frr service restart fails with zebra error- zclient_send_message: buffer_write failed to zclient error

[arjunramu]

Description

frr.log

Our setup:

Site A: Ubuntu 22.04 Linux VM and FRR for BGP
Site B: Catalyst Router
FRR configured with a very minimalistic config - just to exchange routes with neighbors

bgp session established between A to B.
systemctl restart frr failed after bgp session was established.

Issue: FRR service restart failing with zebra error- zclient_send_message: buffer_write failed to zclient error

root@10-1-1-1:~# systemctl restart frr
root@10-1-1-1:~# systemctl status frr
× frr.service - FRRouting
     Loaded: loaded (/lib/systemd/system/frr.service; enabled; vendor preset: enabled)
     Active: failed (Result: start-limit-hit) since Tue 2024-11-12 07:48:55 UTC; 57min ago
       Docs: https://frrouting.readthedocs.io/en/latest/setup.html
    Process: 1494090 ExecStart=/usr/lib/frr/frrinit.sh start (code=exited, status=0/SUCCESS)
    Process: 1495628 ExecStop=/usr/lib/frr/frrinit.sh stop (code=exited, status=0/SUCCESS)
   Main PID: 1494100 (code=exited, status=0/SUCCESS)
        CPU: 1.890s

bgp configuration -

master-10-1-1-1# sh running-config 
Building configuration...

Current configuration:
!
frr version 8.1
frr defaults traditional
hostname master-10-1-1-1
log file /var/log/frr/bgpd.log
log syslog
no ipv6 forwarding
bgp no-rib
service integrated-vtysh-config
username root nopassword
!
router bgp 64512
 bgp router-id 10.1.1.1
 neighbor 10.1.1.2 remote-as 64512
 neighbor 10.1.1.2 update-source 10.1.1.1
 neighbor 10.1.1.2 timers 60 180
 !
 address-family ipv4 unicast
  neighbor 10.1.1.2 next-hop-self
  neighbor 10.1.1.2 route-map DENYALL in
 exit-address-family
exit
!
access-list all seq 5 permit any
!
ip prefix-list denyall seq 5 deny 0.0.0.0/0 le 32
!
route-map DENYALL permit 10
 match ip address prefix-list denyall
exit
!
end

Version

root@10-1-1-1:~# vtysh

Hello, this is FRRouting (version 8.1).
Copyright 1996-2005 Kunihiro Ishiguro, et al.

10-1-1-1# show version
FRRouting 8.1 (master-10-1-1-1).
Copyright 1996-2005 Kunihiro Ishiguro, et al.
configured with:
    '--build=x86_64-linux-gnu' '--prefix=/usr' '--includedir=${prefix}/include' '--mandir=${prefix}/share/man' '--infodir=${prefix}/share/info' '--sysconfdir=/etc' '--localstatedir=/var' '--disable-option-checking' '--disable-silent-rules' '--libdir=${prefix}/lib/x86_64-linux-gnu' '--libexecdir=${prefix}/lib/x86_64-linux-gnu' '--disable-maintainer-mode' '--localstatedir=/var/run/frr' '--sbindir=/usr/lib/frr' '--sysconfdir=/etc/frr' '--with-vtysh-pager=/usr/bin/pager' '--libdir=/usr/lib/x86_64-linux-gnu/frr' '--with-moduledir=/usr/lib/x86_64-linux-gnu/frr/modules' '--disable-dependency-tracking' '--enable-rpki' '--disable-scripting' '--with-libpam' '--enable-doc' '--enable-doc-html' '--enable-snmp' '--enable-fpm' '--disable-protobuf' '--disable-zeromq' '--enable-ospfapi' '--enable-bgp-vnc' '--enable-multipath=256' '--enable-user=frr' '--enable-group=frr' '--enable-vty-group=frrvty' '--enable-configfile-mask=0640' '--enable-logfile-mask=0640' 'build_alias=x86_64-linux-gnu' 'PYTHON=python3'
10-1-1-1#

How to reproduce

Steps to reproduce -

1. Add the neighbor router configuration and establish a bgp session
2. Ensure the bgp session is established between A and B
3. Restart the frr service
4. frr service restart fails.

root@10-1-1-1:~# cat a.sh
while true ; do sleep 3 ; systemctl restart frr ; systemctl status frr | grep running; if [ $? -eq 1 ]; then     exit 1; fi; done
root@10-1-1-1:~#

root@10-1-1-1:~# cat /tmp/a.log
     Active: active (running) since Thu 2024-11-21 08:54:31 UTC; 5ms ago
     Active: active (running) since Thu 2024-11-21 08:54:39 UTC; 5ms ago
     Active: active (running) since Thu 2024-11-21 08:54:48 UTC; 5ms ago
Job for frr.service failed.

Expected behavior

Steps -

1. Add the neighbor router configuration and establish a bgp session
2. Ensure the bgp session is established between A and B
3. Restart the frr service
4. frr restart should be successful

Actual behavior

Steps -

1. Add the neighbor router configuration and establish a bgp session
2. Ensure the bgp session is established between A and B
3. Restart the frr service
4. frr restart failed

Additional context

Workaround is to stop and start the frr service -

Nov 21 08:44:25 10-1-1-1 bgpd[42189]: [YAF85-253AP][EC 100663299] buffer_write: write error on fd 15: Broken pipe
Nov 21 08:44:25 10-1-1-1 bgpd[42189]: [X6B3Y-6W42R][EC 100663302] zclient_send_message: buffer_write failed to zclient fd 15, closing
Nov 21 08:44:25 10-1-1-1 zebra[42184]: [QS0NJ-H5QKJ] Zebra final shutdown
Nov 21 08:44:25 10-1-1-1 frrinit.sh[42335]:  * Stopped staticd
Nov 21 08:44:25 10-1-1-1 frrinit.sh[42336]:  * Stopped bgpd
Nov 21 08:44:25 10-1-1-1 frrinit.sh[42337]:  * Stopped zebra
Nov 21 08:44:25 10-1-1-1 systemd[1]: frr.service: Deactivated successfully.
Nov 21 08:44:25 10-1-1-1 systemd[1]: Stopped FRRouting.
Nov 21 08:44:25 10-1-1-1 systemd[1]: frr.service: Start request repeated too quickly.
Nov 21 08:44:25 10-1-1-1 systemd[1]: frr.service: Failed with result 'start-limit-hit'.
Nov 21 08:44:25 10-1-1-1 systemd[1]: Failed to start FRRouting.
Nov 21 08:44:25 10-1-1-1 systemd[1]: frr.service: Triggering OnFailure= dependencies.
Nov 21 08:44:25 10-1-1-1 systemd[1]: frr.service: Failed to enqueue OnFailure= job, ignoring: Unit heartbeat-failed@frr.service not f>
Nov 21 08:44:52 10-1-1-1 systemd[1]: frr.service: Start request repeated too quickly.
Nov 21 08:44:52 10-1-1-1 systemd[1]: frr.service: Failed with result 'start-limit-hit'.
Nov 21 08:44:52 10-1-1-1 systemd[1]: Failed to start FRRouting.
root@10-1-1-1 :~#
root@10-1-1-1 :~#
root@10-1-1-1 :~# systemctl stop frr
root@10-1-1-1 :~# systemctl start frr
root@10-1-1-1 :~# systemctl status frr
● frr.service - FRRouting
     Loaded: loaded (/lib/systemd/system/frr.service; enabled; vendor preset: enabled)
     Active: active (running) since Thu 2024-11-21 08:50:16 UTC; 2s ago
       Docs: https://frrouting.readthedocs.io/en/latest/setup.html
    Process: 47341 ExecStart=/usr/lib/frr/frrinit.sh start (code=exited, status=0/SUCCESS)
   Main PID: 47350 (watchfrr)
     Status: "FRR Operational"
      Tasks: 13 (limit: 23695)
     Memory: 17.2M
        CPU: 435ms
     CGroup: /system.slice/frr.service
             ├─47350 /usr/lib/frr/watchfrr -d -F traditional zebra bgpd staticd
             ├─47366 /usr/lib/frr/zebra -d -F traditional -A 127.0.0.1 -s 90000000
             ├─47372 /usr/lib/frr/bgpd -d -F traditional --daemon -A 127.0.0.1 -l 10.1.1.1
             └─47379 /usr/lib/frr/staticd -d -F traditional -A 127.0.0.1

Nov 21 08:50:12 10-1-1-1  zebra[47366]: [VTVCM-Y2NW3] Configuration Read in Took: 00:00:00
Nov 21 08:50:12 10-1-1-1  bgpd[47372]: [VTVCM-Y2NW3] Configuration Read in Took: 00:00:00
Nov 21 08:50:12 10-1-1-1  staticd[47379]: [VTVCM-Y2NW3] Configuration Read in Took: 00:00:00
Nov 21 08:50:12 10-1-1-1  watchfrr[47350]: [ZJW5C-1EHNT] restart all process 47351 exited with non-zero status 13
Nov 21 08:50:16 10-1-1-1  watchfrr[47350]: [QDG3Y-BY5TN] bgpd state -> up : connect succeeded
Nov 21 08:50:16 10-1-1-1  watchfrr[47350]: [QDG3Y-BY5TN] zebra state -> up : connect succeeded
Nov 21 08:50:16 10-1-1-1  watchfrr[47350]: [QDG3Y-BY5TN] staticd state -> up : connect succeeded
Nov 21 08:50:16 10-1-1-1  watchfrr[47350]: [KWE5Q-QNGFC] all daemons up, doing startup-complete notify
Nov 21 08:50:16 10-1-1-1 frrinit.sh[47341]:  * Started watchfrr
Nov 21 08:50:16 10-1-1-1 systemd[1]: Started FRRouting.
root@10-1-1-1:~# 

Checklist

• I have searched the open issues for this bug.
• I have not included sensitive information in this report.

[ton31337]

Could you enable debug logging and show us the logs? debug bgp updates, debug bgp neighbor.