[plug] Systemd, good or bad

[Hani Jabr]

Fragile systems?  Heck yes. Not the base OS generally, but apps, hardware (how many lots of firmware patches for hardware (inevitably Intel) this year so far?), disk and cluster migrations, app upgrades, OS upgrades. 

Do you not run production systems?  When you make a change, is it really acceptable for you to take time tracking down an issue caused by a typo or a misconfiguration 6 months or more ago?  Do you not test your changes to make sure the box will come up if it crashes or if you reboot it for some other reason?  Do you not have change windows or someone wondering why stuff was down for longer than you said it would be?

> On 22 Sep 2014, at 16:46, Brad Campbell <brad at fnarfbargle.com> wrote:
> 
>> On 22/09/14 16:28, Hani Jabr wrote:
>> Slow starting apps?   Why wait for a DB to initialise before starting sshd, or vice versa?  I have one box that starts multiple instances of DB2 that could just as easily be started in parallel (the bottleneck is the LUNs, not the HBA).
> 
> I can see that. My boxes bring up ssh and other essentials before any application daemons.
>> When you reboot a production machine, do you really never go back and check if it is working?
>> 
> 
> _Really_. I have far better things to do with my time and I have a system that will let me know if something fails to come up. No need to waste time drumming my fingers, when I could be doing other stuff.
> 
> Let me reverse the question, do you really have systems that fragile you need to check they are working after a reboot?
> 
> In the rare event something fails to come up (and I mean maybe once a year in a bad year) it gets diagnosed and properly fixed. I don't have time to paper over problems with reboots and babysitting.
> 
> I was just reading part of a blog post talking about systemd's "advantage" of being able to automatically restart crashed daemons.
> 
> I'm chasing a memory leak in a CCTV system at the moment, and the software has the interesting feature that if the service is restarted more than 3 times in a 24 hour period it forces a server reboot. I asked the developers what that horrible kludge was doing in their otherwise well engineered application and they replied "We only run on windows, what do you expect?".
> 
> If something crashes it's broken and needs to be fixed, not restarted.
> 
> Brad
> -- 
> Dolphins are so intelligent that within a few weeks they can
> train Americans to stand at the edge of the pool and throw them
> fish.