From: Prasad Pillarisetti prasad.pillarisetti@gmail.com
***Do you perform downtimes just for the purpose of rebooting the systems? ***Is there a recommended interval Linux system should be rebooted?
Okay, here's my short answer:
Regardless of OS, you should _always_ reserve dates/times for preventive maintenance on a regular basis.
But I agree with most others, unless it is a kernel upgrade or GLibC change, there's no absolutely need to reboot. Not even during preventive maintanence should you feel compelled.
-- Bryan J. Smith mailto:b.j.smith@ieee.org
On 6/2/05, Bryan J. Smith b.j.smith@ieee.org thebs413@earthlink.net wrote:
From: Prasad Pillarisetti prasad.pillarisetti@gmail.com
***Do you perform downtimes just for the purpose of rebooting the systems? ***Is there a recommended interval Linux system should be rebooted?
Okay, here's my short answer:
Regardless of OS, you should _always_ reserve dates/times for preventive maintenance on a regular basis.
But I agree with most others, unless it is a kernel upgrade or GLibC change, there's no absolutely need to reboot. Not even during preventive maintanence should you feel compelled.
While I agree with most of your points (the short and long version), I think that when you update software on your production servers during a maintenance window, its generally a good idea to do a reboot then to make sure that your not supprised years later. It has nothing to do with fixing some problem or Linux really needing the reboot, but more of a test to make sure that if/when you loose power or for some other reason you have to reboot, that you some level of confidence that the box will reboot clean and all production services are available.
Just by 2 cents...james
-- Bryan J. Smith mailto:b.j.smith@ieee.org
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Thu, 2005-06-02 at 21:34 -0400, James Olin Oden wrote:
While I agree with most of your points (the short and long version), I think that when you update software on your production servers during a maintenance window, its generally a good idea to do a reboot then to make sure that your not supprised years later.
I never disagreed with this. In fact, if you want to test for boot-time issues, the time to do it _is_ when you have upgraded some details _and_ are in that maintenance window.
I typically find that I'm maintaining a configuration management set whereby I have at least 1 system non-production and I can already test for this, and any variation in production. But if you're going to do it -- e.g., if you suspect that power or other interruptions will possibly occur -- then by all means, that _is_ the time to do it.
I was just basically saying don't reboot just because you have a window. I always reboot my NT servers once a week, and I cringe to see new Linux administrators try to apply the same logic. ;->
It has nothing to do with fixing some problem or Linux really needing the reboot, but more of a test to make sure that if/when you loose power or for some other reason you have to reboot, that you some level of confidence that the box will reboot clean and all production services are available.
And you would know your environment better than I. If that is a real possibility, then that's the time to do it -- when you change something during a maintenance window.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Thu, Jun 02, 2005 at 09:50:22PM -0500, Bryan J. Smith wrote:
On Thu, 2005-06-02 at 21:34 -0400, James Olin Oden wrote:
While I agree with most of your points (the short and long version), I think that when you update software on your production servers during a maintenance window, its generally a good idea to do a reboot then to make sure that your not supprised years later.
I never disagreed with this. In fact, if you want to test for boot-time issues, the time to do it _is_ when you have upgraded some details _and_ are in that maintenance window.
Ehehehe that sounds like me breaking the build tree for Conectiva during development cycles. :)
I typically find that I'm maintaining a configuration management set whereby I have at least 1 system non-production and I can already test for this, and any variation in production. But if you're going to do it -- e.g., if you suspect that power or other interruptions will possibly occur -- then by all means, that _is_ the time to do it.
I second that recomendation.
I was just basically saying don't reboot just because you have a window. I always reboot my NT servers once a week, and I cringe to see new Linux administrators try to apply the same logic. ;->
You have NT servers ? I'm really sorry :P
[]s
- -- Rodrigo Barbosa rodrigob@suespammers.org "Quid quid Latine dictum sit, altum viditur" "Be excellent to each other ..." - Bill & Ted (Wyld Stallyns)
On Fri, 2005-06-03 at 00:30 -0300, Rodrigo Barbosa wrote:
You have NT servers ? I'm really sorry :P
I am a multi-platform architect. Always have been. Always will be.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Fri, Jun 03, 2005 at 12:25:31AM -0500, Bryan J. Smith wrote:
On Fri, 2005-06-03 at 00:30 -0300, Rodrigo Barbosa wrote:
You have NT servers ? I'm really sorry :P
I am a multi-platform architect. Always have been. Always will be.
Me too. I even have Win2K servers here. But NT ? Ick.
[]s
- -- Rodrigo Barbosa rodrigob@suespammers.org "Quid quid Latine dictum sit, altum viditur" "Be excellent to each other ..." - Bill & Ted (Wyld Stallyns)
On Fri, 2005-06-03 at 03:52 -0300, Rodrigo Barbosa wrote:
Me too. I even have Win2K servers here. But NT ? Ick.
NT 5.0 _is_ Windows 2000.
BTW, is this going to stop anytime soon? Or is it your goal to just post to cross me at every opportunity?
I'm _so_sorry_ I crossed you on the fact that run-levels differ between UNIX/Linux systems, insinuated there was a lack of any standard, etc...
Okay?
-
Bryan J. Smith -
Bryan J. Smith <b.j.smith＠ieee.org> -
James Olin Oden -
Rodrigo Barbosa