From afee1a851c5e96c54bfb933bd82fe369df932a88 Mon Sep 17 00:00:00 2001
From: Shi Su <67605788+shi-su@users.noreply.github.com>
Date: Tue, 19 Jan 2021 00:23:36 -0800
Subject: [PATCH] [bgpd]: Check zebra is ready to connect when starting bgpd
 (#6478)

Fix #5026

There is a race condition between zebra server accepts connections and bgpd tries to connect. Bgpd has a chance to try to connect before zebra is ready. In this scenario, bgpd will try again after 10 seconds and operate as normal within these 10 seconds. As a consequence, whatever bgpd tries to sent to zebra will be missing in the 10 seconds. To avoid such a scenario, bgpd should start after zebra is ready to accept connections.
---
 dockers/docker-fpm-frr/Dockerfile.j2          |  1 +
 dockers/docker-fpm-frr/bgpd.sh                | 33 +++++++++++++++++++
 .../frr/supervisord/supervisord.conf.j2       |  2 +-
 3 files changed, 35 insertions(+), 1 deletion(-)
 create mode 100755 dockers/docker-fpm-frr/bgpd.sh

diff --git a/dockers/docker-fpm-frr/Dockerfile.j2 b/dockers/docker-fpm-frr/Dockerfile.j2
index ab8b585899..69e3655d0c 100644
--- a/dockers/docker-fpm-frr/Dockerfile.j2
+++ b/dockers/docker-fpm-frr/Dockerfile.j2
@@ -55,6 +55,7 @@ COPY ["TSB", "/usr/bin/TSB"]
 COPY ["TSC", "/usr/bin/TSC"]
 COPY ["files/supervisor-proc-exit-listener", "/usr/bin"]
 COPY ["critical_processes", "/etc/supervisor"]
+COPY ["bgpd.sh", "/usr/bin/"]
 RUN chmod a+x /usr/bin/TSA && \
     chmod a+x /usr/bin/TSB && \
     chmod a+x /usr/bin/TSC
diff --git a/dockers/docker-fpm-frr/bgpd.sh b/dockers/docker-fpm-frr/bgpd.sh
new file mode 100755
index 0000000000..c61a594a1e
--- /dev/null
+++ b/dockers/docker-fpm-frr/bgpd.sh
@@ -0,0 +1,33 @@
+#!/usr/bin/env bash
+
+addr="127.0.0.1"
+port=2601
+
+function help()
+{
+    echo "This script aims to ensure zebra is ready to accept connections before starting bgpd"
+    echo "Usage: $0 [options] [bgpd options]"
+    echo "Options:"
+    echo " -a   Zebra address"
+    echo " -o   Zebra port"
+    exit 1
+}
+
+while getopts ":a:o:h" opt; do
+    case "${opt}" in
+        h) help
+            ;;
+        a) addr=${OPTARG}
+            ;;
+        o) port=${OPTARG}
+            ;;
+    esac
+done
+shift $((OPTIND-1))
+
+timeout 5s bash -c -- "until </dev/tcp/${addr}/${port}; do sleep 0.1;done"
+if [ "$?" != "0" ]; then
+    logger -p error "Error: zebra is not ready to accept connections"
+fi
+
+exec /usr/lib/frr/bgpd "$@"
diff --git a/dockers/docker-fpm-frr/frr/supervisord/supervisord.conf.j2 b/dockers/docker-fpm-frr/frr/supervisord/supervisord.conf.j2
index 205b57e5f6..e71d6fa713 100644
--- a/dockers/docker-fpm-frr/frr/supervisord/supervisord.conf.j2
+++ b/dockers/docker-fpm-frr/frr/supervisord/supervisord.conf.j2
@@ -51,7 +51,7 @@ dependent_startup=true
 dependent_startup_wait_for=zebra:running
 
 [program:bgpd]
-command=/usr/lib/frr/bgpd -A 127.0.0.1 -M snmp
+command=/usr/bin/bgpd.sh -A 127.0.0.1 -M snmp
 priority=5
 stopsignal=KILL
 autostart=false