<div dir="ltr"><div><div><div><div>First off, forget about heartbeat. It is the name of an old package which was replaced with 3 packages; pacemaker, corosync, and "cluster-glue".<br><br></div>Corosync is a communication service that allows multiple servers to communicate with each other. <br>
<br></div>Pacemaker is a system that uses corosync and cluster-glue to coordinate interdependent activities on multiple servers, for example making sure a service is running somewhere in a cluster of servers but only one place.<br>
</div><br>DRBD is summarized as "RAID-1 over ethernet". It keeps "block" devices synchronized across multiple servers.<br><br><br></div><div>The reason you see all of these together is because the file-systems that use block devices such as those presented by DRBD usually assume they are the only code making changes to that block device. If you mount the replicated DRBD block device on two servers at once it can become seriously corrupted because the filesystem drivers on each of the separate servers are not coordinating their activities. Imagine two authors writing and revising the same book who are not communicating and are not even aware of the others existence, then you have a good sense for the problem. <br>
<br>The reason why you see pacemaker and drbd together a lot is that pacemaker can be used (is is actually built to) ensure a replicated DRBD block device is only mounted on one server at a time. Also, since DRBD is used when availability/up-time of the data is important it is also important to not only make sure the disk is mounted somehwere but you also need to make sure the service that uses that disk (database server, webserver, nfs-server, etc) is started on that node and ensure is not running anywhere else. It can do this all automaticly meaning if you server fails while you are away you will usually find out about a server failure via an automatic email mention services are back up rather than a phone call from marketing asking why the website is down. :)<br>
<br></div></div><div class="gmail_extra"><br clear="all"><div>--Tim</div>
<br><br><div class="gmail_quote">On Thu, May 15, 2014 at 1:42 PM, Jonathan Poole <span dir="ltr"><<a href="mailto:jpoole@digitaljedi.ca" target="_blank">jpoole@digitaljedi.ca</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word">Corosync is a cluster engine…DRBD is distributed replicated block device system..<div><br></div><div>Most setups i’ve used use a combination of Pacemaker, corosync, and DRBD for distributed volumes.</div>
<div><br></div><div><br><div><br></div><div>So, basically DRBD is used for replicating volumes.<br><div><br></div><div><div class="">On May 15, 2014, at 12:11 PM, Joe Wennechuk <<a href="mailto:youcanreachmehere@hotmail.com" target="_blank">youcanreachmehere@hotmail.com</a>> wrote:</div>
<div><br><blockquote type="cite"><div style="font-size:12pt;font-family:Calibri;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px">
<div class=""><div dir="ltr">Any Have any experience with DRBD. It seems to use heartbeat, and not corosync?<span> </span><br></div></div>_______________________________________________<br>kwlug-disc mailing list<br><a href="mailto:kwlug-disc@kwlug.org" target="_blank">kwlug-disc@kwlug.org</a><br>
<a href="http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org" target="_blank">http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org</a></div></blockquote></div><br></div></div></div></div><br>_______________________________________________<br>
kwlug-disc mailing list<br>
<a href="mailto:kwlug-disc@kwlug.org">kwlug-disc@kwlug.org</a><br>
<a href="http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org" target="_blank">http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org</a><br>
<br></blockquote></div><br></div>