vtysh failure result in all bgp related CLI command stuck

[keboliu]

Description

with the latest master image, after switch boot up can observe that all bgp related CLI command stuck like "show bgp summary". After investigated little bit further, found that it because "vtysh" can not be finished and always stuck there. At the same time, the bgp neighbors seem are all established if check from the peer.

root@mtbc-sonic-03-2700:/home/admin# show bgp summary

^C
Aborted! 

root@mtbc-sonic-03-2700:/home/admin# docker exec -it bgp bash
root@mtbc-sonic-03-2700:/# vtysh

Hello, this is FRRouting (version 7.0.1-sonic).
Copyright 1996-2005 Kunihiro Ishiguro, et al.

^C
^C

all tasks inside bgp are running:

root@mtbc-sonic-03-2700:/home/admin# docker exec -it bgp bash
root@mtbc-sonic-03-2700:/# ps -a
  PID TTY          TIME CMD
   59 pts/0    00:00:00 rsyslogd
   69 pts/0    00:00:00 python
   72 pts/0    00:00:28 zebra
   75 pts/0    00:00:00 staticd
   78 pts/0    00:00:02 bgpd
   84 pts/0    00:00:07 fpmsyncd
   96 pts/1    00:00:00 ps

Steps to reproduce the issue:
install the latest master image, there could be a chance to see this issue.

Describe the results you received:
expect bgp related CLI commands shall work.

Describe the results you expected:
bgp related CLI commands all stuck

Additional information you deem important (e.g. issue happens only occasionally):

**Output of `show version`:**

```
root@mtbc-sonic-03-2700:/home/admin# show version

SONiC Software Version: SONiC.HEAD.60-99b0b349
Distribution: Debian 9.9
Kernel: 4.9.0-9-2-amd64
Build commit: 99b0b349
Build date: Wed Aug 14 08:29:23 UTC 2019
Built by: johnar@jenkins-worker-3

Platform: x86_64-mlnx_msn2700-r0
HwSKU: ACS-MSN2700
ASIC: mellanox
Serial Number: MT1533X04568
Uptime: 08:24:00 up 35 min,  2 users,  load average: 3.15, 3.20, 3.29

Docker images:
REPOSITORY                 TAG                 IMAGE ID            SIZE
docker-syncd-mlnx          HEAD.60-99b0b349    b8c3d8732800        369MB
docker-syncd-mlnx          latest              b8c3d8732800        369MB
docker-lldp-sv2            HEAD.60-99b0b349    69f6de70b961        298MB
docker-lldp-sv2            latest              69f6de70b961        298MB
docker-router-advertiser   HEAD.60-99b0b349    27cd8a6f2ea0        281MB
docker-router-advertiser   latest              27cd8a6f2ea0        281MB
docker-dhcp-relay          HEAD.60-99b0b349    96e1f59b6c09        289MB
docker-dhcp-relay          latest              96e1f59b6c09        289MB
docker-database            HEAD.60-99b0b349    5c280cb48068        281MB
docker-database            latest              5c280cb48068        281MB
docker-teamd               HEAD.60-99b0b349    6a2953bb2249        302MB
docker-teamd               latest              6a2953bb2249        302MB
docker-snmp-sv2            HEAD.60-99b0b349    d2a27b71de9e        312MB
docker-snmp-sv2            latest              d2a27b71de9e        312MB
docker-orchagent           HEAD.60-99b0b349    e85db1cc59bf        321MB
docker-orchagent           latest              e85db1cc59bf        321MB
docker-fpm-frr             HEAD.60-99b0b349    30e04664a793        319MB
docker-fpm-frr             latest              30e04664a793        319MB
docker-sonic-telemetry     HEAD.60-99b0b349    997f96a67066        304MB
docker-sonic-telemetry     latest              997f96a67066        304MB
docker-platform-monitor    HEAD.60-99b0b349    ad00733567fa        560MB
docker-platform-monitor    latest              ad00733567fa        560MB

syslog.zip

    **Attach debug file `sudo generate_dump`:**

    ```
    (paste your output here)
    ```

[pavel-shirshov]

Probably fix is here:
#3604

[keboliu]

Probably fix is here:
#3604

Thank you for the notification, will test after PR merged.

[pavel-shirshov]

@keboliu I've merged the PR. Please test it when you have time. Thanks