Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[config reload]: On dual ToR systems, cache ARP and FDB tables #1465

Merged
merged 1 commit into from
Mar 4, 2021
Merged

[config reload]: On dual ToR systems, cache ARP and FDB tables #1465

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
42 changes: 41 additions & 1 deletion config/main.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -911,6 +911,38 @@ def update_sonic_environment():
display_cmd=True
)

def cache_arp_entries():
success = True
cache_dir = '/host/config-reload'
click.echo('Caching ARP table to {}'.format(cache_dir))

if not os.path.exists(cache_dir):
os.mkdir(cache_dir)

arp_cache_cmd = '/usr/local/bin/fast-reboot-dump.py -t {}'.format(cache_dir)
cache_proc = subprocess.Popen(arp_cache_cmd, shell=True, text=True, stdout=subprocess.PIPE)
_, cache_err = cache_proc.communicate()
if cache_err:
click.echo("Could not cache ARP and FDB info prior to reloading")
success = False

if not cache_err:
fdb_cache_file = os.path.join(cache_dir, 'fdb.json')
arp_cache_file = os.path.join(cache_dir, 'arp.json')
theasianpianist marked this conversation as resolved.
Show resolved Hide resolved
fdb_filter_cmd = '/usr/local/bin/filter_fdb_entries -f {} -a {} -c /etc/sonic/configdb.json'.format(fdb_cache_file, arp_cache_file)
filter_proc = subprocess.Popen(fdb_filter_cmd, shell=True, text=True, stdout=subprocess.PIPE)
_, filter_err = filter_proc.communicate()
if filter_err:
click.echo("Could not filter FDB entries prior to reloading")
success = False

# If we are able to successfully cache ARP table info, signal SWSS to restore from our cache
# by creating /host/config-reload/needs-restore
if success:
Copy link
Contributor

@prsunny prsunny Feb 25, 2021 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is this required? in a similar way, swss can check if the (arp.json and fdb.json) file exists and restore, correct?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

SWSS checks for file existence, but also checks /proc/cmdline to make sure fast-reboot actually occurred. I wanted to have a similar check for config reload for redundancy.

restore_flag_file = os.path.join(cache_dir, 'needs-restore')
open(restore_flag_file, 'w').close()
return success

# This is our main entrypoint - the main 'config' command
@click.group(cls=clicommon.AbbreviationGroup, context_settings=CONTEXT_SETTINGS)
@click.pass_context
Expand Down Expand Up @@ -1071,9 +1103,10 @@ def load(filename, yes):
@click.option('-y', '--yes', is_flag=True)
@click.option('-l', '--load-sysinfo', is_flag=True, help='load system default information (mac, portmap etc) first.')
@click.option('-n', '--no_service_restart', default=False, is_flag=True, help='Do not restart docker services')
@click.option('-d', '--disable_arp_cache', default=False, is_flag=True, help='Do not cache ARP table before reloading (applies to dual ToR systems only)')
@click.argument('filename', required=False)
@clicommon.pass_db
def reload(db, filename, yes, load_sysinfo, no_service_restart):
def reload(db, filename, yes, load_sysinfo, no_service_restart, disable_arp_cache):
"""Clear current configuration and import a previous saved config DB dump file.
<filename> : Names of configuration file(s) to load, separated by comma with no spaces in between
"""
Expand Down Expand Up @@ -1112,6 +1145,13 @@ def reload(db, filename, yes, load_sysinfo, no_service_restart):
else:
cfg_hwsku = cfg_hwsku.strip()

# For dual ToR devices, cache ARP and FDB info
localhost_metadata = db.cfgdb.get_table('DEVICE_METADATA')['localhost']
cache_arp_table = not disable_arp_cache and 'subtype' in localhost_metadata and localhost_metadata['subtype'].lower() == 'dualtor'

if cache_arp_table:
cache_arp_entries()

#Stop services before config push
if not no_service_restart:
log.log_info("'reload' stopping services...")
Expand Down