Code optimizations to reduce disk writes on SONiC switches (#20640)

Why I did it
This PR contains various optimizations to reduce disk writes on SONiC switches as mentioned in the High-level Design.

This contributes to the overall reduction in data written to the disk over time, which can help extend the life of the disk and reduce the risk of disk failure.

Work item tracking
Microsoft ADO (number only): 28300356

How I did it
Moved supervisord logs within swss to tmpfs
Optimized away repeated disk writes by vtysh in the bgp container
Moved monit and logrotate writers to tmpfs

How to verify it
Flash image with these optimizations onto a sonic switch and check the following:

Within swss container, supervisord logs are written to /dev/shm/supervisor/
Within bgp container, verify that ~/.history_frr file is not written into repeatedly with show bgp summary json
Verify that monit state file is in /dev/shm/monit
Verify that logrotate status file is in /dev/shm/logrotate

Author: assrinivasan

dockers/docker-orchagent/docker-init.j2

@@ -3,7 +3,7 @@
 mkdir -p /etc/swss/config.d/
 mkdir -p /etc/supervisor/
 mkdir -p /etc/supervisor/conf.d/
-
+mkdir -p /dev/shm/supervisor/
 
 CFGGEN_PARAMS=" \
     -d \

files/build_templates/docker_image_ctl.j2

@@ -633,6 +633,7 @@ start() {
 {%- endif %}
         -e RUNTIME_OWNER=local \
         --uts=host \{# W/A: this should be set per-docker, for those dockers which really need host's UTS namespace #}
+	--tmpfs /var/log/supervisor:rw \
 {%- if install_debug_image == "y" %}
         -v /src:/src:ro -v /debug:/debug:rw \
 {%- endif %}

files/build_templates/sonic_debian_extension.j2

@@ -104,6 +104,11 @@ sudo dpkg --root=$FILESYSTEM_ROOT -i $debs_path/sonic-nettools_*.deb || \
     sudo LANG=C DEBIAN_FRONTEND=noninteractive chroot $FILESYSTEM_ROOT apt-get -y install -f
 sudo setcap 'cap_net_raw=+ep' $FILESYSTEM_ROOT/usr/bin/wol
 
+# This is needed for moving monit logs, state and logrotate status to tmpfs
+sudo bash -c "echo \"d	/dev/shm/monit/ 0755 root root\" > $FILESYSTEM_ROOT/etc/tmpfiles.d/tmpfs-monit.conf"
+sudo bash -c "echo \"d	/dev/shm/logrotate/ 0755 root root\" > $FILESYSTEM_ROOT/etc/tmpfiles.d/tmpfs-logrotate.conf"
+
+
 # Install a patched version of ifupdown2  (and its dependencies via 'apt-get -y install -f')
 sudo dpkg --root=$FILESYSTEM_ROOT -i $debs_path/ifupdown2_*.deb || \
     sudo LANG=C DEBIAN_FRONTEND=noninteractive chroot $FILESYSTEM_ROOT apt-get -y install -f

files/image_config/logrotate/logrotateOverride.conf

@@ -1,2 +1,6 @@
 [Unit]
-Requires=logrotate-config.service
+Requires=logrotate-config.service
+
+[Service]
+ExecStart=
+ExecStart=/usr/sbin/logrotate --state /dev/shm/logrotate/status /etc/logrotate.conf

files/image_config/monit/monitrc

@@ -38,15 +38,15 @@
 ## default the file is placed in $HOME/.monit.id.
 #
 # set idfile /var/.monit.id
-  set idfile /var/lib/monit/id
+  set idfile /dev/shm/monit/id
 #
 ## Set the location of the Monit state file which saves monitoring states
 ## on each cycle. By default the file is placed in $HOME/.monit.state. If
 ## the state file is stored on a persistent filesystem, Monit will recover
 ## the monitoring state across reboots. If it is on temporary filesystem, the
 ## state will be lost on reboot which may be convenient in some situations.
 #
-  set statefile /var/lib/monit/state
+  set statefile /dev/shm/monit/state
 #
 #
 
@@ -91,7 +91,7 @@
 ## available in the back end filesystem).
 #
   set eventqueue
-      basedir /var/lib/monit/events # set the base directory where events will be stored
+      basedir /dev/shm/monit/events # set the base directory where events will be stored
       slots 100                     # optionally limit the queue size
 #
 #

files/initramfs-tools/union-mount.j2

@@ -212,6 +212,9 @@ mkdir -p ${rootmnt}/boot
 mkdir -p ${rootmnt}/host/$image_dir/boot
 mount --bind ${rootmnt}/host/$image_dir/boot ${rootmnt}/boot
 
+## Mount the /tmp directory as tmpfs
+mount -t tmpfs -o rw,nosuid,nodev,size=25% tmpfs ${rootmnt}/tmp
+
 ## Mount loop device or tmpfs for /var/log
 if $logs_inram; then
     # NOTE: some platforms, when reaching initramfs stage, have a small

src/sonic-bgpcfgd/bgpcfgd/managers_bgp.py

@@ -387,7 +387,7 @@ def load_peers():
         Load peers from FRR.
         :return: set of peers, which are already installed in FRR
         """
-        command = ["vtysh", "-c", "show bgp vrfs json"]
+        command = ["vtysh", "-H", "/dev/null", "-c", "show bgp vrfs json"]
         ret_code, out, err = run_command(command)
         if ret_code == 0:
             js_vrf = json.loads(out)

src/sonic-bgpcfgd/bgpmon/bgpmon.py

@@ -77,7 +77,7 @@ def update_new_peer_states(self, peer_dict):
 
     # Get a new snapshot of BGP neighbors and store them in the "new" location
     def get_all_neigh_states(self):
-        cmd = ["vtysh", "-c", 'show bgp summary json']
+        cmd = ["vtysh", "-H", "/dev/null",  "-c", 'show bgp summary json']
         retry_attempt = 0
 
         while retry_attempt < self.MAX_RETRY_ATTEMPTS:

src/sonic-bgpcfgd/tests/test_bgp.py

@@ -32,7 +32,7 @@ def constructor(constants_path, bgp_router_id="", peer_type="general", with_lo0_
     }
 
     return_value_map = {
-        "['vtysh', '-c', 'show bgp vrfs json']": (0, "{\"vrfs\": {\"default\": {}}}", ""),
+        "['vtysh', '-H', '/dev/null', '-c', 'show bgp vrfs json']": (0, "{\"vrfs\": {\"default\": {}}}", ""),
         "['vtysh', '-c', 'show bgp vrf default neighbors json']": (0, "{\"10.10.10.1\": {}, \"20.20.20.1\": {}, \"fc00:10::1\": {}}", "")
     }