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.
**- Why I did it**
We were building a custom version of Supervisor because I had added patches to prevent hangs and crashes if the system clock ever rolled backward. Those changes were merged into the upstream Supervisor repo as of version 3.4.0 (http://supervisord.org/changes.html#id9), therefore, we should be able to simply install the vanilla package via pip. This will also allow us to easily move to Python 3, as Python 3 support was added in version 4.0.0.
**- How I did it**
- Remove Makefiles and patches for building supervisor package from source
- Install Python 3 supervisor package version 4.2.1 in Buster base container
- Also install Python 3 version of supervisord-dependent-startup in Buster base container
- Debian package installed binary in `/usr/bin/`, but pip package installs in `/usr/local/bin/`, so rather than update all absolute paths, I changed all references to simply call `supervisord` and let the system PATH find the executable to prevent future need for changes just in case we ever need to switch back to build a Debian package, then we won't need to modify these again.
- Install Python 2 supervisor package >= 3.4.0 in Stretch and Jessie base containers
* Add bgpmon under sonic-bgpcfgd to be started as a new daemon under BGP docker
* Added bgpmon to be monitored by Monit so that if it crashed, it gets alerted
* use console_scripts entry point to package bgpmon
When stopping the swss, pmon or bgp containers, log messages like the following can be seen:
```
Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,061 ERRO pool dependent-startup event buffer overflowed, discarding event 34
Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,063 ERRO pool dependent-startup event buffer overflowed, discarding event 35
Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,064 ERRO pool dependent-startup event buffer overflowed, discarding event 36
Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,066 ERRO pool dependent-startup event buffer overflowed, discarding event 37
```
This is due to the number of programs in the container managed by supervisor, all generating events at the same time. The default event queue buffer size in supervisor is 10. This patch increases that value in all containers in order to eliminate these errors. As more programs are added to the containers, we may need to further adjust these values. I increased all buffer sizes to 25 except for containers with more programs or templated supervisor.conf files which allow for a variable number of programs. In these cases I increased the buffer size to 50. One final exception is the swss container, where the buffer fills up to ~50, so I increased this buffer to 100.
Resolves https://github.com/Azure/sonic-buildimage/issues/5241
fixes https://github.com/Azure/sonic-buildimage/issues/5026
Explanation:
In the log from the issue I found:
```
I see following in the log
Jul 22 21:13:06.574831 vlab-01 WARNING bgp#bgpd[49]: [EC 33554499] sendmsg_nexthop: zclient_send_message() failed
```
Analyzing source code I found that the error message could be issues only when `zclient_send_rnh()` return less than 0.
```
ret = zclient_send_rnh(zclient, command, p, exact_match,
bnc->bgp->vrf_id);
/* TBD: handle the failure */
if (ret < 0)
flog_warn(EC_BGP_ZEBRA_SEND,
"sendmsg_nexthop: zclient_send_message() failed");
```
I checked [zclient_send_rnh()](88351c8f6d/lib/zclient.c (L654)) and found that this function will return the exit code which the function gets from [zclient_send_message()](88351c8f6d/lib/zclient.c (L266)) But the latter function could return not 0 in two cases:
1. bgpd didn’t connect to the zclient socket yet [code](88351c8f6d/lib/zclient.c (L269))
2. The socket was closed. But in this case we would receive the error message in the log. (And I can find the message in the log when we reboot sonic) [code](88351c8f6d/lib/zclient.c (L277))
Also I see from the logs that client connection was set later we had the issue in bgpd.
Bgpd.log
```
Jul 22 21:13:06.574831 vlab-01 WARNING bgp#bgpd[49]: [EC 33554499] sendmsg_nexthop: zclient_send_message() failed
```
Vs
Zebra.log
```
Jul 22 21:13:12.713249 vlab-01 NOTICE bgp#zebra[48]: client 25 says hello and bids fair to announce only static routes vrf=0
Jul 22 21:13:12.820352 vlab-01 NOTICE bgp#zebra[48]: client 30 says hello and bids fair to announce only bgp routes vrf=0
Jul 22 21:13:12.820352 vlab-01 NOTICE bgp#zebra[48]: client 33 says hello and bids fair to announce only vnc routes vrf=0
```
So in our case we should start zebra first. Wait until it is started and then start bgpd and other daemons.
**- How I did it**
I changed a graph to start daemons in the following order:
1. First start zebra
2. Then starts staticd and bgpd
3. Then starts vtysh -b and bgpeoi after bgpd is started.
