I crash i have follow stack trcae<br><br>#0 0xb77fa424 in __kernel_vsyscall ()<br>(gdb) bt<br>#0 0xb77fa424 in __kernel_vsyscall ()<br>#1 0xb77e98a0 in raise () from /lib/i386-linux-gnu/libpthread.so.0<br>#2 0x08230f8b in ?? ()<br>
#3 <signal handler called><br>#4 0xb77fa424 in __kernel_vsyscall ()<br>#5 0xb70eae71 in raise () from /lib/i386-linux-gnu/libc.so.6<br>#6 0xb70ee34e in abort () from /lib/i386-linux-gnu/libc.so.6<br>#7 0xb73130b5 in __gnu_cxx::__verbose_terminate_handler() () from /usr/lib/i386-linux-gnu/libstdc++.so.6<br>
#8 0xb7310fa5 in ?? () from /usr/lib/i386-linux-gnu/libstdc++.so.6<br>#9 0xb7310fe2 in std::terminate() () from /usr/lib/i386-linux-gnu/libstdc++.so.6<br>#10 0xb731114e in __cxa_throw () from /usr/lib/i386-linux-gnu/libstdc++.so.6<br>
#11 0x0822f8c7 in ceph::__ceph_assert_fail(char const*, char const*, int, char const*) ()<br>#12 0x081cf8a4 in MDSMap::get_health(std::basic_ostream<char, std::char_traits<char> >&) const ()<br>#13 0x0811e8a7 in MDSMonitor::get_health(std::basic_ostream<char, std::char_traits<char> >&) const ()<br>
#14 0x080c4977 in Monitor::handle_command(MMonCommand*) ()<br>#15 0x080cf244 in Monitor::_ms_dispatch(Message*) ()<br>#16 0x080df1a4 in Monitor::ms_dispatch(Message*) ()<br>#17 0x081f706d in SimpleMessenger::dispatch_entry() ()<br>
#18 0x080b27d2 in SimpleMessenger::DispatchThread::entry() ()<br>#19 0x081b5d81 in Thread::_entry_func(void*) ()<br>#20 0xb77e0e99 in start_thread () from /lib/i386-linux-gnu/libpthread.so.0<br>#21 0xb71919ee in clone () from /lib/i386-linux-gnu/libc.so.6<br>
<br><br><br>And when one mone craches all other monitors in cluster will crashes too:-((. So one time in cluster not any alive mons<br><br><br><br><br><div class="gmail_quote">2012/3/16 ruslan usifov <span dir="ltr"><<a href="mailto:ruslan.usifov@gmail.com">ruslan.usifov@gmail.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br><br><div class="gmail_quote"><div class="im">2012/3/16 Florian Haas <span dir="ltr"><<a href="mailto:florian@hastexo.com" target="_blank">florian@hastexo.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div><div>On Fri, Mar 16, 2012 at 10:13 AM, ruslan usifov <<a href="mailto:ruslan.usifov@gmail.com" target="_blank">ruslan.usifov@gmail.com</a>> wrote:<br>
> Hello<br>
><br>
> I search a solution for scalable block device (dist that can extend if we<br>
> add some machines to cluster). Only what i find accepten on my task is ceph<br>
> + RDB, but ceph on my test i very unstable(regulary crash of all it daemons)<br>
> + have poor integration with pacemaker. So does anybody recommend some<br>
> solution???<br>
<br>
</div></div>Which Ceph version are you using? Both the Ceph daemons and RBD are<br>
fully integrated into Pacemaker in upstream git.<br></blockquote></div><div><br>I use ceph from this debian repository <a href="http://ceph.newdream.net/debian/" target="_blank">http://ceph.newdream.net/debian/</a>. So i use version 0.43. The reason why i decide use pacemaker is delays that happens periodically(time to time perhaps when client try to connect to failed mon) if one monitor in cluster will fail (in my case crash). I think that if monitor in fail case will move to health node i remove this delayes. But in may case i doesn't have success:-(((( (i try use haproxy, try to launch one monitor which will migrate to heath node)<br>
</div><div class="im"><blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
<a href="https://github.com/ceph/ceph/tree/master/src/ocf" target="_blank">https://github.com/ceph/ceph/tree/master/src/ocf</a><br>
<br>
You may want to look at <a href="http://www.hastexo.com/category/tags/ceph" target="_blank">http://www.hastexo.com/category/tags/ceph</a> for<br>
upcoming updates on this (RSS feed icon at the bottom).<br>
<br></blockquote></div></div><br>
</blockquote></div><br>