metal/bsp-files/kickstarts/functions.sh
Ovidiu Poncea 7a0a2dac1a Fix issues with controller node Anaconda hang
On some deployments install fails as we keep one FD open
during install. This leads to hangs when Anaconda
'post' stage returns.

On other deployments install fails as udev sometimes creates
multiple links to the same devices in /dev/disk/by-path.
We iterate through this list and, because they are not unique,
we try to run flocks multiple times for the same device.
Locking a device multiple times doesn't work, the second
flock waits for first lock to release.

This commit:
 o removes 'exec {stdout}>&1' from ks-functions.sh so it no
   longer opens FDs in 'post' stage. For the pre stage we open
   it only when needed;
 o makes sure that list of storage devices is unique;
 o increases timeout of udevadm settle from its default of 180s
   to 300s, the value used throughout Anaconda. This helps
   with slower hardware.

Closes-Bug: 1889427
Change-Id: I348f10d96a78ea2c1c25fe6cf48462b0bc31fb84
Signed-off-by: Ovidiu Poncea <ovidiu.poncea@windriver.com>
2020-07-30 14:20:33 +00:00

156 lines
3.4 KiB
Bash

# This file defines functions that can be used in %pre and %post kickstart sections, by including:
# . /tmp/ks-functions.sh
#
cat <<END_FUNCTIONS >/tmp/ks-functions.sh
#
# Copyright (c) xxxYEARxxx Wind River Systems, Inc.
#
# SPDX-License-Identifier: Apache-2.0
#
function wlog()
{
[ -z "\$stdout" ] && stdout=1
local dt="\$(date "+%Y-%m-%d %H:%M:%S.%3N")"
echo "\$dt - \$1" >&\${stdout}
}
function get_by_path()
{
local disk=\$(cd /dev ; readlink -f \$1)
for p in /dev/disk/by-path/*; do
if [ "\$disk" = "\$(readlink -f \$p)" ]; then
echo \$p
return
fi
done
}
function get_disk()
{
echo \$(cd /dev ; readlink -f \$1)
}
function report_pre_failure_with_msg()
{
local msg=\$1
echo -e '\n\nInstallation failed.\n'
echo "\$msg"
exit 1
}
function report_post_failure_with_msg()
{
local msg=\$1
cat <<EOF >> /etc/motd
Installation failed.
\$msg
EOF
echo "\$msg" >/etc/platform/installation_failed
echo -e '\n\nInstallation failed.\n'
echo "\$msg"
exit 1
}
function report_post_failure_with_logfile()
{
local logfile=\$1
cat <<EOF >> /etc/motd
Installation failed.
Please see \$logfile for details of failure
EOF
echo \$logfile >/etc/platform/installation_failed
echo -e '\n\nInstallation failed.\n'
cat \$logfile
exit 1
}
function get_http_port()
{
echo \$(cat /proc/cmdline |xargs -n1 echo |grep '^inst.repo=' | sed -r 's#^[^/]*://[^/]*:([0-9]*)/.*#\1#')
}
function get_disk_dev()
{
local disk
# Detect HDD
for blk_dev in vda vdb sda sdb dda ddb hda hdb; do
if [ -d /sys/block/\$blk_dev ]; then
disk=\$(ls -l /sys/block/\$blk_dev | grep -v usb | head -n1 | sed 's/^.*\([vsdh]d[a-z]\+\).*$/\1/');
if [ -n \$disk ]; then
echo \$disk
return
fi
fi
done
for blk_dev in nvme0n1 nvme1n1; do
if [ -d /sys/block/\$blk_dev ]; then
disk=\$(ls -l /sys/block/\$blk_dev | grep -v usb | head -n1 | sed 's/^.*\(nvme[01]n1\).*$/\1/');
if [ -n \$disk ]; then
echo \$disk
return
fi
fi
done
}
function exec_no_fds()
{
# Close open FDs when executing commands that complain about leaked FDs.
local fds=\$1
local cmd=\$2
local retries=\$3
local interval=\$4
local ret_code=0
local ret_stdout=""
for fd in \$fds
do
local cmd="\$cmd \$fd>&-"
done
if [ -z "\$retries" ]; then
#wlog "Running command: '\$cmd'."
eval "\$cmd"
else
ret_stdout=\$(exec_retry "\$retries" "\$interval" "\$cmd")
ret_code=\$?
echo "\${ret_stdout}"
return \${ret_code}
fi
}
function exec_retry()
{
local retries=\$1
local interval=\$2
local cmd=\$3
let -i retry_count=1
local ret_code=0
local ret_stdout=""
cmd="\$cmd" # 2>&\$stdout"
while [ \$retry_count -le \$retries ]; do
#wlog "Running command: '\$cmd'."
ret_stdout=\$(eval \$cmd)
ret_code=\$?
[ \$ret_code -eq 0 ] && break
wlog "Error running command '\${cmd}'. Try \${retry_count} of \${retries} at \${interval}s."
wlog "ret_code: \${ret_code}, stdout: '\${ret_stdout}'."
sleep \$interval
let retry_count++
done
echo "\${ret_stdout}"
return \${ret_code}
}
END_FUNCTIONS