this issue Jun 25, 2019
Commit b12bf16 "luv-test: update standard output for luv" updated
the default standard output of LUV to ttyS0 for x86 and ttyAMA0
for ARM architectures. However, after this change it was observed that
on some x86 systems, systemd fails to use ttyS0 as Standard Output and
the system appears to hang.

By reverting the above commit, we no longer see this issue, but we do
not see test-suite logs for ARM architecture on the serial console.
Hence, selectively revert the above patch for x86 architecture.

This patch resolves the issue:
#130

Reviewed-By/Tested-By: Naresh Bhat <naresh.bhat at linaro.org>
Signed-off-by: Megha Dey <megha.dey@linux.intel.com>
meghadey pushed a commit that referenced this issue Jun 25, 2019
Commit b12bf16 "luv-test: update standard output for luv" updated
the default standard output of LUV to ttyS0 for x86 and ttyAMA0
for ARM architectures. However, after this change it was observed that
on some x86 systems, systemd fails to use ttyS0 as Standard Output and
the system appears to hang.

By reverting the above commit, we no longer see this issue, but we do
not see test-suite logs for ARM architecture on the serial console.
Hence, selectively revert the above patch for x86 architecture.

This patch resolves the issue:
#130

Reviewed-By/Tested-By: Naresh Bhat <naresh.bhat at linaro.org>
Signed-off-by: Megha Dey <megha.dey@linux.intel.com>
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

No branches or pull requests

2 participants