Hi, A few days ago, I tried puppet for configure management. After watching it for some days, I got a problem. puppetmasterd exited without any log message ( nothing about shutdown/ exit is left in neither /var/log/message nor /var/log/puppet/ puppetmaster.log ). it seems that the daemon was killed forcely, but there is no such a cron job or something like a watcher. my env is CentOS 4.6 ruby/1.8.1 installed by yum puppet and his friends from http://people.redhat.com/dlutter/yum/rhel/4/i386/ puppet clients on other servers run well. is there any advice or suggestion? --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en -~----------~----~----~----~------~----~------~--~---
Do you see any memory related error messages in your logs? Puppet is known to have memory leaks, and it could be that it used all of your memory and eventually the kernel killed it as it had no more memory... cheers, Ohad On Thu, Aug 28, 2008 at 5:33 PM, Hu Hailin <hu.hailin@adways.net> wrote:> > Hi, > > A few days ago, I tried puppet for configure management. After > watching it for some days, I got a problem. > puppetmasterd exited without any log message ( nothing about shutdown/ > exit is left in neither /var/log/message nor /var/log/puppet/ > puppetmaster.log ). it seems that the daemon was killed forcely, but > there is no such a cron job or something like a watcher. > > my env is > > CentOS 4.6 > ruby/1.8.1 installed by yum > puppet and his friends from > http://people.redhat.com/dlutter/yum/rhel/4/i386/ > > puppet clients on other servers run well. > > is there any advice or suggestion? > > > >--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en -~----------~----~----~----~------~----~------~--~---
no, /var/log/message is very clean. and the HotSaNIC graph of memory usage seems not matching the start/ disappear time of puppetmasterd. ( i suppose the time of the last log message in /var/log/puppet/puppetmaster.log is the disappear time ) On Aug 28, 7:04 pm, "Ohad Levy" <ohadl...@gmail.com> wrote:> Do you see any memory related error messages in your logs? > Puppet is known to have memory leaks, and it could be that it used all of > your memory and eventually the kernel killed it as it had no more memory... > > cheers, > Ohad > > On Thu, Aug 28, 2008 at 5:33 PM, Hu Hailin <hu.hai...@adways.net> wrote: > > > Hi, > > > A few days ago, I tried puppet for configure management. After > > watching it for some days, I got a problem. > > puppetmasterd exited without any log message ( nothing about shutdown/ > > exit is left in neither /var/log/message nor /var/log/puppet/ > > puppetmaster.log ). it seems that the daemon was killed forcely, but > > there is no such a cron job or something like a watcher. > > > my env is > > > CentOS 4.6 > > ruby/1.8.1 installed by yum > > puppet and his friends from > >http://people.redhat.com/dlutter/yum/rhel/4/i386/ > > > puppet clients on other servers run well. > > > is there any advice or suggestion?--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en -~----------~----~----~----~------~----~------~--~---
Is your system fairly heavily used? In the past, I had to restart my puppetmasterd every 4 hours or so or it would lock up and then disappear. My puppetmaster runs on a system that also handles lots of concurrent RRD writes (file io) and lots of concurrent snmp polling (load and process count). Since then I''ve moved puppetmaster and the rrd work to a machine with double the ram and double the cores and it''s been solid My experience is the puppetmasterd (same with puppet client) does not perform reliably when under resource constraints. (mem, cpu, i/o, etc.) --Joel On Thu, Aug 28, 2008 at 4:36 AM, Hu Hailin <hu.hailin@adways.net> wrote:> > no, /var/log/message is very clean. > and the HotSaNIC graph of memory usage seems not matching the start/ > disappear time of puppetmasterd. ( i suppose the time of the last log > message in /var/log/puppet/puppetmaster.log is the disappear time ) > > On Aug 28, 7:04 pm, "Ohad Levy" <ohadl...@gmail.com> wrote: >> Do you see any memory related error messages in your logs? >> Puppet is known to have memory leaks, and it could be that it used all of >> your memory and eventually the kernel killed it as it had no more memory... >> >> cheers, >> Ohad >> >> On Thu, Aug 28, 2008 at 5:33 PM, Hu Hailin <hu.hai...@adways.net> wrote: >> >> > Hi, >> >> > A few days ago, I tried puppet for configure management. After >> > watching it for some days, I got a problem. >> > puppetmasterd exited without any log message ( nothing about shutdown/ >> > exit is left in neither /var/log/message nor /var/log/puppet/ >> > puppetmaster.log ). it seems that the daemon was killed forcely, but >> > there is no such a cron job or something like a watcher. >> >> > my env is >> >> > CentOS 4.6 >> > ruby/1.8.1 installed by yum >> > puppet and his friends from >> >http://people.redhat.com/dlutter/yum/rhel/4/i386/ >> >> > puppet clients on other servers run well. >> >> > is there any advice or suggestion? > > > >--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en -~----------~----~----~----~------~----~------~--~---
Joel, thank you for your apply. sometimes memory is heavily used, mainly cached. not sure if it is the cause.i''ll keep watching. On Aug 29, 5:38 am, "Joel Krauska" <jkrau...@gmail.com> wrote:> Is your system fairly heavily used? > > In the past, I had to restart my puppetmasterd every 4 hours or so or > it would lock up and then disappear. > > My puppetmaster runs on a system that also handles lots of concurrent > RRD writes (file io) and lots of concurrent snmp polling (load and > process count). > > Since then I''ve moved puppetmaster and the rrd work to a machine with > double the ram and double the cores and it''s been solid > > My experience is the puppetmasterd (same with puppet client) does not > perform reliably when under resource constraints. > (mem, cpu, i/o, etc.) > > --Joel > > On Thu, Aug 28, 2008 at 4:36 AM, Hu Hailin <hu.hai...@adways.net> wrote: > > > no, /var/log/message is very clean. > > and the HotSaNIC graph of memory usage seems not matching the start/ > > disappear time of puppetmasterd. ( i suppose the time of the last log > > message in /var/log/puppet/puppetmaster.log is the disappear time ) > > > On Aug 28, 7:04 pm, "Ohad Levy" <ohadl...@gmail.com> wrote: > >> Do you see any memory related error messages in your logs? > >> Puppet is known to have memory leaks, and it could be that it used all of > >> your memory and eventually the kernel killed it as it had no more memory... > > >> cheers, > >> Ohad > > >> On Thu, Aug 28, 2008 at 5:33 PM, Hu Hailin <hu.hai...@adways.net> wrote: > > >> > Hi, > > >> > A few days ago, I tried puppet for configure management. After > >> > watching it for some days, I got a problem. > >> > puppetmasterd exited without any log message ( nothing about shutdown/ > >> > exit is left in neither /var/log/message nor /var/log/puppet/ > >> > puppetmaster.log ). it seems that the daemon was killed forcely, but > >> > there is no such a cron job or something like a watcher. > > >> > my env is > > >> > CentOS 4.6 > >> > ruby/1.8.1 installed by yum > >> > puppet and his friends from > >> >http://people.redhat.com/dlutter/yum/rhel/4/i386/ > > >> > puppet clients on other servers run well. > > >> > is there any advice or suggestion?--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en -~----------~----~----~----~------~----~------~--~---