runit don't restart service.

runit don't restart service.

[Yoshinori Sato]

Hello.

I'm using runit-2.1.2. I have a question about this.

I checked the operation at the time of service abnormality.

- before problem.
[root@nodeA ec2-user]# ps -ewwwf |grep lkcheck
root     10647 10592  0 7月13 ?  00:00:00 /opt/LifeKeeper/sbin/runsv lkcheck
root     12521 10647  0 7月13 ?  00:00:02 /opt/LifeKeeper/bin/lkcheck
root     12523 12521  0 7月13 ?  00:00:00 /opt/LifeKeeper/bin/lklogmsg
-p lkcheck -l LK_ERROR -s runit -t  -a uncaught_error -i 0 -c
/opt/LifeKeeper/bin/lkcheck
root     25993 23089  0 02:19 pts/0    00:00:00 grep --color=auto lkcheck

lkcheck is runit managed service process.

- terminate service
[root@nodeA ec2-user]# kill 10647 12521

- restart check
[root@nodeA ec2-user]# ps -ewwwf |grep lkcheck
root     26053 10592  0 02:20 ?        00:00:00
/opt/LifeKeeper/sbin/runsv lkcheck
root     26056 23089  0 02:20 pts/0    00:00:00 grep --color=auto lkcheck

I was hoping that lkcheck was restarting.
However, it has not restarted.

I would like to know if this behavior is a specification.

Thanks.

Re: runit don't restart service.

[Jonathan de Boyne Pollard]

Yoshinori Sato:

> root 10647 10592 0 7月13 ? 00:00:00 /opt/LifeKeeper/sbin/runsv lkcheck
>
> terminate service
> [root@nodeA ec2-user]# kill 10647 12521
>
That is not terminating a service.  That is terminating the service 
supervision subsystem.  If you kill the process that actually does the 
restarting, you really should not expect that process to then restart 
anything.  The behaviour of runsv in response to SIGTERM is documented 
in its manual; as are the ways to properly tell it to terminate a 
service, as opposed to what you did here.

Use the runit manual in preference to the LifeKeeper manual here, too.  
In particular learn the use of runit commands like

     # /opt/LifeKeeper/sbin/sv status /opt/LifeKeeper/etc/service/*

rather than the clumsy and error-prone grepping of the process list that 
is given in the LifeKeeper manual.

Re: runit don't restart service.

[Yoshinori Sato]

OK.

I understood it.
I will try it again.

Thanks.

2018-07-25 17:35 GMT+09:00 Jonathan de Boyne Pollard
<J.deBoynePollard-newsgroups@ntlworld.com>:
> Yoshinori Sato:
>
>> root 10647 10592 0 7月13 ? 00:00:00 /opt/LifeKeeper/sbin/runsv lkcheck
>>
>> terminate service
>> [root@nodeA ec2-user]# kill 10647 12521
>>
> That is not terminating a service.  That is terminating the service
> supervision subsystem.  If you kill the process that actually does the
> restarting, you really should not expect that process to then restart
> anything.  The behaviour of runsv in response to SIGTERM is documented in
> its manual; as are the ways to properly tell it to terminate a service, as
> opposed to what you did here.
>
> Use the runit manual in preference to the LifeKeeper manual here, too.  In
> particular learn the use of runit commands like
>
>     # /opt/LifeKeeper/sbin/sv status /opt/LifeKeeper/etc/service/*
>
> rather than the clumsy and error-prone grepping of the process list that is
> given in the LifeKeeper manual.
>