vr: memory and swap optimizations (#2892)

This tries to provide a threshold based fix for #2873 where swappinness of VR is not used until last resort. By limiting swappiness unless actually needed, the VR system degradation can be avoided for most cases. The other change is around not starting baremetal-vr by default on all VRs, according to the spec https://cwiki.apache.org/confluence/display/CLOUDSTACK/Baremetal+Advanced+Networking+Support only vmware VRs need to run it and that too only as the last step of the setup/completion, so we don't need to run it all the time.

Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>

Author: yadvr

plugins/hypervisors/baremetal/src/com/cloud/baremetal/networkservice/BaremetalKickStartServiceImpl.java

@@ -226,11 +226,16 @@ private boolean preparePxeInAdvancedZone(VirtualMachineProfile profile, NicProfi
             throw new CloudRuntimeException(String.format("please specify 'baremetal.internal.storage.server.ip', which is the http server/nfs server storing kickstart files and ISO files, in global setting"));
         }
 
+        Pair<Boolean, String> ret = SshHelper.sshExecute(mgmtNic.getIPv4Address(), 3922, "root", getSystemVMKeyFile(), null, "systemctl start baremetal-vr");
+        if (!ret.first()) {
+            throw new CloudRuntimeException(String.format("failed to start baremetal agent in virtual router[id:%s]", vr.getId()));
+        }
+
         List<String> tuple =  parseKickstartUrl(profile);
         String cmd =  String.format("/opt/cloud/bin/prepare_pxe.sh %s %s %s %s %s %s", tuple.get(1), tuple.get(2), profile.getTemplate().getUuid(),
                 String.format("01-%s", nic.getMacAddress().replaceAll(":", "-")).toLowerCase(), tuple.get(0), nic.getMacAddress().toLowerCase());
         s_logger.debug(String.format("prepare pxe on virtual router[ip:%s], cmd: %s", mgmtNic.getIPv4Address(), cmd));
-        Pair<Boolean, String> ret = SshHelper.sshExecute(mgmtNic.getIPv4Address(), 3922, "root", getSystemVMKeyFile(), null, cmd);
+        ret = SshHelper.sshExecute(mgmtNic.getIPv4Address(), 3922, "root", getSystemVMKeyFile(), null, cmd);
         if (!ret.first()) {
             throw new CloudRuntimeException(String.format("failed preparing PXE in virtual router[id:%s], because %s", vr.getId(), ret.second()));
         }

systemvm/debian/etc/sysctl.conf

@@ -60,3 +60,6 @@ net.ipv6.conf.all.forwarding = 0
 net.ipv6.conf.all.accept_ra = 0
 net.ipv6.conf.all.accept_redirects = 0
 net.ipv6.conf.all.autoconf = 0
+
+# Minimum swappiness without disabling it
+vm.swappiness=1

systemvm/debian/etc/systemd/journald.conf

@@ -0,0 +1,5 @@
+[Journal]
+Compress=yes
+SystemMaxUse=40M
+SystemMaxFileSize=10M
+RuntimeMaxUse=20M

systemvm/debian/etc/systemd/system/baremetal-vr.service

@@ -0,0 +1,12 @@
+[Unit]
+Description=CloudStack Baremetal VR service
+After=network.target local-fs.target
+
+[Install]
+WantedBy=multi-user.target
+
+[Service]
+Type=simple
+WorkingDirectory=/opt/cloud/bin
+ExecStart=/usr/bin/python /opt/cloud/bin/baremetal-vr.py
+Restart=on-failure

systemvm/debian/etc/systemd/system/cloud-postinit.service

@@ -2,7 +2,7 @@
 Description=CloudStack post-patching init script
 After=cloud-early-config.service network.target local-fs.target
 Before=ssh.service
-Requires=network.service
+Requires=networking.service
 
 [Install]
 WantedBy=multi-user.target

systemvm/debian/opt/cloud/bin/setup/postinit.sh

@@ -21,6 +21,9 @@
 # Eject cdrom if any
 eject || true
 
+# Restart journald for setting changes to apply
+systemctl restart systemd-journald
+
 TYPE=$(grep -Po 'type=\K[a-zA-Z]*' /var/cache/cloud/cmdline)
 if [ "$TYPE" == "router" ] || [ "$TYPE" == "vpcrouter" ] || [ "$TYPE" == "dhcpsrvr" ]
 then
@@ -59,8 +62,3 @@ fi
 systemctl enable --now --no-block ssh
 
 date > /var/cache/cloud/boot_up_done
-
-if [ "$TYPE" == "router" ]
-then
-    python /opt/cloud/bin/baremetal-vr.py &
-fi