[CRIU] [PATCH] scripts: restart all systemd automount services

[Stanislav Kinsburskiy]

I like the idea in general.
But it's racy, because mount point is moved aside for some time on a 
running container, while service is restarting
and nothing prevents a process try to access this mount which it's not 
available.

08.06.2016 16:49, Andrew Vagin пишет:
> Cc: Stanislav
>
> On Mon, Jun 06, 2016 at 06:42:13PM +0300, Eugene Kolomeetz wrote:
>> No need to hardcode list of autofs services,
>> we can query it dynamically via systemctl.
>> https://jira.sw.ru/browse/PSBM-46672
>>
>> Signed-off-by: Eugene Kolomeetz <klm at virtuozzo.com>
>> ---
>>   scripts/systemd-autofs-restart.sh |   11 +++--------
>>   1 files changed, 3 insertions(+), 8 deletions(-)
>>
>> diff --git a/scripts/systemd-autofs-restart.sh b/scripts/systemd-autofs-restart.sh
>> index ba32f1f..6ef2261 100755
>> --- a/scripts/systemd-autofs-restart.sh
>> +++ b/scripts/systemd-autofs-restart.sh
>> @@ -39,7 +39,8 @@ JOIN_CT="$NS_ENTER -t $CRTOOLS_INIT_PID -m -u -p"
>>   # Skip container, if it's not systemd based
>>   [ "$($JOIN_CT basename -- $($JOIN_CT readlink /proc/1/exe))" == "systemd" ] || exit 0
>>   
>> -AUTOFS_SERVICES="proc-sys-fs-binfmt_misc.automount"
>> +AUTOFS_SERVICES="$($JOIN_CT systemctl --no-legend  -t automount \
>> +	 --state=active list-units | awk '{ print $1 }')"
>>   
>>   bindmount=""
>>   
>> @@ -144,13 +145,7 @@ function restart_service {
>>   }
>>   
>>   for service in $AUTOFS_SERVICES; do
>> -	status=$($JOIN_CT systemctl is-active $service)
>> -
>> -	if [ $status == "active" ]; then
>> -		restart_service $service
>> -	else
>> -		echo "$service skipped ($status)"
>> -	fi
>> +	restart_service $service
>>   done
>>   
>>   exit 0
>> -- 
>> 1.7.6
>>
>> _______________________________________________
>> CRIU mailing list
>> CRIU at openvz.org
>> https://lists.openvz.org/mailman/listinfo/criu