<html dir="ltr">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="text-align:left; direction:ltr;">
<div>In an effort to answer the actual question, in my experience the Gluster internals captures the address the first time you probe another node.</div>
<div>So if you're logged into the first node and probe the second using an IP-address, that is what will "forever" be displayed by gluster status, and if you use a hostname that's what will be shown.</div>
<div>Brick paths are captured when the brick is registered, so using a path with IP will always show the IP as part of the path, and hostname will show that, etc.</div>
<div><br>
</div>
<div>I haven't verified, but the second node I believe will attempt a reverse lookup of the first node (when probing first->second) and record that name (if any) as the "primary" name of the first node.</div>
<div>Also good to know, nodes can have multiple names, the primary name is the one "configured" during setup, and secondary names can be added by probing them afterwards.</div>
<div>All IP/hostname/FQDN parts of the brick-path has to be known to the cluster, by probing that IP/hostname/FQDN.</div>
<div><span><br>
<br>
Best regards<br>
<b>THORGEIR MARTHINUSSEN</b> </span></div>
<div><br>
</div>
<div>-----Original Message-----</div>
<div><b>From</b>: TomK <<a href="mailto:TomK%20%3ctomkcpr@mdevsys.com%3e">tomkcpr@mdevsys.com</a>></div>
<div><b>Reply-To</b>: <a href="mailto:tomkcpr@mdevsys.com">tomkcpr@mdevsys.com</a></div>
<div><b>To</b>: <a href="mailto:gluster-users@gluster.org">gluster-users@gluster.org</a></div>
<div><b>Subject</b>: Re: [Gluster-users] Where does Gluster capture the hostnames from?</div>
<div><b>Date</b>: Mon, 23 Sep 2019 21:31:19 -0400</div>
<div><br>
</div>
<pre>Hey All,</pre>
<pre><br></pre>
<pre>My hosts below:</pre>
<pre><br></pre>
<pre>[root@mdskvm-p01 ~]# cat /etc/hosts</pre>
<pre>127.0.0.1 localhost localhost.localdomain localhost4 </pre>
<pre>localhost4.localdomain4</pre>
<pre>::1 localhost localhost.localdomain localhost6 </pre>
<pre>localhost6.localdomain6</pre>
<pre>[root@mdskvm-p01 ~]# hostname</pre>
<pre>mdskvm-p01.nix.mds.xyz</pre>
<pre>[root@mdskvm-p01 ~]# hostname -f</pre>
<pre>mdskvm-p01.nix.mds.xyz</pre>
<pre>[root@mdskvm-p01 ~]#</pre>
<pre><br></pre>
<pre><br></pre>
<pre>[root@mdskvm-p02 ~]# cat /etc/hosts</pre>
<pre>127.0.0.1 localhost localhost.localdomain localhost4 </pre>
<pre>localhost4.localdomain4</pre>
<pre>::1 localhost localhost.localdomain localhost6 </pre>
<pre>localhost6.localdomain6</pre>
<pre>[root@mdskvm-p02 ~]# hostname</pre>
<pre>mdskvm-p02.nix.mds.xyz</pre>
<pre>[root@mdskvm-p02 ~]# hostname -f</pre>
<pre>mdskvm-p02.nix.mds.xyz</pre>
<pre>[root@mdskvm-p02 ~]#</pre>
<pre><br></pre>
<pre><br></pre>
<pre>My take on the /etc/hosts file discussion:</pre>
<pre><br></pre>
<pre>1) If hostname / hostname -f returns any valid values, the software </pre>
<pre>should capture it.</pre>
<pre><br></pre>
<pre><br></pre>
<pre>2) There is no benefit or need to use /etc/hosts in a small setup. </pre>
<pre>Larger setups resolving hosts against an enterprise DNS behind many </pre>
<pre>switches could be a problem. Managing our /etc/hosts files using </pre>
<pre>Ansible helped to reduce some of these problems esp since lookups are </pre>
<pre>logged against the connection tracking tables, that can get full, </pre>
<pre>network response time could vary etc. ("Semi static" I guess might </pre>
<pre>describe this approach best?) These are populated, if changes are </pre>
<pre>needed, via an initial DNS lookup once a day. Invariably, managing </pre>
<pre>/etc/hosts is time consuming and messy.</pre>
<pre><br></pre>
<pre><br></pre>
<pre>3) Running a good DNS cluster, something like a two node IPA cluster </pre>
<pre>that I run for a small setup, prevents such outages. This particularly </pre>
<pre>when also placing a VIP across the nodes and locating cluster nodes </pre>
<pre>across different hardware and locations.</pre>
<pre><br></pre>
<pre><br></pre>
<pre>4) Point 2) should be no reason why an application cannot obtain or </pre>
<pre>resolve proper DNS entries in 1).</pre>
<pre><br></pre>
<pre><br></pre>
<pre>Having said that, decided to check if there's any benefit to having </pre>
<pre>entries in /etc/hosts:</pre>
<pre><br></pre>
<pre>[root@mdskvm-p01 ~]# time $(dig mdskvm-p01.nix.mds.xyz >/dev/null)</pre>
<pre><br></pre>
<pre>real 0m0.092s</pre>
<pre>user 0m0.087s</pre>
<pre>sys 0m0.005s</pre>
<pre>[root@mdskvm-p01 ~]# time $(dig mdskvm-p02.nix.mds.xyz >/dev/null)</pre>
<pre><br></pre>
<pre>real 0m0.092s</pre>
<pre>user 0m0.084s</pre>
<pre>sys 0m0.008s</pre>
<pre>[root@mdskvm-p01 ~]# cat /etc/hosts</pre>
<pre>127.0.0.1 localhost localhost.localdomain localhost4 </pre>
<pre>localhost4.localdomain4</pre>
<pre>::1 localhost localhost.localdomain localhost6 </pre>
<pre>localhost6.localdomain6</pre>
<pre>192.168.0.60 mdskvm-p01.nix.mds.xyz mdskvm-p01</pre>
<pre>192.168.0.39 mdskvm-p02.nix.mds.xyz mdskvm-p02</pre>
<pre>[root@mdskvm-p01 ~]# vi /etc/hosts</pre>
<pre>[root@mdskvm-p01 ~]# time $(dig mdskvm-p01.nix.mds.xyz >/dev/null)</pre>
<pre><br></pre>
<pre>real 0m0.093s</pre>
<pre>user 0m0.082s</pre>
<pre>sys 0m0.010s</pre>
<pre>[root@mdskvm-p01 ~]# time $(dig mdskvm-p02.nix.mds.xyz >/dev/null)</pre>
<pre><br></pre>
<pre>real 0m0.093s</pre>
<pre>user 0m0.085s</pre>
<pre>sys 0m0.007s</pre>
<pre>[root@mdskvm-p01 ~]# time $(dig mdskvm-p01.nix.mds.xyz >/dev/null)</pre>
<pre><br></pre>
<pre>real 0m0.094s</pre>
<pre>user 0m0.084s</pre>
<pre>sys 0m0.010s</pre>
<pre>[root@mdskvm-p01 ~]# time $(dig mdskvm-p02.nix.mds.xyz >/dev/null)</pre>
<pre><br></pre>
<pre>real 0m0.092s</pre>
<pre>user 0m0.081s</pre>
<pre>sys 0m0.011s</pre>
<pre>[root@mdskvm-p01 ~]# cat /etc/hosts</pre>
<pre>127.0.0.1 localhost localhost.localdomain localhost4 </pre>
<pre>localhost4.localdomain4</pre>
<pre>::1 localhost localhost.localdomain localhost6 </pre>
<pre>localhost6.localdomain6</pre>
<pre>[root@mdskvm-p01 ~]#</pre>
<pre><br></pre>
<pre><br></pre>
<pre>So with /etc/hosts file entries present makes little difference in small </pre>
<pre>setup when governed by /etc/nsswitch.conf .</pre>
<pre><br></pre>
<pre>Having entries in /etc/hosts, doesn't affect how gluster displays the </pre>
<pre>entries when calling gluster volume status .</pre>
<pre><br></pre>
<pre><br></pre>
<pre>Cheers,</pre>
<pre>TK</pre>
<pre><br></pre>
<pre>On 9/23/2019 11:36 AM, Joe Julian wrote:</pre>
<blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex">
<pre>Perhaps I misread the intent, I apologize if I did. I read "static </pre>
<pre>entries" as "ip addresses" which I've seen suggested (from my </pre>
<pre>perspective) far too often. /etc/hosts is a valid solution that can </pre>
<pre>still adapt if the network needs to evolve.</pre>
<pre><br></pre>
<pre>On 9/23/19 8:29 AM, ROUVRAIS Cedric wrote:</pre>
<blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex">
<pre>Hello,</pre>
<pre><br></pre>
<pre>I guess everyone sort of has his perspective on this topic.</pre>
<pre><br></pre>
<pre>I don't want to take this thread on an off-topic conversation </pre>
<pre>(discussing the merits of having a local hosts file) but I do dissent, </pre>
<pre>and therefore had to respond, on the shortcut that using a local </pre>
<pre>etc/host file creates a fixed network configuration that can never </pre>
<pre>adapt as business needs change. I'm running a k8s infrastructure and </pre>
<pre>actually have local conf files, FWIW.</pre>
<pre><br></pre>
<pre>Regards,</pre>
<pre><br></pre>
<pre>Cédric</pre>
<pre><br></pre>
<pre><br></pre>
<pre>-----Original Message-----</pre>
<pre>From: </pre>
<a href="mailto:gluster-users-bounces@gluster.org">
<pre>gluster-users-bounces@gluster.org</pre>
</a>
<pre> </pre>
<pre><</pre>
<a href="mailto:gluster-users-bounces@gluster.org">
<pre>gluster-users-bounces@gluster.org</pre>
</a>
<pre>> On Behalf Of Joe Julian</pre>
<pre>Sent: lundi 23 septembre 2019 17:06</pre>
<pre>To: </pre>
<a href="mailto:gluster-users@gluster.org">
<pre>gluster-users@gluster.org</pre>
</a>
<pre><br></pre>
<pre>Subject: Re: [Gluster-users] Where does Gluster capture the hostnames </pre>
<pre>from?</pre>
<pre><br></pre>
<pre>I disagree about it being "best practice" to lock yourself in to a </pre>
<pre>fixed network configuration that can never adapt as business needs </pre>
<pre>change.</pre>
<pre>There are other resilient ways of ensuring your hostnames resolve </pre>
<pre>consistently (so that your cluster doesn't run loose ;-)).</pre>
<pre><br></pre>
<pre>On 9/23/19 7:38 AM, Strahil wrote:</pre>
<blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex">
<pre>Also,</pre>
<pre><br></pre>
<pre>It's more safe to have static entries for your cluster - after all if </pre>
<pre>DNS fails for some reason - you don't want to loose your cluster.A </pre>
<pre>kind of 'Best Practice'.</pre>
<pre><br></pre>
<pre>Best Regards,</pre>
<pre>Strahil NikolovOn Sep 23, 2019 15:01, TomK <</pre>
<a href="mailto:tomkcpr@mdevsys.com">
<pre>tomkcpr@mdevsys.com</pre>
</a>
<pre>> wrote:</pre>
<blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex">
<pre>Do I *really* need specific /etc/hosts entries when I have IPA?</pre>
<pre><br></pre>
<pre>[root@mdskvm-p01 ~]# cat /etc/hosts</pre>
<pre>127.0.0.1 localhost localhost.localdomain localhost4</pre>
<pre>localhost4.localdomain4</pre>
<pre>::1 localhost localhost.localdomain localhost6</pre>
<pre>localhost6.localdomain6</pre>
<pre>[root@mdskvm-p01 ~]#</pre>
<pre><br></pre>
<pre>I really shouldn't need too. ( Ref below, everything resolves fine.</pre>
<pre>)</pre>
<pre><br></pre>
<pre>Cheers,</pre>
<pre>TK</pre>
<pre><br></pre>
<pre><br></pre>
<pre>On 9/23/2019 1:32 AM, Strahil wrote:</pre>
<blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex">
<pre>Check your /etc/hosts for an entry like:</pre>
<pre>192.168.0.60 mdskvm-p01.nix.mds.xyz mdskvm-p01</pre>
<pre><br></pre>
<pre>Best Regards,</pre>
<pre>Strahil NikolovOn Sep 23, 2019 06:58, TomK <</pre>
<a href="mailto:tomkcpr@mdevsys.com">
<pre>tomkcpr@mdevsys.com</pre>
</a>
<pre>> </pre>
<pre>wrote:</pre>
<blockquote type="cite" style="margin:0 0 0 .8ex; border-left:2px #729fcf solid;padding-left:1ex">
<pre>Hey All,</pre>
<pre><br></pre>
<pre>Take the two hosts below as example. One host shows NFS Server on</pre>
<pre>192.168.0.60 (FQDN is mdskvm-p01.nix.mds.xyz).</pre>
<pre><br></pre>
<pre>The other shows mdskvm-p02 (FQDN is mdskvm-p02.nix.mds.xyz).</pre>
<pre><br></pre>
<pre>Why is there no consistency or correct hostname resolution? Where</pre>
<pre>does gluster get the hostnames from?</pre>
<pre><br></pre>
<pre><br></pre>
<pre>[root@mdskvm-p02 glusterfs]# gluster volume status Status of</pre>
<pre>volume: mdsgv01 Gluster process TCP</pre>
<pre>Port RDMA Port Online Pid</pre>
<pre>-------------------------------------------------------------------</pre>
<pre>----------- Brick mdskvm-p02.nix.mds.xyz:/mnt/p02-d01/g</pre>
<pre>lusterv02 49153 0 Y</pre>
<pre>17503</pre>
<pre>Brick mdskvm-p01.nix.mds.xyz:/mnt/p01-d01/g</pre>
<pre>lusterv01 49153 0 Y</pre>
<pre>15044</pre>
<pre>NFS Server on localhost N/A N/A N</pre>
<pre>N/A Self-heal Daemon on localhost N/A N/A</pre>
<pre>Y</pre>
<pre>17531</pre>
<pre>NFS Server on 192.168.0.60 N/A N/A N</pre>
<pre>N/A Self-heal Daemon on 192.168.0.60 N/A N/A</pre>
<pre>Y</pre>
<pre>15073</pre>
<pre><br></pre>
<pre>Task Status of Volume mdsgv01</pre>
<pre>-------------------------------------------------------------------</pre>
<pre>-----------</pre>
<pre>There are no active volume tasks</pre>
<pre><br></pre>
<pre>[root@mdskvm-p02 glusterfs]#</pre>
<pre><br></pre>
<pre><br></pre>
<pre><br></pre>
<pre><br></pre>
<pre>[root@mdskvm-p01 ~]# gluster volume status Status of volume:</pre>
<pre>mdsgv01 Gluster process TCP Port RDMA</pre>
<pre>Port Online Pid</pre>
<pre>-------------------------------------------------------------------</pre>
<pre>----------- Brick mdskvm-p02.nix.mds.xyz:/mnt/p02-d01/g</pre>
<pre>lusterv02 49153 0 Y</pre>
<pre>17503</pre>
<pre>Brick mdskvm-p01.nix.mds.xyz:/mnt/p01-d01/g</pre>
<pre>lusterv01 49153 0 Y</pre>
<pre>15044</pre>
<pre>NFS Server on localhost N/A N/A N</pre>
<pre>N/A Self-heal Daemon on localhost N/A N/A</pre>
<pre>Y</pre>
<pre>15073</pre>
<pre>NFS Server on mdskvm-p02 N/A N/A N</pre>
<pre>N/A Self-heal Daemon on mdskvm-p02 N/A N/A</pre>
<pre>Y</pre>
<pre>17531</pre>
<pre><br></pre>
<pre>Task Status of Volume mdsgv01</pre>
<pre>-------------------------------------------------------------------</pre>
<pre>-----------</pre>
<pre>There are no active volume tasks</pre>
<pre><br></pre>
<pre>[root@mdskvm-p01 ~]#</pre>
<pre><br></pre>
<pre><br></pre>
<pre><br></pre>
<pre>But when verifying everything all seems fine:</pre>
<pre><br></pre>
<pre><br></pre>
<pre>(1):</pre>
<pre>[root@mdskvm-p01 glusterfs]# dig -x 192.168.0.39 ;; QUESTION</pre>
<pre>SECTION:</pre>
<pre>;39.0.168.192.in-addr.arpa. IN PTR</pre>
<pre><br></pre>
<pre>;; ANSWER SECTION:</pre>
<pre>39.0.168.192.in-addr.arpa. 1200 IN PTR </pre>
<pre>mdskvm-p02.nix.mds.xyz.</pre>
<pre>[root@mdskvm-p01 glusterfs]# hostname -f mdskvm-p01.nix.mds.xyz</pre>
<pre>[root@mdskvm-p01 glusterfs]# hostname -s</pre>
<pre>mdskvm-p01</pre>
<pre>[root@mdskvm-p01 glusterfs]# hostname mdskvm-p01.nix.mds.xyz</pre>
<pre>[root@mdskvm-p01 glusterfs]#</pre>
<pre><br></pre>
<pre><br></pre>
<pre>(2):</pre>
<pre><br></pre>
<pre>[root@mdskvm-p02 glusterfs]# dig -x 192.168.0.60 ;; QUESTION</pre>
<pre>SECTION:</pre>
<pre>;60.0.168.192.in-addr.arpa. IN PTR</pre>
<pre><br></pre>
<pre>;; ANSWER SECTION:</pre>
<pre>60.0.168.192.in-addr.arpa. 1200 IN PTR </pre>
<pre>mdskvm-p01.nix.mds.xyz.</pre>
<pre><br></pre>
<pre>[root@mdskvm-p02 glusterfs]# hostname -s</pre>
<pre>mdskvm-p02</pre>
<pre>[root@mdskvm-p02 glusterfs]# hostname -f mdskvm-p02.nix.mds.xyz</pre>
<pre>[root@mdskvm-p02 glusterfs]# hostname mdskvm-p02.nix.mds.xyz</pre>
<pre>[root@mdskvm-p02 glusterfs]#</pre>
<pre><br></pre>
<pre><br></pre>
<pre>Gluster version used is:</pre>
<pre><br></pre>
<pre>[root@mdskvm-p01 glusterfs]# rpm -aq|grep -Ei gluster</pre>
<pre>glusterfs-server-3.12.15-1.el7.x86_64</pre>
<pre>glusterfs-client-xlators-3.12.15-1.el7.x86_64</pre>
<pre>glusterfs-rdma-3.12.15-1.el7.x86_64</pre>
<pre>glusterfs-3.12.15-1.el7.x86_64</pre>
<pre>glusterfs-events-3.12.15-1.el7.x86_64</pre>
<pre>libvirt-daemon-driver-storage-gluster-4.5.0-10.el7_6.12.x86_64</pre>
<pre>glusterfs-libs-3.12.15-1.el7.x86_64</pre>
<pre>glusterfs-fuse-3.12.15-1.el7.x86_64</pre>
<pre>glusterfs-geo-replication-3.12.15-1.el7.x86_64</pre>
<pre>python2-gluster-3.12.15-1.el7.x86_64</pre>
</blockquote>
</blockquote>
</blockquote>
<pre>________</pre>
<pre><br></pre>
<pre>Community Meeting Calendar:</pre>
<pre><br></pre>
<pre>APAC Schedule -</pre>
<pre>Every 2nd and 4th Tuesday at 11:30 AM IST</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>NA/EMEA Schedule -</pre>
<pre>Every 1st and 3rd Tuesday at 01:00 PM EDT</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>Gluster-users mailing list</pre>
<a href="mailto:Gluster-users@gluster.org">
<pre>Gluster-users@gluster.org</pre>
</a>
<pre><br></pre>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users">
<pre>https://lists.gluster.org/mailman/listinfo/gluster-users</pre>
</a>
<pre><br></pre>
</blockquote>
<pre>________</pre>
<pre><br></pre>
<pre>Community Meeting Calendar:</pre>
<pre><br></pre>
<pre>APAC Schedule -</pre>
<pre>Every 2nd and 4th Tuesday at 11:30 AM IST</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>NA/EMEA Schedule -</pre>
<pre>Every 1st and 3rd Tuesday at 01:00 PM EDT</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>Gluster-users mailing list</pre>
<a href="mailto:Gluster-users@gluster.org">
<pre>Gluster-users@gluster.org</pre>
</a>
<pre><br></pre>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users">
<pre>https://lists.gluster.org/mailman/listinfo/gluster-users</pre>
</a>
<pre><br></pre>
<pre>________</pre>
<pre><br></pre>
<pre>Community Meeting Calendar:</pre>
<pre><br></pre>
<pre>APAC Schedule -</pre>
<pre>Every 2nd and 4th Tuesday at 11:30 AM IST</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>NA/EMEA Schedule -</pre>
<pre>Every 1st and 3rd Tuesday at 01:00 PM EDT</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>Gluster-users mailing list</pre>
<a href="mailto:Gluster-users@gluster.org">
<pre>Gluster-users@gluster.org</pre>
</a>
<pre><br></pre>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users">
<pre>https://lists.gluster.org/mailman/listinfo/gluster-users</pre>
</a>
<pre><br></pre>
</blockquote>
<pre>________</pre>
<pre><br></pre>
<pre>Community Meeting Calendar:</pre>
<pre><br></pre>
<pre>APAC Schedule -</pre>
<pre>Every 2nd and 4th Tuesday at 11:30 AM IST</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>NA/EMEA Schedule -</pre>
<pre>Every 1st and 3rd Tuesday at 01:00 PM EDT</pre>
<pre>Bridge: </pre>
<a href="https://bluejeans.com/118564314">
<pre>https://bluejeans.com/118564314</pre>
</a>
<pre><br></pre>
<pre><br></pre>
<pre>Gluster-users mailing list</pre>
<a href="mailto:Gluster-users@gluster.org">
<pre>Gluster-users@gluster.org</pre>
</a>
<pre><br></pre>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users">
<pre>https://lists.gluster.org/mailman/listinfo/gluster-users</pre>
</a>
<pre><br></pre>
</blockquote>
<pre><br></pre>
<pre><br></pre>
</body>
</html>